Graphene-Enhanced Methacrylated Alginate Gel Films for Sustainable Dye Removal in Water Purification

Self-standing nanocomposite films were prepared by three-dimensional UV-induced radical copolymerization of methacrylated alginate (MALG) with acrylic acid (AA) and reinforced with graphene oxide (GO) to improve both mechanical strength and dye adsorption capacity in wastewater decontamination operations. Dynamic mechanical–thermal analysis revealed variations in storage modulus: the higher the GO content, the higher the storage modulus (E′) values. Also, the higher the temperature (associated with a lower and lower water content of films), the larger values of E′ for the films of the same composition (E′(25 °C) = 676.6–1538.7 MPa; E′(100 °C) = 886.9–2066.6 MPa), providing insights into the compatibility between GO and the MALG/AA matrix, as well as, assessing the improvement in the nanocomposite’s final mechanical properties. These crosslinked films in a dry state exhibited rapid water uptake and relatively short drying times (ca. 30 min at room temperature for the MALG/AA/GO composites) resulting from the swelling–drying studies and water contact angle measurements. The efficacy of methylene blue removal from water assessed via UV–VIS spectrometry revealed excellent results, expressed as an adsorption yield of 70–80% and 85–98% after 30 h and 258 h, respectively, of immersion time of films into an MB aqueous solution of 12.5 mg/L (as the contaminated water model). The reusability of the same films was evaluated by consecutive extraction processes of MB from the composite membranes when the content of desorbed dye was also spectrophotometrically monitored and conducted in acidic conditions (HCl aqueous solutions of pH 2). Overall, the introduction of GO in the developed self-standing MALG/AA nanocomposite films exhibited enhanced mechanical properties and increased efficiency for dye removal applications. Their great reutilization potential was highlighted by low drying times and a good ability to release the dye initially adsorbed. Thus, the prepared films could be suitable materials for sustainable and effective water treatment technologies.


Introduction
Polysaccharides stand out as biopolymers with considerable potential applications [1][2][3] based on their widespread availability, relatively low costs, and eco-friendliness.In recent years, there has been a growing interest in alginate, a naturally occurring substance abundantly found in the cell walls of numerous algae species, providing remarkable flexibility and toughness, enabling these algae to withstand the relentless forces of continuous waves and tidal activities [4].Structurally, alginate is a linear copolymer of β-d-mannuronic acid (M) and its C-5 epimer α-l-guluronic acid (G) linked by ether 1→4 bonds.Each alginate species can be characterized by its M/G ratio, as it may greatly influence the material's final mechanical characteristics [5,6].The distinctive characteristics of alginate are related to its thus reinforcing the global structure and providing additional adsorption sites in the GO sheets' interlayered space.Based on the relatively high market price of GO, only two GOcontaining formulations have been chosen, developed, and studied, MALG/AA/GO1.25 and MALG/AA/GO5, where 1.25 and 5 denote the percentages of GO with respect to the MALG component in the formulations aforementioned.
The different amounts of graphene oxide (GO) were incorporated into methacrylated alginate/acrylic acid (MALG/AA) films to evaluate their impact on the resulting mechanical properties and MB adsorption efficiency.The overall efficiencies and capacities were quantified through UV-VIS spectrophotometric analysis of aqueous MB solutions in which the fabricated films were immersed.In response to the demand for enhanced adsorption properties of materials concerning water pollutants, a study of the reutilization potential of the fabricated films was carried out involving the characterization of swelling-drying behavior and associated timeframes, as well as the evaluation of the extraction efficiency of adsorbed MB (as pollutant model in wastewater) under acidic conditions.

Synthesis and Characterization of Alginate and Its Methacrylated Derivative
The M/G ratio of the alginate influences the final mechanical properties of the materials [5,6].Considering this, the M/G ratio of the used alginate was determined by 1 H-NMR.The yield of the synthesis of methacrylated alginate was 78%.For the determination of the methacrylation degree (%MD), as suggested by Jensen et al. [25], MALG samples 1 H-NMR spectra were acquired at an elevated temperature of 80 • C for two main reasons.Firstly, at 80 • C, the sample viscosity decreased, resulting in narrower signal widths.Secondly, the water resonance signal shifted out of the spectral region of interest (Figure 1), originally positioned between 4.5-5 ppm.This shift is very relevant as it resolves the overlap with signals from the anomeric protons in alginate G and M units, hindering the determination of %MD through these integration methodologies.

