Fabrication of biodegradable chicken feathers into ecofriendly-functionalized biomaterials: characterization and bio-assessment study

This study aims to investigate novel applications for chicken feather waste hydrolysate through a green, sustainable process. Accordingly, an enzymatically degraded chicken feather (EDCFs) product was used as a dual carbon and nitrogen source in the production medium of bacterial cellulose (BC). The yield maximization was attained through applying experimental designs where the optimal level of each significant variable was recorded and the yield rose 2 times. The produced BC was successfully characterized by FT-IR, XRD and SEM. On the other hand, sludge from EDCFs was used as a paper coating agent. The mechanical features of the coated papers were evaluated by bulk densities, maximum load, breaking length, tensile index, Young’s modulus, work to break and coating layer. The results showed a decrease in tensile index and an increase in elongation at break. These indicate more flexibility of the coated paper. The coated paper exhibits higher resistance to water vapor permeability and remarkable oil resistance compared to the uncoated one. Furthermore, the effectiveness of sludge residue in removing heavy metals was evaluated, and the sorption capacities were ordered as Cu ++  > Fe ++  > Cr ++  > Co ++ with high affinity (3.29 mg/g) toward Cu ++ and low (0.42 mg/g) towards Co ++ in the tested metal solution.

www.nature.com/scientificreports/ had shown a favorable positive effect on biocellulose biosynthesis, however, peptone, Na 2 HPO 4 , temp and EDCFS had a negative impact. The nine variables were investigated using a linear multiple regression analysis approach and the percent confidence level was derived using the formula: confidence level (percent) = (1− p value)* 100. In addition, the main effect was estimated as the difference between the average measurements of each variable taken at a high (+ 1) and low (1) level. In this experiment, the value of the determination coefficient R 2 was 0.87 for BC production, indicating a high degree of correlation between the experimental and predicted values. The polynomial model describes the relationship between the 9 factors and the biocellulose concentration yield as follows: Y cellulose yield = 0.3705 + 0.0115X 1 + 0.01483X 2 −0.0535X 3 −0.0195X 4 + 0.0135X 5 + 0.08283X 6 −0.198X 7 + 0.0138X 8 −0.0345X 9 .  Table 1. PB experimental design for evaluating factors influencing on biocellulose concentration produced by Komagataeibacter hansenii AS.5 using EDCFs as a carbon and nitrogen source. Levels of independent variables (X1-X9) presented between brackets are expressed in terms of g/L for X1-X5, value for X6-X8 and volume (ml/L) for X9. − 1 and + 1 constitute the low level and high level and cellulose conc. expressed as g/L.
Trial X 1 (Glucose) X 2 (YE) X 3 (Peptone) X 4 (Na 2 HPO 4 ) X 5 (Citric acid) X 6 (pH) X 7 (Temp) X 8  www.nature.com/scientificreports/ We employed EDCFs lysate as a medium backbone (carbon and nitrogen source) for BC production since the usage of substituted nutrients greatly lowers the production costs and also diminishes environmental degradation brought on by improperly discarding industrial waste.
The pH, temperature, and cultivation method are physical factors that affect the BC's microbial productivity. Also, the medium's compositions (carbon, nitrogen and additives) are another critical parameters affecting this bioprocess. Two sequential optimization techniques (PB and BB) are considered popular choices for product optimization [41][42][43][44] . Herein, BC production was optimized using the EDCFs as sources of carbon and nitrogen through applying statistical experimental designs. The PB design offers a convenient, quick screening process and quantitatively calculates the significance of a huge number of factors. This saves time and preserves evidence concerning each element 45 . Even if this model does not provide for interaction, the screening programme does not place a high premium on looking into how these numerous elements interact with one another 43 .
Of all studied variables, only the most effective and advantageous ones would be selected for further optimization, while those having a significant detrimental impact on the bioprocess might be eliminated from all future tests. The wide range of PB outcomes in this investigation, from 0.34 to 0.65 g/L of BC, highlights the significance of medium tuning for achieving higher productivity.
