Synthesis and Characterization of Slow-Release Fertilizer Hydrogel Based on Hydroxy Propyl Methyl Cellulose, Polyvinyl Alcohol, Glycerol and Blended Paper

In this study, biodegradable slow-release fertilizer (SRF) hydrogels were synthesized from hydroxyl propyl methyl cellulose (HPMC), polyvinyl alcohol (PVA), glycerol and urea (SRF1) and HPMC, PVA, glycerol, urea and blended paper (SRF2). The fertilizer hydrogels were characterized by SEM, XRD and FTIR. The swelling capacity of the hydrogels in both distilled and tap water as well as their water retention capacity in sandy soil were evaluated. The hydrogels had good swelling capacity with maximum swelling ratio of 17.2 g/g and 15.6 g/g for SRF1 and SRF2 in distilled, and 14.4 g/g and 15.2 g/g in tap water, respectively. The water retention capacity of the hydrogels in sandy soil exhibited higher water retention when compared with soil without the (SRFs). The soil with the hydrogels was found to have higher water retention than the soil without the hydrogels. The slow-release profile of the hydrogels was also evaluated. The result suggested that the prepared fertilizer hydrogels has a good controlled release capacity. The blended paper component in SRF2 was observed to aid effective release of urea, with about 87.01% release in soil at 44 days compared to the pure urea which was about 97% release within 4 days. The addition of blended paper as a second layer matrix was found to help improve the release properties of the fertilizer. The swelling kinetic of the hydrogel followed Schott’s second order model. The release kinetics of urea in water was best described by Kormeye Peppas, suggesting urea release to be by diffusion via the pores and channels of the SRF, which can be controlled by changing the swelling of the SRF. However, the release mechanism in soil is best described by first order kinetic model, suggesting that the release rate in soil is depended on concentration and probably on diffusion rate via the pores and channels of the SRF.


Introduction
The world population is growing rapidly and is expected to reach 9.5 billion by the end of 2050 [1]. Meanwhile industrialization, desertification and land degradation caused by heavy flooding have diminished arable land and consequently threatened global food security [2]. Amidst all these, global food requirements have risen and is likely to double by 2050 [3]. Many steps are being taken to resolve such issues so as to modify and improve our Gels 2021, 7, 262 3 of 16 an important concept which has not been reported in literature. Furthermore, the use of paper implies the possibility of recycling waste paper thus reducing pollution due to paper disposal. In addition, incorporation of paper in the coating makes it biodegradable. The novelty of using paper is its biodegradability so that after serving its purpose, it can be degraded without constituting another environmental nuisance.
In our continuous effort [38] to design materials that can effectively deliver urea to plants in a controlled manner, we present here the synthesis of SRF hydrogel based HPMC/PVA/Glycerol with blended paper as the second layer. Furthermore, the effects of the blended paper on release, swelling and water retention in soil were investigated.

Results and Discussions
2.1. Characterization 2.1.1. Surface Morphology Analysis SEM technique was used to observe the surface morphologies of the hydrogel (SRF1, SRF2) and the starting materials (HPMC), and the results are presented in Figure 1. From Figure 1A, it can be seen that HPMC powder is a porous spongy particle like materials which has a rough, less dense surface that is no more recognizable in both hydrogels (SRF1 and SRF2) because of adsorption of urea fertilizer to its surface (microstructure). This porous nature aids to increase the surface area and can enhance the adsorption capacity of HPMC by yielding more sites for more interaction. Figure 1B (SRF1) has a uniform and smooth surface without any crack except for the existence of a defect on the surface. The SRF2 ( Figure 1C) has a rough packed surface having no porous cracks on the surface, which may have enhanced its water uptake and retention and consequently the release capacity. The dense and compact structure of the hydrogel is due to the co-polymerization of the polymer and the composite blends resulting from the crosslinking and incorporation of the paper blends into the matrix. This is because polymer matrices can either be chemically cross-linked through covalent or hydrogen bonds [39]. HPMC, PVA and glycerol to formulate a second layer coating for controlled release of fertilizer is an important concept which has not been reported in literature. Furthermore, the use of paper implies the possibility of recycling waste paper thus reducing pollution due to paper disposal. In addition, incorporation of paper in the coating makes it biodegradable. The novelty of using paper is its biodegradability so that after serving its purpose, it can be degraded without constituting another environmental nuisance. In our continuous effort [38] to design materials that can effectively deliver urea to plants in a controlled manner, we present here the synthesis of SRF hydrogel based HPMC/PVA/Glycerol with blended paper as the second layer. Furthermore, the effects of the blended paper on release, swelling and water retention in soil were investigated.