Contact Angle Measurements
The water wettability of the individual films is highly dependent on the surface roughness but also on their chemical functionalities and possible additives involved in their formulations.Water contact angle (WCA) results were obtained as the mean values of five measurements, and some representative images of sessile water drops are presented in Figure 3.After analysing five samples, the proportion of G units was found to be 65 ± 1%, resulting in an M/G ratio of 0.539 and a %MD = 44.4± 2.9%.

Contact Angle Measurements
The water wettability of the individual films is highly dependent on the surface roughness but also on their chemical functionalities and possible additives involved in their formulations.Water contact angle (WCA) results were obtained as the mean values of five measurements, and some representative images of sessile water drops are presented in Figure 3.

Contact Angle Measurements
The water wettability of the individual films is highly dependent on the surface roughness but also on their chemical functionalities and possible additives involved in their formulations.Water contact angle (WCA) results were obtained as the mean values of five measurements, and some representative images of sessile water drops are presented in Figure 3.The water contact angle results evidenced the decrease in the hydrophilicity of these polysaccharide-based films (Figure 3) related to the GO increase in the formulation.The The water contact angle results evidenced the decrease in the hydrophilicity of these polysaccharide-based films (Figure 3) related to the GO increase in the formulation.The lowest value of WCA (39.2 • ) was obtained for pure alginate films.Furthermore, due to the poor physical crosslinking of this highly hydrophilic biopolymer, the sessile drop was rapidly absorbed by the film (local dissolution of alginate), leaving a hole on the surface.The stability problem of water drops was solved once chemical crosslinking was introduced into these systems.A first crosslinking approach consisted of three-dimensional radical homopolymerization of methacrylate alginate induced by UV-irradiation, which allowed to obtain higher WCA values (52.40 • ), but with the persistence of partial dissolution of the substrate on which the water drops were placed.On the other hand, by using a small amount of AA, the crosslinking process consisting of three-dimensional radical copolymerization of methacrylated alginate with AA most likely led to higher crosslinking density, with a direct consequence in increasing the value of contact angle up to 69.40 • (the highest value).Instead, incorporating 1.25 wt.% of graphene oxide (GO) as a hydrophilic filler to enhance mechanical properties and dye adsorption capacity determined a reduction of water contact angle (WCAs) to 59 • .At the same time, increasing GO content from 1.25 to 5% with respect to the MALG amount led to higher values of contact angle of 62.60 • , due to the potential for physical crosslinking of these materials with the alginate-based matrix and the barrier effect of the impermeable GO nanoplatelets.
2.3.Mechanical Properties of MALG/AA and MALG/AA/GO Films Dynamic mechanical thermal analysis (DMTA) was conducted on the self-standing films composed of MALG/AA, with various quantities of graphene oxide (GO) in their formulations.The results are presented in Figure 4, with the determined storage modulus values at two particular temperatures gathered in Table 1.
tion of the substrate on which the water drops were placed.On the other hand, by using a small amount of AA, the crosslinking process consisting of three-dimensional radical copolymerization of methacrylated alginate with AA most likely led to higher crosslinking density, with a direct consequence in increasing the value of contact angle up to 69.40° (the highest value).Instead, incorporating 1.25 wt.% of graphene oxide (GO) as a hydrophilic filler to enhance mechanical properties and dye adsorption capacity determined a reduction of water contact angle (WCAs) to 59°.At the same time, increasing GO content from 1.25 to 5% with respect to the MALG amount led to higher values of contact angle of 62.60°, due to the potential for physical crosslinking of these materials with the alginatebased matrix and the barrier effect of the impermeable GO nanoplatelets.

