Polyvinyl alcohol-carboxymethyl cellulose blended hydrogel membranes were prepared using citric acid as chemical crosslinker through esterification reaction as displayed in Scheme 1., where the feature peaks verifying the chemical crosslinking is proven by FTIR analysis. While physical interaction between PVA and CMC as is explained in Scheme1 is also verified by FTIR analysis.
FTIR spectra of pure PVA, pure CMC, PVA-CMC non-crosslinked membrane and citric acid crosslinked hydrogel membrane are represented in Fig. 1a. The peaks associated with PVA shows C-H alkyl-asymmetric stretching vibration at ν 2850 cm−1. A notable stretching band for –CH2 appears at ν 1560–1450 cm−1 was found. Also, C–O stretching band appeared approximately at ν 1100 cm−1 .
For CMC, aliphatic C–H band appears approximately at ν 2900 cm−1, a new peak represents asymmetric stretching vibration of carbonyl group of COO– at ν 1600-1640 cm−1 and symmetric stretches at approximately 1400 cm−1, also a band at 1070 cm−1 is assigned to C–O–C stretching, were clearly detected .
FT-IR spectrum of PVA/CMC membranes crosslinked by 10 (wt%) of CA shows abroad band appeared at ν 3200–3600 cm−1, which represents O–H stretching vibration  also –OH peak intensity decreased with the crosslinking reaction. This is due to the esterification reaction with citric acid and formation of ester bonds. In addition, new peaks are observed at ν 1750 and 1300 cm−1, which represent the formation of C=O and C–O stretching of ester bonds, respectively; because of the crosslinking reaction . However, IR spectra were difficult to be evaluated due of the peaks overlapping occurred with carboxylic group bands of CMC, and addition of citric acid protonation reaction of carboxymethyl groups of CMC . Also, the spectrum of pristine attapulgite clay and interaction between attapulgite clay and PVA polymeric matrix is relatively clear and was lately explained and discussed well in our previous published work .
Figure 1b represents the effect of citric acid concentration, when CA concentration increases from 7, 10, to 12 (wt%) in membrane composition, the intensity of carbonyl band increases. This is owing to a result of increase the number of ester crosslinks and also free –COOH groups are associated to the increase of CA concentration, where it can’t be significantly determined . Interestingly, esterification mechanism leads to the formation of crosslinks between PVA and CMC, whereas a cyclic anhydride intermediate is formed by post heat-treatment to complete the crosslinking cycle. This intermediate is responsible for the formation of crosslinks with CMC by esterification of –OH group of CMC resulting in another carboxylic acid occurring intramolecular cyclic anhydride with neighboring carboxylate group .
The surface morphology of crosslinked PVA-CMC composite hydrogel membranes with different CA concentrations (7, 10 and 12 wt%) were investigated using SEM, as represented in Fig. 2. SEM micrographs showed that, citric acid content is clearly responsible for changing the surface morphology of tested membrane, since the amount of PVA and CMC were uniform in all films. At membrane of (7 wt%) CA concentration, few pores with smooth, regular and homogenous surface is observed due to low crosslinking degree in the obtained composite membranes. By increasing the CA concentration up to (10 wt%), the film exhibits more porous network that enables the diffusion of water through the pores. These results might be attributed to the presence of three (-COOH) groups for each CA molecule. Thus, it was found with increasing CA percentage, increases the number of –COOH groups available for crosslinking with –OH group of PVA leading to increasing of the crosslinking density and hence, promoting the porosity. This process decreases the water absorption of PVA composite hydrogels. Further, increase of CA ratio up to (12 wt%), hyper-crosslinking was triggered due to further increasing available free –COOH groups of CA for crosslinking process, causing much pores forming and limited water uptake of hydrogel membranes occurred. These results are supporting swelling findings which were discussed below and consistent with previous findings of Nataraj et al. .
Figure 3 displays the surface morphology alternations due to incorporation of attapulgite clay in different contents (1, 2, 4 and 5 wt%) into PVA-CMC composite hydrogel membranes. As clearly seen, the low attapulgite contents (i.e. 1 and 2 wt%) showed compacted, smooth and uniform shape structure. However, high incorporated clay contents (4 and 5 wt%) showed clay aggregation onto the membrane surface. The high clay incorporation behavior is similar and consistent with previous results obtained by Hussein et al. , regarding the aggregated high concentrations of CNCs onto PVA/HA hydrogel membranes.