Surface Morphology Analysis
SEM technique was used to observe the surface morphologies of the hydrogel (SRF1, SRF2) and the starting materials (HPMC), and the results are presented in Figure 1. From Figure 1A, it can be seen that HPMC powder is a porous spongy particle like materials which has a rough, less dense surface that is no more recognizable in both hydrogels (SRF1 and SRF2) because of adsorption of urea fertilizer to its surface (microstructure). This porous nature aids to increase the surface area and can enhance the adsorption capacity of HPMC by yielding more sites for more interaction. Figure 1B (SRF1) has a uniform and smooth surface without any crack except for the existence of a defect on the surface. The SRF2 ( Figure 1C) has a rough packed surface having no porous cracks on the surface, which may have enhanced its water uptake and retention and consequently the release capacity. The dense and compact structure of the hydrogel is due to the co-polymerization of the polymer and the composite blends resulting from the crosslinking and incorporation of the paper blends into the matrix. This is because polymer matrices can either be chemically cross-linked through covalent or hydrogen bonds [39].

FTIR Spectra
The FTIR spectra of the blended paper (BP), HPMC, PVA, Urea and SRF's (SRF1 and SRF2) are shown in Figure 2. In the spectrum of blended paper, peaks at 3272 cm −1 and 1028 cm −1 are associated with stretching vibration of O-H and C-O-C in C-C out of plane stretching vibration [37]. The peaks at 2759 cm −1 and 2114 cm −1 are assigned to -CH-in CH2 stretching vibration. The absorption band at 1425 cm −1 is ascribed to -CH2-stretching. In the spectrum of HPMC absorption bands at 3424 cm −1 is ascribed to the stretching vibration of O-H. The peak at 2887 cm −1 represents the stretching of CH2, and C=O stretching is observed at 1648 cm −1 . The band at 1452 cm −1 represents CH2 and 1048 cm −1 is assigned to C-O-C stretching [40]. In the spectrum of PVA, characteristic peaks of PVA occurred