Mechanical Properties of MALG/AA and MALG/AA/GO Films
Dynamic mechanical thermal analysis (DMTA) was conducted on the self-standing films composed of MALG/AA, with various quantities of graphene oxide (GO) in their formulations.The results are presented in Figure 4, with the determined storage modulus values at two particular temperatures gathered in Table 1.Due to the high hydrophilicity of these systems, it was noticed that the ambient humidity absorbed by the films before the DMTA experiments influenced their mechanical properties.This was evidenced by an increase in the storage modulus from 25 to 100 °C for all samples as the water content diminished.For this reason, two results of the storage modulus are presented, one at 25 °C and one at 100 °C.The latter values are considered  Due to the high hydrophilicity of these systems, it was noticed that the ambient humidity absorbed by the films before the DMTA experiments influenced their mechanical properties.This was evidenced by an increase in the storage modulus from 25 to 100 • C for all samples as the water content diminished.For this reason, two results of the storage modulus are presented, one at 25 • C and one at 100 • C. The latter values are considered indicative of completely dried films, reflecting the observed maximum storage modulus.It could be observed that the introduction of the fillers influenced the mechanical properties of the films.A substantial increase in the storage modulus was measured for film specimens bearing GO when compared to those without it.This effect was more evident at higher GO proportions, with values almost three times higher than those for the pristine polymer at the two presented temperatures.The high compatibility of GO functionality with the biopolymeric matrix facilitated more intimate filler-matrix interactions, thereby reinforcing the final structure of the material.

Dye Adsorption Onto MALG/AA and MALG/AA/GO Films
The potential applicability of these films for dye decontamination was monitored by UV-VIS spectroscopy.Thus, allowing for the calculation of the adsorption capacity per gram of material and the total amount of dye adsorbed or adsorption yield, the results are presented in Figure 5.
The potential applicability of these films for dye decontamination was monitored by UV-VIS spectroscopy.Thus, allowing for the calculation of the adsorption capacity per gram of material and the total amount of dye adsorbed or adsorption yield, the results are presented in Figure 5.The capacity for MB adsorption appeared to be related to the total water incorporation capacity of the fabricated films, following the same trend as in the swelling behavior (Figure 6).MALG/AA films showed higher capacities (17 mg MB/g film).In contrast, the capacities for the more intricate structures of MALG/AA/GO films were 10 mg MB/g film for 1.25 wt.% GO containing films and 12 mg MB/g film for films with 5 wt.% of GO, evidencing a compensation effect due to the MB molecules incorporated through interactions inbetween the GO layered structure.The apparent low reproducibility, as shown by large error bars (Figure 5a), was found to be most likely due to the significant influence of film thickness (or mass) exerted on this parameter.Thus, the reproducibility of the thickness of the films was low during the different sample fabrications due to the ad hoc molds used and the evaporation of the solvent during the crosslinking reaction.In other words, samples with different thicknesses produced a broad dispersion of adsorption capacity results, so that a better control over film thickness is expected to improve the consistency of measurements.However, the total efficiency for MB adsorption was increased in films with GO content from 88% of MALG/AA films to 90% on those bearing 1.25 wt.% of GO and 97% on those with a 5 wt.% of GO.Such an ascending tendency of MB adsorption with increasing GO content could be due to the great affinity of this dye to GO via multiple interaction types (hydrogen bonding, electrostatic interactions, π-π stacking), as suggested in Figure 7 [26].The capacity for MB adsorption appeared to be related to the total water incorporation capacity of the fabricated films, following the same trend as in the swelling behavior (Figure 6).MALG/AA films showed higher capacities (17 mg MB/g film).In contrast, the capacities for the more intricate structures of MALG/AA/GO films were 10 mg MB/g film for 1.25 wt.% GO containing films and 12 mg MB/g film for films with 5 wt.% of GO, evidencing a compensation effect due to the MB molecules incorporated through interactions inbetween the GO layered structure.The apparent low reproducibility, as shown by large error bars (Figure 5a), was found to be most likely due to the significant influence of film thickness (or mass) exerted on this parameter.Thus, the reproducibility of the thickness of the films was low during the different sample fabrications due to the ad hoc molds used and the evaporation of the solvent during the crosslinking reaction.In other words, samples with different thicknesses produced a broad dispersion of adsorption capacity results, so that a better control over film thickness is expected to improve the consistency of measurements.However, the total efficiency for MB adsorption was increased in films with GO content from 88% of MALG/AA films to 90% on those bearing 1.25 wt.% of GO and 97% on those with a 5 wt.% of GO.Such an ascending tendency of MB adsorption with increasing GO content could be due to the great affinity of this dye to GO via multiple interaction types (hydrogen bonding, electrostatic interactions, π-π stacking), as suggested in Figure 7 [26].For comparison and a better understanding of this process of MB adsorption, a brief collection of composites accompanied by specific mechanical properties and corresponding MB adsorption yields is listed in Table 2.For comparison and a better understanding of this process of MB adsorption, a brief collection of composites accompanied by specific mechanical properties and corresponding MB adsorption yields is listed in Table 2.In addition, swelling-drying behavior was also evaluated, the results for each formulation being displayed in Figure 7.
The swelling-drying results evidenced good stability of the films, barely losing mass due to an almost negligible breaking or decomposition of the sample during the process.Additionally, the films presented a remarkable ability to spontaneously dry at ambient temperature, losing weight during the drying stage considerably faster, hinting at a good reusability in water-related applications.The incorporation of GO in the MALG/AA films had the effect of hindering the water adsorption, which led to the most substantial decrease of the equilibrium swelling for the films with 1.25% GO by ca.33% compared to the equilibrium swelling of the films devoid of GO (Figure 7).This effect was slightly compensated by increasing the GO content to 5% and, as a result of the hydrophilic nature of the filler, the reduction in the equilibrium swelling of these systems compared to that of the films without GO was only 17%.During the drying stage, the films without GO were able to recover their initial weight in 30 min at ambient temperature in a ventilated room.Also, the same films showed shorter drying times, being 20 min for MALG/AA/GO1.25 films and only 15 min for MALG/AA/GO5 films.For this last case, the drying time of 15 min was equal to that characteristic of the entire swelling phase.The results revealed higher Gels 2024, 10, 25 8 of 13 evaporation rates of water molecules physically bonded to the GO filler-containing films than those bonded to the alginate-acrylic acid matrix, the effect being more pronounced at an elevated filler content.