To assess the mechanical properties of crosslinked PVA/CMC membranes with different CA concentrations (7, 10 and 12 wt%), the maximum tensile strength and elongation to break were tested and obtained data was listed in Table 1. Unexpectedly, maximum tensile strength and elongation to break decreases with increasing the citric acid concentration and they show the same mechanical pattern behavior. These results refer to addition of citric acid to PVA-CMC hydrogel membranes might destabilize and accelerate the elongation-to-break of membranes resulting insignificant decreasing in the maximum tensile strength. These results are almost matched with results obtained by Jiang et al. , who demonstrated that increasing CA concentration up to 9 wt% boosted the CA molecules available for crosslinking leading to an increase in mechanical strength. However, further increase of CA concentration triggered the hyper-crosslinking reaction resulting in sudden mechanical deterioration of CA crosslinked zein NFs .
Different attapulgite clay contents (1, 2, 4, and 5 wt%) were incorporated into PVA-CMC composite hydrogel membranes and their mechanical stability were tested and displayed in Table 2. It was found that, addition of clay up to 1(wt%) significantly increased the entire parameters of mechanical properties of PVA/CMC/clay membrane (Table 2). It was observed that, further increase of attapulgite clay ≥ 1.0 (wt%) might result in an agglomeration of clay occurring the weak interaction between PVA-CMC chains, followed by reducing the crosslinking degree of PVA/CMC due to spacing of polymeric chains accompanied with a sudden mechanical deterioration of hybrid hydrogel membranes [11, 29]. Hence, 1 wt% clay was considered as optimal concentration in terms of cost and high mechanical stability of composite membranes. Similar results were previously reported by Du et al.. They have tested mechanical properties of high Laponite clay contents in Laponite-poly (acrylic acid) nanocomposite membranes.
Thermal Properties (TGA Measurements)
Thermal stability of PVA/CMC hydrogel membranes crosslinked with different CA concentration (7, 10, and 12 wt%) were conducted by TGA measurement (Fig. 4). It was shown that, (9.5, 8.1 and 2.4% of weight loss (%) of PVA/CMC membranes crosslinked with CA (7, 10 and 12 wt%), respectively occurred up to 66–107 oC, owing to evaporation of free water, bounded water and stored humidity. The highest crosslinked membranes with 12 (wt%) of CA was stable until 180 °C, then a steep and sharp weight loss (%) occurred until 459 °C, as known Tonset or the second decomposition stage. This biggest weight loss (%) around 79% was allied with organic matters decomposition and destruction of PVA/CMC chain linkage. Meanwhile, in this second thermal decomposition; some volatile compounds e.g. CO2 were thermally-decomposed. Finally, the pyrolysis or the third complete volatilization stage is detected from 459 to 550 °C with total weight loss (%) of 82% was detected (Fig. 4). Overall, increasing of used CA for crosslinking PVA/CMC membranes sharply improved crosslinking degree, accompanied with enhancing the thermal stability of membranes.
As shown in Fig. 5, the weight loss of PVA/CMC/clay composite membranes sharply reduced from 20 to 2% with increasing the incorporated attapulgite clay contents from 1 to 5 (wt%) owing to evaporation of free water, bounded water and stored humidity. While Tonset of the beginning of second decomposition stage increased from 166 to 193oC, when the incorporated clay content was increased from 1 to 5 (wt%), respectively. Notably, the total weight loss % after pyrolysis thermal decomposition stage (i.e. third decomposition stage) decreased from 92 to 81% with increasing the incorporated clay contents in membranes from 1 to 5 (wt%). These findings refer to the addition of attapulgite clay into PVA-CMC membranes remarkably the entire thermal stability behavior of composite membrane, in addition prolonged the time of decomposition and the total weight loss decreased due to existing of inorganic matters. These results are consistent with obtained results of Du et al. , where who demonstrated that thermal stability of poly (acrylic acid) nanocomposite membranes were sharply enhanced with increasing Laponite clay contents in membranes.