FTIR Spectra
The FTIR spectra of the blended paper (BP), HPMC, PVA, Urea and SRF's (SRF1 and SRF2) are shown in Figure 2. In the spectrum of blended paper, peaks at 3272 cm −1 and 1028 cm −1 are associated with stretching vibration of O-H and C-O-C in C-C out of plane stretching vibration [37]. The peaks at 2759 cm −1 and 2114 cm −1 are assigned to -CH-in CH 2 stretching vibration. The absorption band at 1425 cm −1 is ascribed to -CH 2stretching. In the spectrum of HPMC absorption bands at 3424 cm −1 is ascribed to the stretching vibration of O-H. The peak at 2887 cm −1 represents the stretching of CH 2 , and C=O stretching is observed at 1648 cm −1 . The band at 1452 cm −1 represents CH 2 and 1048 cm −1 is assigned to C-O-C stretching [40]. In the spectrum of PVA, characteristic peaks of PVA occurred at 3358 cm −1 is assigned to -OH stretching vibration of hydroxyl group in alcohol, 2956 cm −1 -2873 cm −1 characterized the aliphatic -CH symmetric stretching vibration. 1726 cm −1 , 1450 cm −1 and 1158 cm −1 correspond to C=O carbonyl, -CH 2 , and C-O, C-O-C [30]. In the spectrum of Urea, two peaks at 3426 cm −1 and 3326 cm −1 are related to the stretching vibration of -NH and -OH group in urea. Furthermore, peaks appeared at 2799 cm −1 , 1586 cm −1 , 1453 cm −1 , 1148 cm −1 and 1001 cm −1 , indicating the -CH 2 stretching vibration, -NH angular deformation, CH stretching in -CH 2 aliphatic group and -C-O stretching in -C-O-C group. In the spectrum of SRF1 and SRF2 new peaks appeared around 3425 cm −1 and 3328 cm −1 in SRF1, and around 3689 cm −1 , 3616 cm −1 , and 3345 cm −1 in SRF2 after incorporation of urea into the hydrogel. These peaks may be attributed to the formation of intermolecular hydrogen bond between -OH group of HPMC and the carbonyl group of PVA and urea, which confirms the encapsulation of the urea in the matrix via physical entrapment and its interaction with the polymer (30). This interaction was indicated by the shifting of peaks to higher wavelengths and the carbonyl group of (PVA and HPMC) [41], and as well due to the presence of -NH 2 that is part of O=C-NH 2 functional group in urea [42]. Most of the characteristic peaks shifted to higher frequencies in both hydrogels (SRF1 and SRF2) because the hydrogen bond between water and glycerol was replaced with hydrogen bond between HPMC and glycerol which is weak [30]. The appearance of urea characteristic peaks in FTIR spectra of SRF1 and SRF2 confirms the successful encapsulation of urea fertilizer compound within the matrix hydrogel. at 3358 cm −1 is assigned to -OH stretching vibration of hydroxyl group in alcohol, 2956 cm −1 -2873 cm −1 characterized the aliphatic -CH symmetric stretching vibration. 1726 cm −1 1450 cm −1 and 1158 cm −1 correspond to C=O carbonyl, -CH2, and C-O, C-O-C [30]. In the spectrum of Urea, two peaks at 3426 cm −1 and 3326 cm −1 are related to the stretching vi bration of -NH and -OH group in urea. Furthermore, peaks appeared at 2799 cm −1 , 1586 cm −1 , 1453 cm −1 , 1148 cm −1 and 1001 cm −1 , indicating the -CH2 stretching vibration, -NH angular deformation, CH stretching in -CH2 aliphatic group and -C-O stretching in -C-O-C group. In the spectrum of SRF1 and SRF2 new peaks appeared around 3425 cm −1 and 3328 cm −1 in SRF1, and around 3689 cm −1 , 3616 cm −1 , and 3345 cm −1 in SRF2 after incorpo ration of urea into the hydrogel. These peaks may be attributed to the formation of inter molecular hydrogen bond between -OH group of HPMC and the carbonyl group of PVA and urea, which confirms the encapsulation of the urea in the matrix via physical entrap ment and its interaction with the polymer (30). This interaction was indicated by the shift ing of peaks to higher wavelengths and the carbonyl group of (PVA and HPMC) [41], and as well due to the presence of -NH2 that is part of O=C-NH2 functional group in urea [42] Most of the characteristic peaks shifted to higher frequencies in both hydrogels (SRF1 and SRF2) because the hydrogen bond between water and glycerol was replaced with hydro gen bond between HPMC and glycerol which is weak [30]. The appearance of urea char acteristic peaks in FTIR spectra of SRF1 and SRF2 confirms the successful encapsulation of urea fertilizer compound within the matrix hydrogel.   that HPMC is crystalline with crystalline peak. As it can be seen, all the samples exhibit typical cellulose diffraction peaks [43]. In the pattern of SRF2, a shift in the basalt diffraction peak from 29.3 to 29.4 was observed but became weaker. This might be due to high compatibility between the blended paper, HPMC that led to the insertion of paper into HPMC layers which occurred as a result of cationic exchange between hydrogen bond and COO [44]. This slight change indicates that urea was absorbed into the HPMC-blended paper matrix, which slowed the release of the nutrient. However, most of the characteristic peaks of blended paper and HPMC were seen to disappear, while in SRF1 no peak was observed with disappearance of characteristic peaks of HPMC. Similarly, a broad peak of amorphous structure of polymer present shows the compositions of organic materials formed and that the blended paper was dispersed in SRF [45].
Gels 2021, 7, x FOR PEER REVIEW a shift in the basalt diffraction peak from 29.3 to 29.4 was observed but became w This might be due to high compatibility between the blended paper, HPMC that le insertion of paper into HPMC layers which occurred as a result of cationic excha tween hydrogen bond and COO [44]. This slight change indicates that urea was ab into the HPMC-blended paper matrix, which slowed the release of the nutrient. Ho most of the characteristic peaks of blended paper and HPMC were seen to dis while in SRF1 no peak was observed with disappearance of characteristic peaks of Similarly, a broad peak of amorphous structure of polymer present shows the co tions of organic materials formed and that the blended paper was dispersed in SR

