Degradation of Acrylic Acid/acrylamide Superabsorbent Copolymer

Poly(potassium acrylate), P(KA), and poly[acrylamide-co-(potassium acrylate)], P(Am-co-KA), were synthesized and an effective degradation technique of the polymers via chemical and biological processes were pursued. Signicant reductions in dry mass and water absorbency were observed after P(KA) (53% and 54%, respectively) and P(Am-co-KA) (43% and 40%, respectively) were buried in the soil for ten weeks. The living fungal culture failed to degrade the polymers, but the enzymatic treatment using crude peroxidase (20 U/g) for 16 h signicantly decreased the dry mass (15%; 15.0±0.3 g) and water absorbency (13%; 16.0±1.0 g/g) of P(KA). Chemical oxidation using H 2 O 2 at high temperature with/without peroxidase eciently degraded P(KA) and P(Am-co-KA). The maximum degradation of P(KA) (99.84% weight loss) was obtained when incubated with 12.8% (v/w) H 2 O 2 at 65 ºC for 7.3 h while 98.43% weight loss was achieved after P(Am-co-KA) was incubated with 14.8% (v/w) H 2 O 2 at 68 ºC for 9.2 h. No signicant inhibition was observed in seed germination of mung bean grown on the untreated polymers but sweet corn was slightly inhibited. The effects of degraded products on mung bean germination were not signicantly different from the control and untreated polymers. On sweet corn, the degraded products were apparently less toxic than did the untreated polymers. These results suggested that the rapid and ecient degradation of polyacrylate and its copolymer by the thermo-oxidation of H 2 O 2 could be applied for a larger scale of SAP waste management.


Introduction
Superabsorbent polymers (SAPs) are basically the hydrophilic three-dimensional network of waterinsoluble polymer chains that can absorb uids such as water, electrolyte solution, and biological uids as high as 1,000 g/g of their own dried weight whereas the absorption capacity of common cotton pulp is not more than 1 g/g [1,2]. Thus, SAPs are employed for a multitude of applications. Commercial production of SAPs began in 1978 for use in feminine napkins and further employing in baby diapers [3], agriculture [4] and commodity products [5]. Currently, the uses of SAP for biomedical purposes such as wound healing and drug carrier are the main goal of their applications [6].
SAPs are prepared on a laboratory scale from a variety of speci cally hydrophilic petrochemical-based monomers and natural polymers that contain hydroxyl, carboxylic, amide, sulphonic groups etc., such as, poly(acrylic acid), polyacrylamide, starch, cellulose, etc. On the industrial scale of SAPs, acrylic acid (AA), its sodium or potassium salts, and acrylamide (AM) are the major and common monomers widely employed for the production of PA or the copolymer of acrylate and amide moieties [3]. The desired features of PAs are fast absorption and swelling, large water absorption capacity, high swelling capacity, good mechanical strength of the swollen gel [6] and adequate release rate of retained water as moisture.
Due to their outstanding properties, most of the SAPs available on the market are PA-based products in both homopolymeric and copolymeric forms [7]. The market for the PAs is predicted to continually grow as a result of rise of global population and changes in human lifestyle [8]. As the use of PA disposable items has increased, environmental problems related to the waste disposal have been concerned and needed prompt remedies [5].
Although both PA homopolymer and the copolymer between AA and AM (AA-AM) are reportedly biodegradable in soil, the degradation rates of the PA main chain are very slow and the calculated amount was only 0.12 to 0.24% (w/w) per six months [9,10]. In addition, Mai et al., [11] reported that the mineralization rate of PA ranged from 4 to 7% (w/w) after one year. Based on those reports, the biological degradation amount and rate of PAs have been considered too slow and too low compared to a much larger amount of PA wastes deposited each year. Consequently, the quick accumulation of PA wastes in land lls raises serious environmental concerns. Therefore, an alternative and "green" technique that can e ciently shorten the degradation period is much desired.
On the degradation of superabsorbent polymers, one has to apply an effective treatment method to break the covalent crosslinked network in order to improve their mineralization either by chemical or biological means or even both means. To the best of the available information currently, there are very few documents reporting on effective non-biological degradations of PA. Thus, the purpose of the current study is to explore a new and effective method for the degradations of homopolymer of AA (potassium salt) and its copolymer of AA/AM. Due to the acidic nature of the PA gels, hydrogen peroxide was selected as the chemical oxidizer. The oxidation of the PA polymers was carried out at various temperatures and the degradation was assessed as the loss of hydrogel property. The contributions of biological factors, both living white rot fungus and its peroxidase enzyme, on the degradation were also evaluated. The structures of the degraded PAs were observed via FTIR and 1 H-NMR analyses. Phytotoxicity of the degraded products was also investigated by seed germination test. Here we reported the rapid, e cient, and environmentally safe protocol for the degradation of P(KA) superabsorbent polymers that can be further optimized and implemented in a larger scale of PA waste treatment.