Analysis of the regression coefficients, t test, and p values for the nine parameters revealed that glucose, yeast extract, citric acid, pH and time had positive effects on the biosynthesis of BC, but peptone, Na 2 HPO4, temperature, and EDCFs had adverse effects. This is in agreement with Hegde et al. 46 who found the most effective parameters for BC production were glucose, yeast extract, but is in disagreement with Singh et al. 47 who reported that pH had a negative effect on bacterial cellulose production.
Temperature, pH, and EDCFs were chosen because they significantly affect BC synthesis and have a greater confidence level, as shown by a pareto chart (Fig. 2). This chart was illustrated by using EdrowMax software (https:// www. edraw soft. com/ downl oad-edraw max. html). In order to simplify the medium and depend mostly on EDCFs lysate as a medium backbone, two verification tests were applied. The first used the typical medium and conditions recommended by the PB equation model, while the second did not include peptone. Both experiments gave approximately the same yield of BC (1 g/L). Therefore, peptone was omitted from the medium in the subsequent experiment.
Accordingly, the major independent variables (X1; temp, X2; pH, X3; EDCFs waste lysate) were further studied at three levels, using Box-Behnken design (BBD) (the second approach) to find the optimum response region for biocellulose synthesis in terms of g/L ( Table 2). The three variables were studied using a linear multiple regression analysis approach with fourteen trials, and the percentage confidence levels (percent) were calculated as previously described. The determination coefficient R 2 = 0.87 for BC concentration, which is a measure of model fit, indicating that only 13% of the responses could not be fitted within the applied model. The use of surface plots to present experimental data demonstrates that higher levels of cellulose production were achieved with lower EDCFs lysate and higher levels of temperatures and pH, as shown in Fig. 3. A second-order polynomial function was fitted to the experimental results for estimating the optimal point of variable within experimental constraints (non-linear optimization algorithm: The optimal levels of the three examined variables, as determined by the polynomial model's maximum point, were temperature, 23; pH, 8; and EDCFs waste-lysate, 75.26 ml/L, with a predicted cellulose concentration of 1.86 g/L. Finally, to test the quadratic polynomial's accuracy, a verification experiment was conducted under predicted optimal conditions, measuring biocellulose production in the optimized medium. Y cellulose yield = 1.2825 + 0.4065 X 1 + 0.02 X 2 −0.122 X 3 + 0.434 X 1 X 2 −0.032 X 1 X 3 −0.24 X 2 X 3 −0.33(X 1 ) 2 −0.081(X 2 ) 2 −0.272(X 3 ) 2 . www.nature.com/scientificreports/ The percent accuracy of the model was obtained using the following formula to demonstrate its accuracy: model accuracy = [Y Experiment/Y Calculated] 100. The Y value is 2 g/L, according to bench scale studies. The accuracy of the calculated model was 107.5%. In this work, a statistical methodology based on a combination of PB and BB designs was found to be successful and accurate in identifying statistically important components and determining their optimum concentrations. As a result, the following medium composition (g/L) is estimated to be close to the optimum: glucose, 6; yeast extract, 5; Na 2 HPO 4 , 2; citric acid, 1.5; pH, 8; temperature, 23 °C; EDCFs waste, 75.26 ml; and time, 10 days under static condition, where BC concentration was 2 g/L and the yield increased twofold compared with production from the PB design.
Experimental data presented as surface plots reveals that high BC production yield is supported by higher temperatures and pH levels, even at relatively low EDCFs lysate waste levels.
For BC production, the R 2 value was 0.87, indicating a strong association between experimental and forecast values. Experimentally, the confirmed optimal conditions from the optimization experiment were compared to the model's anticipated value. The estimated BC concentration was 2.1 g/L, and the predicted value from the polynomial model was 1.87 g/L. This high level of accuracy (107.5%) shows that the model was validated under ideal circumstances; furthermore, the BC concentration determined by BBD was 2 times higher than PBD. This demonstrates the significance and necessity of the optimization process. Our research supports the findings of 46,48 , they claim that the RSM is a generally accepted modern statistical strategy for the optimization of the experiment's overall circumstances and the solution of the analysis problems. RSM assists in identifying the critical variables for studying interactions, determining the optimal number of variables, and ensuring the best output in a limited number of tests 49 . In conclusion, employing EDCFs for BC manufacture can save production costs and environmental pollution. Statistical analysis. "JMPIN Version: 4.0.4 software was used to perform the experimental designs and statistical analysis and STATISTICA software version 7 was used to draw 3D surface plots".