Absorbed MB Extraction from MALG Films in Acidic Conditions
To continue with the reusability characterization of the fabricated films, the possibility of extracting the absorbed MB was studied.MB bears ionizable tertiary amine functionalities in its chemical structure under acidic conditions (Figure 8) [30].Thus, the selected experimental conditions for MB extraction include submerging the films in HCl solutions (0.01 M, pH = 2) for predefined periods of time when the dye positively charged is less and less adsorbed on an alginate matrix with an electrical charge increasingly close to zero in such an acidic medium, as reported by Othman et al. [31] and also employed by Fang et al. [32].
the filler, the reduction in the equilibrium swelling of these systems compared to that the films without GO was only 17%.During the drying stage, the films without GO we able to recover their initial weight in 30 min at ambient temperature in a ventilated roo Also, the same films showed shorter drying times, being 20 min for MALG/AA/GO1.films and only 15 min for MALG/AA/GO5 films.For this last case, the drying time of min was equal to that characteristic of the entire swelling phase.The results reveal higher evaporation rates of water molecules physically bonded to the GO filler-containi films than those bonded to the alginate-acrylic acid matrix, the effect being more pr nounced at an elevated filler content.

Absorbed MB Extraction from MALG Films in Acidic Conditions
To continue with the reusability characterization of the fabricated films, the possib ity of extracting the absorbed MB was studied.MB bears ionizable tertiary amine fun tionalities in its chemical structure under acidic conditions (Figure 8) [30].Thus, the s lected experimental conditions for MB extraction include submerging the films in H solutions (0.01 M, pH = 2) for predefined periods of time when the dye positively charg is less and less adsorbed on an alginate matrix with an electrical charge increasingly clo to zero in such an acidic medium, as reported by Othman et al. [31] and also employed Fang et al. [32].The MB desorption (assessed via UV-Vis spectrometry) was conducted under acid conditions (Figure 9), determining the MB content in desorption solutions, namely, H aqueous solutions of pH 2, over a period of time of 258 h by using MALG/AA/GO5 film This specific formulation was chosen due to its demonstrated higher adsorption efficien within the same timeframe (Figure 5).These tests found good stability of the films in acid media (pH = 2) during exposure time.However, as can be seen in Figure 10, only 42% the total MB adsorbed by the MALG/AA/GO5 film was extracted over 258 h of immersi with a pronounced extraction during the first 24 h (ca.26%), which indicates that a furth improvement to these methodology needs to be achieved for a more efficient extraction the initially adsorbed dye.The MB desorption (assessed via UV-Vis spectrometry) was conducted under acidic conditions (Figure 9), determining the MB content in desorption solutions, namely, HCl aqueous solutions of pH 2, over a period of time of 258 h by using MALG/AA/GO5 films.This specific formulation was chosen due to its demonstrated higher adsorption efficiency within the same timeframe (Figure 5).These tests found good stability of the films in acidic media (pH = 2) during exposure time.However, as can be seen in Figure 10, only 42% of the total MB adsorbed by the MALG/AA/GO5 film was extracted over 258 h of immersion with a pronounced extraction during the first 24 h (ca.26%), which indicates that a further improvement to these methodology needs to be achieved for a more efficient extraction of the initially adsorbed dye.

Additional Nanocomposite MALG/AA/GO Films Characterizations