Synthesis of acrylate superabsorbent polymers
Two types of superabsorbent polymers, poly(potassium acrylate) (P(KA)) and poly[acrylamide-co-(potassium acrylate)] (P(Am-co-KA)), were synthesized via a free radical chain polymerization in aqueous solution of monomers with MBA cross-linker while APS and TEMED were used as an initiator and a coinitiator, respectively. All the reactions were performed under N 2 atmosphere. A KA solution was prepared by neutralization of 27% (v/v) AA in deionized water (110 mL) with 40 mL of 30% (w/v) KOH. The suspension was stirred at room temperature (25±3°C) for 90 min and then the polymerization of KA was conducted by adding APS (2.7 mg), MBA (17.2 mg) and TEMED (0.18 mL). The reaction was heated up to 65°C with vigorous stirring for 90 min to form a viscous gel.
For the copolymer synthesis of P(Am-co-KA), the monomer solution was prepared by dissolving AM (3 g) in KA solution at the ratio of 5:95 and stirring at room temperature (28±3°C) for 90 min. APS (14.3 mg), MBA (16.1 mg) and TEMED (0.18 mL) were then added and stirred at 65°C for 10 min to form a viscous gel. Each viscous gel was soaked in acetone to remove the absorbed water and unreacted monomers.
The gels were oven dried at 65°C until a constant weight was reached, and then ground to reduce the particle sizes and collected the size portions by sieving through a set of ASTM E11-certi ed steel test sieve of 63, 108 and 212 mm (Retsch, Germany).

Polymer structure and property determination
The functional groups of P(KA) and P(Am-co-KA) were examined by Fourier transform infrared spectroscopy (FTIR; Perkin Elmer, Spectrum One, USA) with KBr disc for a direct compression over a wavenumber ranging from 400 to 4,000 cm −1 , with 32 scans. For water absorbency analysis, the dry P(KA) and P(Am-co-KA) weighing 0.1 g were each immersed in deionized water (500 mL) at room temperature for 24 h to reach an equilibrium swelling. The residual water was removed by ltration through an 80-mesh stainless steel screen. The retained polymer was left on the screen for at least 1 h to completely drain off the unabsorbed water. The water absorption was determined using Eq. (1) as follows: Water absorbency (g/g) = (W s - where W d is the weight of the dried SAP and W s is the weight of the swollen sample.

Biological degradation
The biodegradations of P(KA) and P(Am-co-KA) were carried out using a soil burial method (Sharma et al., 2014). Five samples (150 g each) of each swollen gel were separately put in a 14 ´ 14 cm linen bag and buried in the natural garden soil at a depth of ten cm with a three-cm spacing between bags. The soil moisture was maintained by adding 200 mL of tap water every 24 h. To determine the extent of biodegradability, the samples were taken out from the soil after ten weeks, rinsed thoroughly with distilled water to remove the contaminated soil particles and dried in hot air oven at 60°C till a constant weight was obtained. The percentage of weight loss was evaluated based on the following Eq. (2): where W i and W f are dry weights of the polymer before and after the treatment, respectively. The structural changes were determined by comparing the FTIR spectra of each sample before and after the biological degradation. Further the changes in the morphological appearance were evaluated by scanning electron microscopy (SEM) and the presence of element was identi ed by energy dispersive X-ray spectrometry (EDXS) (JSM-6610; JEOL, Tokyo, Japan).

Effects of peroxidase, hydrogen peroxide and temperature on SAP degradation
The equilibrium swellings of each P(KA) and P(Am-co-KA) were performed by soaking 0.1 g of each SAP in deionized water (100 mL) for 24 h, ltrated through an 80-mesh screen and the products were used as initial polymers for the following studies. These swollen SAPs were incubated in a culture of T. polyzona (100 mL of minimal medium) under a static condition at room temperature (28±3°C) for 20 days. Then, the treated SAPs were ltrated and dried. The degradation e ciency was investigated by the weight loss and water absorbency as described above. To evaluate the enzymatic degradation, the crude peroxidase at different concentrations (0, 5, 10, 15 and 20 U/g) in the same working volume of 50 mL was incubated with each SAP at room temperature for 16 h. The effect of hydrogen peroxide pretreatment was investigated by adding 50 mL of hydrogen peroxide with the different concentrations at 0%, 4%, 8% and 12% (v/w of the dry polymer) to each SAP and incubated at room temperature for 2 h. Weight loss and water absorbency were analyzed from each treatment and the best effective concentration of hydrogen peroxide was used as a basis in the next optimization step. The impact of temperatures was investigated by incubation of the SAP with hydrogen peroxide at the selected concentration for 2 h at different temperatures from 40 to 60°C. The degradation was evaluated according to weight loss and water absorbency reduction as mentioned above. All experiments were performed in triplicate and the results shown are means ± one standard deviation.