Swelling Ratio % of Hydrogel Membranes
It was found, citric acid concentration affects significantly on the swelling ratio or water uptake of prepared PVA/CMC composite hydrogel membranes, as shown in Fig. 6a. It was proven previously that, swellability of hydrogel membranes increases with increasing the concentration of citric acid, due to increase the hydrophilicity which associated with increasing the carboxyl content in membranes . Notably, the swelling ration of membranes decreased with increase the used concentration of CA, due to increase the crosslinking degree of membranes. These findings could be explained that fact that, increasing clay content might fill the hydrogel pores and forming compacted and condensed interior structure which hinder the water exchange and adsorption, resulting a significant reducing of hybrid hydrogel swellability. On the other side, increase the CA caused further crosslinking for both PVA and CMC until certain concentration (10 wt% CA). However, membranes crosslinked with ≥ 12 wt% of CA show an opposite swelling effect, where the swelling increased significantly owing to imaginably termination reaction occurred. Thus, 10 wt% of CA was chosen as an optimum concentration for further experiments. Our resuts also matched with swelling results obtained by Demitri et al. , who reported that the optimal degree of swelling for practical applications was achieved using low CA concentrations.
As presented in Fig. 6b, it was observed that with increasing attapulgite clay contents, the swelling ability of hydrogel membranes decreases. This might be ascribed to the blocking available pores of membranes and formation of a very compacted interior-structure of hydrogel network causing swelling reduction. In the same context, Mahdavinia et al.  reported that the incorporation of Laponite clay reduced dramatically the swelling ability. Because of, addition of clay acts as additional physical crosslinker preventing from water absorption. Despite decreasing water uptake, but the results are still with acceptable for water uptake capacity to absorb wound exudates and keeping high water resistance, in case employing the prepared membranes as topical wound dressings.
Protein Adsorption (%)
The assessing of adsorbed BSA onto the surface of composite hydrogel membranes is a real significant factor for the ability of physiological and cells attachment manner, which also verify the biocompatibility of tested biomaterials. The influence of both CA crosslinker concentrations and contents of incorporated clay on the amounts of adsorbed BSA onto the surface of membranes after incubation for 48 h, were assessed and shown in Fig. 7. As seen, the adsorbed BSA onto the membrane surfaces generally increased significantly with the low CA concentration and low contents of incorporated attapulgite clay (7 wt% CA and 1 wt% clay, respectively). It is due to the fact that, the adsorbed BSA onto membrane surface increases significantly with the highest swollen composite hydrogel membranes, or with other mean the lowest crosslinked membranes. Notable, the presented findings here are fully consistent with the presented results of swelling measurements in Fig. 6. Also, the speculation of reduction of adsorbed protein was found with the highest crosslinked membranes and lowest values of swollen membranes, which is fully consistent with our previous and reported findings [15, 18, 19, 31].
The hydrolytic degradation or weight loss (%) of crosslinked PVA/CMC hydrogel membranes were assessed as function of different CA concentrations (7, 10, and 12 wt%) and different attapulgite clay contents (1, 2, 4 and 5 wt%), is shown in Fig. 8. It was obviously shown that, both the high CA concentration and attapulgite clay content, might hinder and resist the degradation behavior of hydrogel membranes, compared to the low CA concentration and clay content, respectively. Notably, CA concentration ≥ 10 (wt%) and incorporated clay ≥ 4 (wt%) have progressively kept the mechanical stability of crosslinked membranes, where membranes lost almost 40% and 25–30% of their weights after 15 days, respectively. This implies that, both using CA as crosslinker and incorporation of attapulgite clay have improved dramatically mechanical stability and hindered the hydrolytic degradation behavior of crosslinked PVA/CMC/attapulgite composite membranes. The current degradation results are entirely consistent with our previous reported results [15, 18, 19].
The antimicrobial activity of crosslinked PVA/CMC and PVA/CMC/clay hydrogel membranes was investigated using disc diffusion assay against different human pathogens, as function of different CA and clay contents; was presented in Fig. 7; Table 3.