Additional characterization techniques employed during the study of the fabricated films, such as cross-sectional SEM images and powder XRD diffractograms, are provided in the supporting information.The XRD diffractogram for a methacrylated alginate film (Figure S1) shows a characteristic highly amorphous signal for alginate with a maximum of around 26 • with a shoulder to his left around 12-15 • , corresponding to the GO maximum peak.This diffractogram is similar to others reported in the literature for alginate-based films containing GO [28].Meanwhile, in the cross-sectional SEM images provided in Figure S2, the thickness of two film samples can be measured, resulting in 21 µm for pristine MALG film and 29 µm for a MALG film containing 5 wt.% of GO.Additionally, the platellets of GO, mostly oriented parallel to the film surfaces, can be observed in the images for the MALG/GO5 films.

Additional Nanocomposite MALG/AA/GO Films Characterizations
Additional characterization techniques employed during the study of the fabricated films, such as cross-sectional SEM images and powder XRD diffractograms, are provided in the supporting information.The XRD diffractogram for a methacrylated alginate film (Figure S1) shows a characteristic highly amorphous signal for alginate with a maximum of around 26° with a shoulder to his left around 12-15°, corresponding to the GO maximum peak.This diffractogram is similar to others reported in the literature for alginatebased films containing GO [28].Meanwhile, in the cross-sectional SEM images provided in Figure S2, the thickness of two film samples can be measured, resulting in 21 µm for pristine MALG film and 29 µm for a MALG film containing 5 wt.% of GO.Additionally, the platellets of GO, mostly oriented parallel to the film surfaces, can be observed in the images for the MALG/GO5 films.

Conclusions
This research work probed the possibility of easily fabricating alginate-based hydrogel-like films to treat dye-contaminated wastewater.The use of these adsorption technologies has already shown high efficiencies for dye removal, which, together with their composition based on abundant and inexpensive biopolymers, may produce new solutions that are affordable even for developing countries where these problems tend to be more acute.However, alginate films lack the mechanical properties to withstand their manipulation and integration in complex water treatment systems.Thus, this research showed how the inclusion of graphene oxide in alginate-based films can greatly improve their mechanical properties, leading to storage modulus values increased by a factor of 2.5.This improvement may allow, for example, the use of these self-standing films in applications

Conclusions
This research work probed the possibility of easily fabricating alginate-based hydrogellike films to treat dye-contaminated wastewater.The use of these adsorption technologies has already shown high efficiencies for dye removal, which, together with their composition based on abundant and inexpensive biopolymers, may produce new solutions that are affordable even for developing countries where these problems tend to be more acute.However, alginate films lack the mechanical properties to withstand their manipulation and integration in complex water treatment systems.Thus, this research showed how the inclusion of graphene oxide in alginate-based films can greatly improve their mechanical properties, leading to storage modulus values increased by a factor of 2.5.This improvement may allow, for example, the use of these self-standing films in applications with higher water flow rates.Furthermore, the affinity of GO for some dyeing products, such as methylene blue, that can be trapped onto/between GO layers, has the effect of greatly increasing their adsorption efficiency.Although some studies have been made for the potential reusability of these films, with good results for their drying times (35 min) once unsubmerged, improvements on the absorbed dye extraction strategy need to be achieved, as the treatment with acidic solutions proved not to be enough by extracting barely half of the amount of the absorbed dye.