Optimization of SAP degradation
The optimal conditions for SAP degradation were separately determined for the swollen P(KA) and P(Amco-KA) by the Response Surface Methodology (RSM). The Box-Behnken experimental design with three levels of the three variable factors including three replicates at the center point is shown in Table 1 (Box and Behnken, 1960). The percentages of SAP degradation were taken as the response in each reaction. Statistical analysis of the data was performed by Design-Expert (version 8.0.7.1, Stat-Ease, Inc., Minneapolis, USA) to evaluate the analysis of variance (ANOVA) and to determine the signi cance of each term in the equations. The tted polynomial equation was then expressed in the form of threedimensional plots to illustrate the main and interactive effects of the independent variables on the dependent variables. To verify the accuracy of the predicted model, the experiment was repeated in triplicate using the predicted optimal condition. The structures of the treated SAPs at the optimal condition were identi ed by 1 H-NMR using a Bruker spectrometry (400 MHz).

Toxicity assessment of the products from SAP degradation
The phytotoxicities of the untreated and treated SAPs were evaluated via the seed germination test using mung bean (Vigna radiata L.), and sweet corn (Zea mays L.). The certi ed seeds were purchased from Chia Tai Co., Ltd. (Thailand) and imbibed in distilled water overnight before use. The experiments were carried out at room temperature by placing 25 seeds in each petri dish (15 ´ 100 mm.) containing 10 mL of either distilled water, the untreated or treated SAPs (5 g/L) and incubated for three days in the dark. All experiments were performed in ve replicates and the results obtained are means ± one standard deviation. The percentages of germination were calculated using the following Eq. monomer radicals, the covalently cross-linked networks of the resulting SAP are produced continuously until the polymerization reaction is terminated. For P(AM-co-KA), a copolymerization between Am and KA competes for the free radical centers to add more reactive monomer. The extent of KA and AM amounts covalently crosslinked network depends mainly on the reactivity of both monomers which also depends signi cantly on the reaction condition. In our case, AA was partially neutralized by KOH in an attempt to enhance the rate of polymerization and to reduce the residual monomers remained in the system [13,14]. The chain polymerization is caused by the radicals, generated by the thermal initiator (APS) under the heating condition, added to the double bonds of the vinyl monomers through the radical routes (KA with/without AM). Simultaneously, the double bonds in the MBA created the cross-linked structures [15]. The chemical structures of the obtained polymers were analyzed by FTIR-spectrophotometry and the IR spectra of P(KA) and P(Am-co-KA) are shown in Fig. 1 (dashed lines).
Both IR spectra were very similar between the two polymers and it might be due to the low amount of AM available in P ( The presence of absorbed water was seen by the band at 3350-3366 cm − 1 in all spectra. Water absorbency of the obtained polymers was determined according to Dispat et al. [17]. After 24-h of equilibrium swelling, P(KA) exhibited a slightly higher water absorbency (387 ± 12 g/g) than that of P(Amco-KA) (338 ± 16 g/g) which might be contributed by the greater degree of ionization of the carboxylic groups originated from KA solely in distilled water. Basically, the extent of water absorbency related directly to the cross-link density which controls the expansion of polymer network to enable penetration of water molecules [18]. Electrostatic interactions between the amino cations originated from AM and the carboxylate anions formed a rather rigid network structure that prohibited the expansion of polymer networks [19]. In addition, the water absorbency of both SAPs is in the same range of commercial SAPs that has been used in agricultural applications [20].