Swelling Studies of the Hydrogels in Different Solutions
Water absorption is one of the most important characteristics of SRF hydrog Figure 4 shows the water absorption in distilled and tap water. The results show the swelling ability in distilled water was higher compared to tap water, this cou a result of the increase in the osmotic pressure difference between the polymeric n and immersion medium. According to Donnan equilibrium theory, osmotic pre the driving force for swelling in hydrogel. The result of this study is consistent w findings of Enayatifard et al., which showed that the presence of a highly watercompound in HPMC matrix generated an additional osmotic gradient, thereby re in a faster rate of polymer swelling and a large increase in gel thickness [32].

Swelling Studies of the Hydrogels in Different Solutions
Water absorption is one of the most important characteristics of SRF hydrogels [45]. Figure 4 shows the water absorption in distilled and tap water. The results showed that the swelling ability in distilled water was higher compared to tap water, this could be as a result of the increase in the osmotic pressure difference between the polymeric network and immersion medium. According to Donnan equilibrium theory, osmotic pressure is the driving force for swelling in hydrogel. The result of this study is consistent with the findings of Enayatifard et al., which showed that the presence of a highly water-soluble compound in HPMC matrix generated an additional osmotic gradient, thereby resulting in a faster rate of polymer swelling and a large increase in gel thickness [32].  Similarly swelling degree might as well be influenced by the ions present in the sw ing medium, where by the presence of these ions (Mg 2+ , Ca 2+ and Na + ) increased the io crosslinking density of the matrix blend causing a decrease in water absorption as repor by [45]. This observation is common in ionic hydrogels static swelling experiments [46]. T swelling equilibrium of the SRF's hydrogel in both medium was attained at 72 h except SR in tap water which attained its equilibrium at 48 h with swelling rate of 17.2 (g/g) and 1 (g/g) for SRF1 and SRF2 in distilled water while 14.4 (g/g) and 15.2 (g/g) for SRF1 and SR in tap water respectively. The swelling rate of SRF2 compared to SRF1 in distilled and water was high. This might be as a result of the additional hydroxyl group found in netw of cellulose paper, since paper network contains randomly laid cellulose fibrous and n fibrous (filler) materials which contains complicated set of cavity pore channels with vari of capillaries dimension, which make its readily permeable to liquids [37].

Swelling Kinetics
The swelling kinetics of the hydrogels in tap and distilled water was determined a the result is shown in Figure 5. A straight-line graph was obtained from the plot of t versus t with a good linear correlation coefficient as given in Table 1. Seq and Kis w calculated from the slope and intercept obtained from fitted straight lines. The res shows that the swelling behavior of hydrogel in distilled and tap water matched Schott's second order swelling kinetic model. The data presented for Seq and Kis in water were lower compared to that in distilled water. Similarly swelling degree might as well be influenced by the ions present in the swelling medium, where by the presence of these ions (Mg 2+ , Ca 2+ and Na + ) increased the ionic crosslinking density of the matrix blend causing a decrease in water absorption as reported by [45]. This observation is common in ionic hydrogels static swelling experiments [46]. The swelling equilibrium of the SRF's hydrogel in both medium was attained at 72 h except SRF2 in tap water which attained its equilibrium at 48 h with swelling rate of 17.2 (g/g) and 15.6 (g/g) for SRF1 and SRF2 in distilled water while 14.4 (g/g) and 15.2 (g/g) for SRF1 and SRF2 in tap water respectively. The swelling rate of SRF2 compared to SRF1 in distilled and tap water was high. This might be as a result of the additional hydroxyl group found in network of cellulose paper, since paper network contains randomly laid cellulose fibrous and non-fibrous (filler) materials which contains complicated set of cavity pore channels with variety of capillaries dimension, which make its readily permeable to liquids [37].

Swelling Kinetics
The swelling kinetics of the hydrogels in tap and distilled water was determined and the result is shown in Figure 5. A straight-line graph was obtained from the plot of t/St versus t with a good linear correlation coefficient as given in Table 1. S eq and K is were calculated from the slope and intercept obtained from fitted straight lines. The result shows that the swelling behavior of hydrogel in distilled and tap water matched the Schott's second order swelling kinetic model. The data presented for S eq and K is in tap water were lower compared to that in distilled water.