Methacrylation of Polysaccharides
Methacrylated alginate was synthesized through esterification of hydroxyl groups of alginate with MA.Briefly, a 2% w/v SA solution in deionized water was prepared and adjusted to pH = 8 using 5 M aqueous solution of NaOH.In the course of the first two hours of reaction time, MA at 20-fold excess was added slowly, while maintaining the temperature at 4 • C under vigorous stirring.During this time, pH was continuously monitored to be over a value of 8, adjusting when necessary by adding 5 M NaOH.Then, the solution was kept in a refrigerator for 22 h at 40 • C for the reaction to complete.Modified alginate was purified by dialysis (cellulose membranes, molecular weight cut-off −3.5 kDa, Medicell Membranes Ltd., London, UK) against 2 L of deionized water, with water periodically changed 8 times over 96 h to remove unreacted MA.By freeze-drying the dialyzed MA solution (−50 • C, 2.5 L Freeze Dryer, LAB-CONCO, Kansas City, MO, USA), the final product was obtained as a white sponge-like material with roughly 78% yielding.

Films Preparation
Initially, MALG solutions (10 mL, 2% w/v) were prepared in deionized water.These solutions were mechanically stirred for 24 h and stored at 4 • C until further use.Meanwhile, aqueous dispersions of GO (1 mg/mL) were prepared, mechanically stirred (400 rpm, 15 min), and bath-sonicated (60 min) to ensure adequate dispersions.Before adding GO dispersions over polymer solutions, they were subjected to ultrasound processing (10 min, 40% amplitude, 20 kHz).For each film formulation preparation (MALG/AA75, MALG/AA75/GO1.25, and MALG/AA75/GO5), where 75 refers to the weight percentage of AA with respect to the MALG content), an appropriate amount of filler aqueous dispersion (0-2.5-10mL) was added dropwise over the correspondent polysaccharide solution vigorous stirring to obtain certain weight percentages of GO with respect to the MALG amount (0, 1.25, and 5%, respectively).Then, a volume of 150 µL AA (AA75, 75 wt.%with respect to MALG amount) and 20 mg of the photo-initiator previously dissolved in 1 mL methanol were added onto every single MALG/GO aqueous dispersion under stirring.Finally, film-forming mixtures were introduced in a sonicating bath 10 additional minutes to ensure homogeneous dispersions of the components and eliminate bubbles that may produce defects during the curing process.The mixtures were then poured with the help of a syringe on the glass/Teflon molds built ad hoc (Figure 11).Each mold was then placed under a UV lamp (65 W, 365 nm) for 30 min.After the UV exposure, the residual solvent was evaporated by introducing the molds into an oven (65 °C) until they were completely dried.Films were recovered by peeling them from the glass substrates after removing pincers and Teflon rings.

Alginate and Methacrylated Alginate Characterization
1 H-NMR spectroscopy (500 MHz, D2O) was employed to determine the starting ALG mannuronic acid/guluronic residues ratio (M/G) as well as the modification degree (%MD) achieved during MALG synthesis (500 MHz, D2O, 80 °C) on a Bruker AVANCE 500 spectrometer (Billerica, MA, USA).Each mold was then placed under a UV lamp (65 W, 365 nm) for 30 min.After the UV exposure, the residual solvent was evaporated by introducing the molds into an oven (65 • C) until they were completely dried.Films were recovered by peeling them from the glass substrates after removing pincers and Teflon rings.

Figure 1 .
Figure 1.MALG 1 H-NMR spectra showing the displacement of water resonance signal in the experiment performed at 80 °C.