Biological degradation
Compared to another widely used poly(sodium acrylate), PKA is far more suitable for applications in agriculture, horticulture and soil care since it does not increase soil salinity [21]. Although there have been a number of reports on biodegradability of P(KA), the fate of P(KA) in the soil still cannot be clearly drawn, both chemically and biologically, and there have been of environmental concerns about its accumulation in agricultural lands [9][10][11]22]. Therefore, biodegradations of P(KA) and P(Am-co-KA) were evaluated after being buried in soil compared with the unburied one but incubated at room temperature for the same period. After ten weeks, the reductions of dry mass were 53 ± 3% for P(KA) and 43 ± 2% for P(Am-co-KA) compared to the unburied negative control ( Table 1).
The difference in the degradation rate between P(KA) and P(Am-co-KA) is probably related to the crystallinity of the two polymers [23] and the lower degradation rate of P(Am-co-KA) was suggestively attributed by the chain stiffness of its cross-linked structure. Changes in water absorbency of the buried SAPs were also in accordance with the weight loss. An approximately 54% (183 ± 7 g/g) and 40% (135 ± 11 g/g) of water absorbency were lost from P(KA) and P(Am-co-KA), respectively. Decrement in water absorbency of both SAPs could be described by the partial destruction of network linkages that led to the loss of polymer density and the generation of lower molecular weight fragments and disintegration of its chain exibility that could not have the same degree of water absorption capability [2].
To determine the changes in chemical structure of the buried polymers, FTIR spectra were compared between the treated SAP and the control, the unburied one (Fig. 1).

Effects of crude peroxidases, hydrogen peroxide and temperature on the SAP degradation
It has been reported that certain white rot fungi such as Phanerochaete chrysosporium and Pleurotus ostreatus can degrade some acrylic copolymers including P(guaiacol-Am-AA) and P(3,4-dihydroxybenzoic acid-AM-AA) within 15 − 18 days [11,25]. Therefore, the swollen P(KA) and P(Am-co-KA) were separately incubated in a liquid culture of the white rot fungus T. polyzona. After incubation for 20 days, T. polyzona did not cause a signi cant decrease in the dry weight and water absorbency compared to the controls incubated in the same medium without white rot fungus (Supplementary data I). The growth of the fungus was strongly inhibited by both SAPs and the activity of extracellular peroxidase that has been reported to oxidize a broad range of chemicals including lignin, PAM, and PAA into the unstable free radicals [2,22] was not detectable in the fungal cultures containing P(KA) and P(Am-co-KA). Therefore, the crude peroxidase was produced from T. polyzona and used to treat P(KA) and P(Am-co-KA) at different concentrations ranging from 5 to 20 U/g. A slight but signi cant reduction of the dry weight (6.5 and 15%) and water absorbency (8 and 13%) were observed in the P(KA) treated with the crude peroxidase at 15 and 20 U/g, respectively, while no signi cant difference was detected in the treated P(Am-co-KA) ( Table 2). The slight changes after 16-h peroxidase treatment suggested that a pretreatment step might be required prior to enzymatic digestion.
Most of the reported processes including mechanical, thermal, and chemical treatments to break the main chain (C-C) of PA and PAM have been involved with the activation of polymer by free radicals. The polymer radicals (P • ) are generated from the attack of free radicals at the secondary and tertiary carbons, as well as the carbon of primary amine. These polymer radicals reacted well with the dissolved oxygen gas to form the polymer peroxyl radicals (PO 2 • ) that could react to each other to generate the polymer fragments [26,27]. Hydrogen peroxide is recognized as the reactive oxygen species that can derive free radicals in water, especially under acidic condition, and considered as a sustainable or green liquid oxidant [27]. Under the conditions applied in this experiment, increasing the hydrogen peroxide concentration enhanced the degradation of P(KA) and P(Am-co-KA) as observed from the signi cant reductions of weight loss and water absorbency ( Table 3). The highest degree of degradation was found in the reaction with 12% (v/w) of hydrogen peroxide; therefore, this concentration was used for the further experiments.
The impact of temperatures was investigated by incubation of P(KA) and P(Am-co-KA) with hydrogen peroxide (12% (v/w)). The corresponding increments in weight loss of both SAPs were related to the rising temperature (Table 4). Although the weight loss of P(Am-co-KA) was not signi cantly different at 30ºC, 40ºC and 50ºC, the signi cant differences in water absorbency were observed in each temperature indicating that the structural changes did occur. These results corresponded well with the previous studies that reported a much higher rate of polyacrylamide chain scission reactions by free radicals under higher temperature conditions ranging from 50ºC to 80ºC [28,29]. After the incubation at 50ºC with 12% (v/w) of hydrogen peroxide for 2 h, the pretreated P(KA) and P(Am-co-KA) were subsequently incubated with the crude peroxidase at 20 U/g for 16 h at room temperature (28±3°C). However, the degradations of the pretreated SAPs with and without the enzyme were not signi cantly different (Supplementary data II); therefore, the crude peroxidase was omitted in the further optimization experiments.