Water Retention Properties of the SRF Hydrogel in Soil
Water retention of hydrogel is an important feature of SRF in soil. Figure 6 sh water retention behaviour of sandy soil with and without SRF. By adding the hyd soil, the water retention increased for some time while the water content of soil the SRF completely vaporized after the 6th day. The water retention of soil co SRF2 was high compared to the soil with SRF1. The water retention ratio of soil SRF hydrogel remained 51.9 and 0.04 on the 2nd and 5th days, respectively, w with SRF1 and SRF2 remained 54.6, 0.8 and 56.2, 1.0, respectively. A similar phen was also observed by others [47]. This study finds out that the addition of blende as second layer in SRF2 helps in increasing the water retention in soil and decreas transpiration. That is, it withholds sufficient water, which would aid its efficien tion for crop cultivation thereby minimize irrigation water for farming.

Water Retention Properties of the SRF Hydrogel in Soil
Water retention of hydrogel is an important feature of SRF in soil. Figure 6 shows the water retention behaviour of sandy soil with and without SRF. By adding the hydrogel in soil, the water retention increased for some time while the water content of soil without the SRF completely vaporized after the 6th day. The water retention of soil containing SRF2 was high compared to the soil with SRF1. The water retention ratio of soil without SRF hydrogel remained 51.9 and 0.04 on the 2nd and 5th days, respectively, while that with SRF1 and SRF2 remained 54.6, 0.8 and 56.2, 1.0, respectively. A similar phenomenon was also observed by others [47]. This study finds out that the addition of blended paper as second layer in SRF2 helps in increasing the water retention in soil and decreases water transpiration. That is, it withholds sufficient water, which would aid its efficient utilization for crop cultivation thereby minimize irrigation water for farming.

Release Profile of SRF Hydrogels and Their Kinetics of Release
The cumulative release rate of pure urea and the hydrogel fertilizers; SRF 1 and SR in water and soil is shown in Figures 7 and 8, respectively. The result indicates 99% a 97% release of pure urea used as control in water and soil within one hour and four da respectively. The urea release in distilled water; being a static medium was faster co pared to that in soil. A cumulative release of approximately 37%, 82.2% and 85.4% we observed for SRF1 in water at 1, 6 and 24 h, while 28.6%, 59.6% and 64.4% were observ for SRF2 at the same time in water. The result suggests that the blended paper and HPM used as second layer coating may be responsible for slower release observed in SRF2 co pared to SRF1. The blended paper in the matrix might have created a physical barrier the form of membrane resistance of the matrix, thereby withstanding osmotic pressu and causing a slower release by diffusion as a result of the concentration or pressure g dient or both [37,48,49]. The cumulative urea release in soil is shown in Figure 8, the lease is significantly lower than that in water. The sustained release rate observed in s is expected, unlike in distilled water which is static the soil medium is dynamic. For SR cumulative release rates of 27.1%, 64.5% and 67.1% were observed at 6, 30 and 40 da respectively. The cumulative release of urea in SRF1 in soil reaches 67.1% at 44 days. Co paratively, SRF2; the hydrogel with blended paper and HPMC as second layer coati had a higher cumulative percentage release of 30.8%, 82.3% and 87.0% at 6, 30 and 44 da respectively. The blended paper in SRF2 might have aid absorption of water thereby d solving the urea in the matrix and consequently resulted in maximum release of 87% the encapsulated urea at 44 days compared to 67.1% of urea in SRF1 at 44 days.
Four different kinetic models namely; Korsmeyer-Peppas, Higuchi, Zero and First der kinetic models were applied to analyze the cumulative release data in both water a soil. The kinetic release parameters in water and soil are presented in Tables 2 and 3, resp tively and the plots for all the kinetic models is attached in supplementary document (F ures S1-16). From the tables it can be seen that Korsmeye peppas kinetic equation is t most suitable model to describe the release mechanism of fertilizer in water (R 2 > 0.85) K Figure 6. Water retention behaviour of sandy soil with and without hydrogel.