Crosslinked PVA/CMC hydrogel membranes show remarked resistances against C. albicans with all used CA concentrations (Fig. 9; Table 3). It was observed that, increasing the concentration of citric acid was matched with the increase of the measured inhibition zones (Tale 3). When the CA concentration was raised from 7 to 12 (wt%), shows an increasing in the inhibition zones from 18 to 25 mm. On the other hand, prepared membranes show the lowest resistance against B. cereus microbe (Table 3). However, increasing CA concentrations from 7 to 12 (wt%) results in decreasing in the measured inhibition zones from 17 to 11 mm, respectively. While prepared hydrogel membranes exhibit a moderately resistance toward E. coli and Klebsiella pneumonia. A maximum inhibition zone was recorded of 22 mm with 12 (wt%) CA against K. pneumoniae and 12 mm as the lowest measured inhibition zone with 10 (wt%) CA was applied against E. coli. These results are consistent with obtained results by Mahdavinia et al.  and Siregar et al., who reported that CA crosslinked PVA-CMC films have powerful antimicrobial properties. These results could be attributed to citric acid possesses non dissociated form (–COOH) that can permeate the bacterial cell membrane, and allowing to donate hydrogen ions to the system. To maintain the intracellular pH, H+ ions are released resulting in weak pH, these acidic conditions lead to deformation and damage to cells also damage enzymatic activity, protein structure and DNA of the microorganism .
Also, crosslinked PVA/CMC/clay composite hydrogel membranes show remarked resistances against all tested microbes (Table 3). C. albicans was the most affected microbe by hydrogel membranes, while B. cereus was the lowest affected one. It was observed that, the membranes that lack clay was generally most effective against the tested microbes than clay-containing membranes. The clay-lacking membrane succeeded to affect the pathogens C. albicans, K. pneumonia and B. cereus and show inhibition zones measuring of 25, 24, and 17 mm, respectively. However, the pathogen of E. coli is highly affected by 5% incorporated clay that resulted in 20 mm inhibition zone, compared to 15 mm inhibition zone of clay-lacking membrane was applied. Thus, it could be concluded that each prepared membrane resulted in the formation of clear zones that are considered pathogen dependent. These results are consistent with results obtained by El-Bassyoni et al., who demonstrated that PVA-HES-attapulgite composite hydrogel membranes tested for wound dressing showed significant inhibition zones with antimicrobial potency. It was varied according to the tested microbe and these results might be attributed to the presence of some metals in attapulgite clay composition which have antimicrobial character.
The effect of citric acid concentration on the biocompatibility of prepared membranes was investigated according to the hemolysis percentage of a tested healthy blood sample. As shown in Table 4 (up), the percentage of citric acid is significantly affecting the blood hemolysis percentage. It was found that the lowest used concentration of citric acid (7 wt%) results in the highest hemolysis of tested membrane at (66%). Increasing citric acid concentration to 10 (wt%) shows a slightly reduction of hemolysis at (64%). However, the lowest hemolysis percentage is recorded as 35.5%, when the citric acid concentration is elevated to 12 (wt%). It could be concluded that the descending order of hemolysis would be summarized as: CA concentrations of 7 > 10 > 12 (wt%). Our results are matched with the biocompatible nature of citric acid as previously reported by Salihu et al. . Also, Mali et al. demonstrated that, hemolysis study of CMC composite hydrogel films crosslinked with citric acid indicated their hemocompatibility making them effective for drug delivery application. These results are attributed to the presence of three carboxylic groups in CA structure, so crosslinking of polymers with CA provides some pendant free carboxylic groups which is responsible for enhancing biocompatibility .
Different clay contents were tested for their biocompatibility depending on their behavior towards a healthy blood sample. As depicted in Table 4 (down), the percentage of blood hemolysis was varied according to the tested clay concentration. It was shown that, the lowest clay content in membranes of (1 wt%) results in a hemolytic percentage at 74%. While, increasing clay concentration to 4 and 5 (wt%) results in a gradual increase of hemolysis percentages of 75.7 and 81%, respectively. However, 2 (wt%) of clay provides the lowest blood hemolysis of 52.3%. According to these results, the ascending order of the clay concentration that resulted in high percentages of blood hemolysis could be summarized as: clay contents in membranes (2 ˂ 1˂ 4 ˂ 5 (wt%)). Our results are consistent with results of Golafshan et al. , who demonstrated that the hemolysis ratio of PVA-alginate-Laponite clay hydrogel was promoted by increasing Laponite clay content for wound healing application. Also, it was reported that the clay incorporation into PVA hydrogels improves the hemolysis percentage and coagulation activity of blood making the hydrogel potential candidate for wound healing application [38, 39]. Moreover, gelatin-PVA-hydroxyapatite composite hydrogel membranes have high hemocompatibility making them suitable for biomedical applications .