Figure 2 .
Figure 2. MALG 1 H-NMR spectra with the integrated signals used in the characterization of the G and M units in MALG.

Figure 1 . 1 )Figure 1 .
Figure 1.MALG 1 H-NMR spectra showing the displacement of water resonance signal in the experiment performed at 80 • C. All calculations for the M/G ratio and %MD were carried out following equations 1 and 2 and considering the integration of the signals for the anomeric protons in G (I G = 4.9-5.0ppm) and M (I M = 4.55-4.6ppm) units, as well as the signals for vinylic protons in the methacrylated units (I A = 5.6-5.7,I B = 6.0-6.1 ppm) (Figure 2): %G = I G I M + I G × 100% (1)

Figure 2 .
Figure 2. MALG 1 H-NMR spectra with the integrated signals used in the characterization of the G and M units in MALG.

Figure 3 .Figure 2 .
Figure 3. Values of the contact angle measured by the method of sessile water drops deposited on the film surfaces of (a) ALG, (b) MALG, (c) MALG/AA, (d) MALG/AA/GO1.25 and (e) MALG/AA/GO5.The water contact angle results evidenced the decrease in the hydrophilicity of these polysaccharide-based films (Figure3) related to the GO increase in the formulation.The

Figure 1 .
Figure 1.MALG 1 H-NMR spectra showing the displacement of water resonance signal in the experiment performed at 80 °C.

Figure 2 .
Figure 2. MALG 1 H-NMR spectra with the integrated signals used in the characterization of the G and M units in MALG.

Figure 4 .
Figure 4. DMTA thermograms for the prepared MALG/AA films without and with the incorporation of different GO contents.

Figure 4 .
Figure 4. DMTA thermograms for the prepared MALG/AA films without and with the incorporation of different GO contents.

Figure 5 .
Figure 5. Methylene blue adsorption capacity (a) and overall adsorption yield (b) of the prepared MALG films as a function of time of exposure to MB aqueous solutions.

5 Figure 5 .
Figure 5. Methylene blue adsorption capacity (a) and overall adsorption yield (b) of the prepared MALG films as a function of time of exposure to MB aqueous solutions.

Figure 6 .
Figure 6.Weight variations during swelling-drying for the specified films.

Figure 6 .
Figure 6.Weight variations during swelling-drying for the specified films.

Figure 7 .
Figure 7.A possible adsorption mechanisms of GO for MB, a positive charge is allocated in the S atom, as it is one of the possible resonance forms of the MB molecule.(Red: Funtional groups present in GO)

Figure 7 .
Figure 7.A possible adsorption mechanisms of GO for MB, a positive charge is allocated in the S atom, as it is one of the possible resonance forms of the MB molecule.(Red: Funtional groups present in GO).

Figure 10 .
Figure 10.Accumulated MB extraction from a MALG/AA/GO5 used film during 258 h of its immersion in different HCl aqueous solutions of pH 2.

Figure 9 .
Figure 9. Image of the desorption solutions during the course of MB desorption from MALG/AA/GO5 films in HCl (0.01 M, pH = 2) aqueous solutions.

Figure 9 .
Figure 9. Image of the desorption solutions during the course of MB desorption from MALG/AA/GO5 films in HCl (0.01 M, pH = 2) aqueous solutions.

Figure 10 .
Figure 10.Accumulated MB extraction from a MALG/AA/GO5 used film during 258 h of its immersion in different HCl aqueous solutions of pH 2.

Figure 10 .
Figure 10.Accumulated MB extraction from a MALG/AA/GO5 used film during 258 h of its immersion in different HCl aqueous solutions of pH 2.

14 Figure 11 .
Figure 11.(a) Film curing process on the glass/Teflon molds and (b) two of the obtained self-standing film specimens.

Figure 11 .
Figure 11.(a) Film curing process on the glass/Teflon molds and (b) two of the obtained self-standing film specimens.

Table 2 .
Published results on different mechanical properties and MB adsorption yields of some alginate-based systems.

Table 2 .
Published results on different mechanical properties and MB adsorption yields of some alginate-based systems.