Release Profile of SRF Hydrogels and Their Kinetics of Release
The cumulative release rate of pure urea and the hydrogel fertilizers; SRF 1 and SRF2 in water and soil is shown in Figures 7 and 8, respectively. The result indicates 99% and 97% release of pure urea used as control in water and soil within one hour and four days, respectively. The urea release in distilled water; being a static medium was faster compared to that in soil. A cumulative release of approximately 37%, 82.2% and 85.4% were observed for SRF1 in water at 1, 6 and 24 h, while 28.6%, 59.6% and 64.4% were observed for SRF2 at the same time in water. The result suggests that the blended paper and HPMC used as second layer coating may be responsible for slower release observed in SRF2 compared to SRF1. The blended paper in the matrix might have created a physical barrier in the form of membrane resistance of the matrix, thereby withstanding osmotic pressure, and causing a slower release by diffusion as a result of the concentration or pressure gradient or both [37,48,49]. The cumulative urea release in soil is shown in Figure 8, the release is significantly lower than that in water. The sustained release rate observed in soil is expected, unlike in distilled water which is static the soil medium is dynamic. For SRF1 cumulative release rates of 27.1%, 64.5% and 67.1% were observed at 6, 30 and 40 days respectively. The cumulative release of urea in SRF1 in soil reaches 67.1% at 44 days. Comparatively, SRF2; the hydrogel with blended paper and HPMC as second layer coating had a higher cumulative percentage release of 30.8%, 82.3% and 87.0% at 6, 30 and 44 days, respectively. The blended paper in SRF2 might have aid absorption of water thereby dissolving the urea in the matrix and consequently resulted in maximum release of 87% of the encapsulated urea at 44 days compared to 67.1% of urea in SRF1 at 44 days. However, the release mechanism in soil is best describe by first order kinetic model (R 0.73), suggesting that the release rate of urea in soil is dependent on its concentration [4 Reports from previous studies [50][51][52] revealed that release from any loaded material fro polymeric excipient could be physically due to diffusion, polymer degradation and dis lution or polymer swelling.  However, the release mechanism in soil is best describe by first order kinetic model (R 2 0.73), suggesting that the release rate of urea in soil is dependent on its concentration [4 Reports from previous studies [50][51][52] revealed that release from any loaded material fro polymeric excipient could be physically due to diffusion, polymer degradation and diss lution or polymer swelling.  Four different kinetic models namely; Korsmeyer-Peppas, Higuchi, Zero and First order kinetic models were applied to analyze the cumulative release data in both water and soil. The kinetic release parameters in water and soil are presented in Tables 2 and 3, respectively and the plots for all the kinetic models is attached in supplementary document ( Figures S1-S16). From the tables it can be seen that Korsmeye peppas kinetic equation is the most suitable model to describe the release mechanism of fertilizer in water (R 2 > 0.85) K × 10 2 is 0.5 and n > 0. 45. This suggest that urea release was by diffusion via the pores and channels of the SRF, which can be controlled by changing the swelling of the SRF [45]. However, the release mechanism in soil is best describe by first order kinetic model (R 2 > 0.73), suggesting that the release rate of urea in soil is dependent on its concentration [45]. Reports from previous studies [50][51][52] revealed that release from any loaded material from polymeric excipient could be physically due to diffusion, polymer degradation and dissolution or polymer swelling.

Conclusions
In this work, slow release and water retention fertilizer hydrogels from hydroxyl propyl methyl cellulose (HPMC)/polyvinyl alcohol (PVA)/glycerol were prepared with blended paper as a second layer matrix which encapsulated urea fertilizer. The hydrogels were analyzed by SEM and XRD. The water absorption in different media and water retention in soil were calculated. The SEM and XRD characterizations confirmed that urea had been successfully absorbed or encapsulated within layer of the matrix, and paper had been introduced into the fertilizer. The swelling behavior of SRF was studied in distilled and tap water. The SRF2 hydrogel had water absorbency of 15.6 (g/g) and 15.2 (g/g) in distilled and tap water, respectively. The water retention was found to increase in soil with SRF hydrogels. Similarly, the slow-release behavior of urea was studied in distilled water and soil. As expected, the swelling ability in distilled water was higher compared to tap water. The swelling kinetics of the hydrogel followed Schott's second order equation. The cumulative release fraction of urea was slower in soil compared to that in water. The release kinetic of urea follows Korsmeye peppas kinetic equation in water (R 2 > 0.85) and first order kinetic model (R 2 >0.56) in soil. These results suggest that the hydrogel SRF from HPMC/PVA/Gly with paper as second layer will go a long way to improve agricultural and horticultural productivity through efficient fertilizer utilization and improved water conservation thereby reducing the amount of urea fertilizer and water for irrigation purposes.