Optimization of the SAP degradation
The optimum concentration of hydrogen peroxide, incubation times and temperatures for the degradation of P(KA) and P(Am-co-KA) were analyzed by RSM according to the Box-Behnken experimental design. The observed and predicted responses on weight loss of P(KA) and P(Am-co-KA) are shown in Tables 5  and 6, respectively. The maximum percentages of weight loss were found at 99.2 ± 6.2% (trial number 12 in Table 5) from the treated P(KA) and 99.7 ± 6.4% (trial number 10 in Table 6) from the treated P(Am-co-KA). The second-order regression equations that provided the percentage of weight loss as the function of variables are presented in terms of coded factors in the following equations: where Y 1 and Y 2 are the response values of weight loss (%) of P(KA) and P(Am-co-KA), respectively. A, B and C are hydrogen peroxide concentration (% v/w), time (h) and temperature (ºC), respectively. Comparison of the predicted and experimental values revealed a good correspondence between them. In this case, the model showed insigni cant lack of t (P 1 = 0.15 and P 2 = 0.83). It indicated that the secondorder model equation was adequate for the prediction of weight loss across the speci ed range of variables employed. According to the statistical analysis, the models in this study were signi cant. In addition, the coe cient of determination (R 2 ) was calculated to be in the range of 0.96 to 0.97, indicated that these models could explain more than 95% of the variability.
Three-dimensional response plots and their corresponding contour plots were drawn to investigate the interaction among the variables and to determine the optimum condition of each factor for the maximum degradation (Figs. 3 and 4). The canonical analysis revealed a complete degradation with >99% of the weight loss of P(KA) under the optimal condition of 12.8% (v/w) hydrogen peroxide at 65 ºC for 7.3 h while the predicted condition to optimally degrade P(Am-co-KA) was at 14.8% (v/w) hydrogen peroxide at 68 ºC for 9.2 h. The validity of predicted results by the regression model was con rmed by repeating the experiment under the optimal concentrations. The results obtained from three replications showed that the degradations of P(KA) (99.84 ± 4.40%) and P(Am-co-KA) (98.43 ± 5.95%) were close to each other and were not signi cantly different from the predicted value of 100.00 ± 1.9% and 98.98 ± 2.2%, respectively. It implied that the empirical models derived from RSM can be used to adequately describe the relationship between the factors and response in the degradations of P(KA) and P(Am-co-KA). After being treated under these optimized conditions, both polymers completely lost their hydrogel appearance and were turned into clear liquids suggesting that the chains were shorten and dissolved. When the optimized hydrogen peroxide treatment was applied to two different SAPs, signi cant reductions in dry weight and water absorbency were observed although they were not completely dissolved. This nding suggested that the process can be applied to other polyacrylate hydrogels but to achieve the maximal e ciency, further optimization speci c to each polymer is needed.
The structures of the treated SAPs under the optimal conditions were identi ed by 1 H-NMR and the spectra are shown in Fig. 5. The degraded products exhibited the peaks representing the carboxylate group (n-a, b; Fig. 5) at 1.6-1.8 ppm in both of P(KA) and P(Am-co-KA), and the amide group (m-a,b; Fig. 6B) at 2.2-2.4 ppm only in P(Am-co-KA). Since the degraded products from both SAPs were completely liquidized, it is suggestively indicated that these peaks represented the shortened chains of the SAPs. These treated products were used for the further assessment of their toxicity.
3.5. Toxicity assessment of the products from SAP degradation The phytotoxicity of the untreated and treated SAPs evaluated via seed germination is shown in Table 7. There were no signi cant differences in seed germination of mung bean among all the treatments.
However, the untreated P(KA) and P(Am-co-KA) moderately (36 %) to slightly (14 %) inhibited sweet corn germination suggesting the higher sensitivity of this plant to the polymers. Signi cantly less inhibition of seed germination was found in sweet corn treated with the degraded P(KA) and P(Am-co-KA) suggesting that the degraded products were less toxic to the plant than the untreated polymers. Overall, the results clearly indicated the less toxic nature of P(KA) and P(Am-co-KA) after thermo-oxidized by hydrogen peroxide, compared to the untreated controls, to germinating seeds of both dicotyledonous and monocotyledonous plant representatives. Our current research conforms to part of the Goal 15, the Department of Economic and Social Affairs, United Nations under the Sustainable Development on the point of "halt and reverse land degradation and halt biodiversity loss" [30].

Conclusion
In conclusion, the simple, rapid and effective thermo-oxidation for the degradation of P(KA) and P(Am-co-KA) was rstly reported. Both SAPs were completely liquidized within ten hours with little chemical and energy input. The degraded products obtained from the process were evidently safe for plants. Since this thermo-oxidation by hydrogen peroxide can also degrade other PA polymers in a preliminary attempt, it has a high potential to be further optimized for a larger scale for polyacrylate waste treatment.