Materials
Hydroxy propyl methyl cellulose (viscosity 25-100 mpa.s) was purchased from Shanghai Chemical Factory, Shanghai, China. Polyvinyl alcohol (degree of hydrolysis, 87-89%, Viscosity of 4% aqueous solution at 20 • C, 35-45 cP) and glycerol were purchased from Guangdang Guanghua Chemical Factory, co ltd, Shanfai, China and Kermel. A4 Paper was obtained from a business center Modibbo Adama University commercial center Yola, Nigeria. Thiosemicarbazide was purchased from Kermel, Tianji, Jinnan District China, deacetyl monoxime was purchased from Blue Horizone, Gopal Niwas, Mumbai, Maharashtra, India, phosphoric acid and sulfuric acid were obtained from commercial outlet; Sigma Aldrich, St. Louis, MO, USA and British Drug Houses, London, UK. All chemicals' reagents were of analytical grade and were used without further purification. Distilled water was obtained from Chemical Engineering laboratory, Modibbo Adama University, Yola, Nigeria and was used throughout the experiment.

Preparation of SRF Hydrogel Based on HPMC, PVA, Glycerol with Paper as Second Layer
Calculated amount of HPMC, PVA and glycerol were mixed at various ratio and 50 mL distilled water was slowly added to the mixture at room temperature under magnetic stirring for 20 min to form dispersion. When completely mixed, the temperature of the mixture was slowly raised to 90 • C maintaining stirring for another 5 min to completely gelatinize the HPMC. Two grams (2 g) of urea fertilizer was thoroughly blended with the mixture with glass rod, there after it was dried at 25 • C forming a single layer matrix. The above procedure was repeated and blended with 0.5 g of paper as the second layer and as well without paper as the second layer. Paper was milled into finer particles by soaking the plain paper in distilled water and was blended using a blender until a homogenous mixture was obtained. The mixture was sieve and air dried. The dried paper was later ground into smaller particles and was used in the matrix formulation. Calculated amount of HPMC, PVA and Glycerol were mixed at ratios given in Table 4, 50 mL distilled water was slowly added to the mixture at room temperature under magnetic stirring for 20 min to form dispersion. The temperature was slowly raised to 90 • C maintaining stirring for 5 min to completely gelatinized the HPMC. Then 0.5 g of the blended paper was added and thoroughly mixed and applied on the first layer forming the second layer and was dried at room temperature before it was later oven dried at 50 • C for 4 h. The matrix was later dried at room temperature and then oven dried at 50 • C for 4 h. Table 4 contains the matrix formulation as reported by Sofyane et al., [30] but with a small modification.

Characterization
The surface morphology of the prepared hydrogels was observed by scanning electron microscope PRO:X: Phenonm World (Model:800-07334), Eindhoven, Netherlands with backscattered electrons detector (BSD), electrons beam voltage of 15 kV. The FTIR spectra of the HPMC, PVA, blended paper and hydrogels matrix were measured by Agilent cary 600 series, Santa Clara, CA, USA at wavelength of 400 to 4000 cm −1 . X-ray diffraction patterns of HPMC, paper and SRF hydrogel were recorded by X-ray diffractometer, thermoscientific (Model: ARL X'TRA), Eindhoven, Waltham, USA in the 2θ range of 10 • to 70 • . Urea content in solution was determined by ultraviolet spectrophotometry (Jenway 6705), Shanghai, China based on standard curve of urea obtained using reaction of urea with diacetylmonoxim colorimetric method [9].

Swelling Behavior of Hydrogels
About 0.5 g of dried samples were weighed, enclosed in 200 mesh bags and immersed in distilled and tap water at 37 • C. After regular time intervals, the swollen samples were removed from the water, dried off of any excess water with a filter paper and weighed. The procedure was replicated three times for SRF1 and SRF2. The water uptake was calculated according to Equation (1).
where, M and M j are weights of swollen and dried samples, respectively.

Swelling Kinetics
The Swelling kinetics of the matrix was determined using Schott's second order swelling kinetic model given as where, S t (g/g) is the water absorption capacity at any time t (h), while S eq (g/g) stands for the theoretical equilibrium water absorption and K is (g/g) represents a constant for the initial swelling rate. From the plot of t/St versus t, a straight line was obtained, which shows that the swelling obeys Schott's second order model and the values of S eq and K is were obtained from the slope and intercept.

Water Retention of Hydrogels in Soil
Dried samples (0.5 g) were weighed and added into 100 g of dried sandy soil in a 250 mL beaker. At the same time, 100 g of dried sandy soil without the hydrogel was put in another beaker as control. Then 50 mL of distilled water was added to each beaker was and weighed. The beakers were weighed every day at ambient temperature (at 37 • C) until it got to a constant mass. The water retention Wr (%) of soil was calculated by Equation (3). Three different sample for SRF1 and SRF2 were used.
where, W r is the water retention rate and W is the total mass of sandy soil and beaker. W o is the total mass of sand, beaker and sample after adding distilled water. W i is the total mass of sand, beaker, and sample at regular intervals.

Slow-Release Behavior of Hydrogels
The release behavior of the matrix was first studied in distilled water, where 1 g of dried sample was immersed in a 250 mL beaker placed in a tight container of 200 mL distilled water maintained at 37 • C. At each preset time interval, the samples were transferred into another beaker containing 200 mL fresh water, this continued until all the urea was released. Then 1 mL of the solution was withdrawn from the beaker and the urea content in the solution was determined by UV-Vis Spectrophotometer using the diacetylmonoxime calorimetry method based on the standard curve obtained. Three samples of SRF1 and SRF2 were used.

Release Kinetic of Hydrogel
The release behavior was analyzed by four different empirical equations that is the power equation by Korsmeyer-Peppas, Higuchi, zero order and first order mathematical models. The power equation is given as Equation (4).
where K i , t, and n are diffusion content, time and diffusion exponent indicative of the release mechanism, respectively. M t /M ∞ is the fraction of fertilizer released at time t. Nutrient release mechanism is classified according to different values of n, if n < 0.45, then it is Fickian diffusion mechanism; if 0.45 < n < 0.89 then is non Fickian diffusion release; if n = 1 then is zero order release, if n > 0.89 then is case II transport [42]. The Higuchi kinetic model can be written as Equation (5).
The Zero order kinetic model is given as follows The First order model is given as below where Q t is the amount of nutrient release at time t, Q o , is the amount of nutrient initially in the matrix, t is the time, K H is the Higuchi dissolution constant, K o and K are the zero order and first order release constant, respectively.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/gels7040262/s1, Figure S1: Graph of Zero order kinetic model of ln Qt vs. ln t for SRF1 in water, Figure S2: Graph of Zero order kinetic model of ln Qt vs. ln t SRF2 in water, Figure S3: Graph of Higuchi model of ln Qt vs. ln t0.5 for SRF1 in water, Figure S4: Graph of Higuchi model of ln Qt vs. ln t0.5 for SRF2 in water, Figure S5: Graph of Korsmeyer-Peppas of ln Mt/ M∞ vs. ln t for SRF1 in water, Figure S6: Graph of Korsmeyer-Peppas of ln Mt/ M∞ vs. ln t for SRF2 in water, Figure S7: Graph of First order kinetic model of ln Qt vs. Time for SRF 1 in water, Figure S8: Graph of First order kinetic model of ln Qt vs. Time for SRF 2 in water, Figure S9: Graph of Zero order kinetic model of ln Qt vs ln t for SRF1 in soil, Figure S10: Graph of Zero order kinetic model of ln Qt vs. ln t for SRF2 in soil, Figure S11: Graph of Higuchi model of ln Qt vs ln t0.5 for SRF1 in soil, Figure S12: Graph of Higuchi model of ln Qt vs. ln t0.5 for SRF2 in soil, Figure S13: Graph of First order kinetic model of ln Qt vs Time for SRF 1 in water, Figure S14: Graph of First order kinetic model of ln Qt vs. Time for SRF 2 in water, Figure S15: Graph of Korsmeyer-Peppas of ln Mt/ M∞ vs. ln t for SRF1 in soil, Figure  S16: Graph of Korsmeyer-Peppas of ln Mt/ M∞ vs. ln t for SRF2 in soil, Figure S17: FTIR spectra of blended paper.