Chemical structure characterization of BC membranes. Fourier transform infrared (FTIR) spectroscopy-analysis. The chemical structure of the BC membrane based on EDCFs medium was compared with the standard HS medium, where FTIR spectra of both BC membranes are shown in Fig. 4. Basically, no differences in characteristic peaks of BC in either HS medium or based on EDCFs medium were observed 50 . It was observed that the absorption band assigned to the -OH groups of cellulose appears at ν 3350 cm −1 . Other cellulose-specific peaks were found at ν 2894, 1427, 1350, 870 and 655 cm −1 which are assigned to -CH stretching bands, (HCH, OCH) bending inside of plane vibration, -CH deformation vibration, (COC, CCO, CCH) deformation mode stretching vibrations, and C-OH banding out of plane; respectively 50,51 .
X-ray diffraction (XRD) analysis. Figure 5 illustrates the XRD diffraction patterns of both two BC obtained by HS medium and by EDCFs optimized medium. As seen in Fig. 5, no differences in the characteristic peak patterns of two BC membranes are observed. The diffraction diagrams of BC reveal more than two characteristic diffraction peaks, indicating that BC contains I α and I β crystal cellulose. Generally, two BC membranes show a high degree of crystallinity or crystallinity-index between 90-95%, where two BC membranes reveal the characteristic diffraction peaks at 2θ = 16° and 25° with interplaner spacing (d-spacing) 3.91 and 2.32, respectively 50 . Meanwhile, other diffraction peaks are observed at 2θ = 38°, 45°, 64° and 78° which indicate the presence of unreacted or unconverted glucose or amino acids residues of BC obtained from HS medium or produced BC from EDCFs medium, respectively 50,52 . Table 2. BB experimental design for optimizing the most significant variables influencing on biocellulose concentration produced by Komagataeibacter hansenii AS.5 using EDCFs as a carbon and nitrogen source. Trials X 1 (Temp) X 1 (Temp) X 3 (EDCFs) X 1 X 2 X 1 X 3 X 2 X 3 X1 2 X2 2 X3 2 Cellulose conc. (g/L) www.nature.com/scientificreports/ Scanning electron microscopy (SEM) investigation. The morphological properties of BC membranes, which were produced by both HS and EDCFs medium, were examined by SEM with different original magnifications at 10,000 and 20,000X, as shown in Fig. 6. As seen, the surface morphology of BC membrane based on EDCFs seems to be more uniform, with a filamentous shape structure, ribbon-fibril networks, and pours less, particularly with high magnification, compared to traditional bacterial cellulose produced by HS medium (Fig. 6). Notable, fibers of BC produced by EDCFs medium diameters were found to be an average of 50-100 nm,    www.nature.com/scientificreports/ whereas diameters of bacterial cellulose produced by HS medium were found to be between 100-150 nm. The current SEM investigation is almost consistent with SEM investigations as previously reported by Tsouko et al. 53 and Zhang et al. 54 .

Characterization of SEDCFs. FT-IR analysis.
For understanding the chemical structure of pristine CFs and SEDCFs, FT-IR spectra were conducted (Fig. 7). After comparison of control CFs and treated SEDCFs samples, it was observed that the characteristic peaks are mostly similar to each other and are comparable with each other. On the other hand, the chemical structure of SEDCFs exhibits little effect on the chemical structure of protein after degradation (Fig. 7, right). The transmission band region between ν 3500-3200 cm −1 in CFs, was shifted to ν 3500-3000 cm −1 in SEDCFs due to starching vibration of -O-H and -N-H of amide A. However, bands that appeared in the range between ν 3000-2800 cm −1 were related to symmetrical -CH 3 stretching vibration in treated SEDCFs, only 52,53 . Interestingly, in case treated SEDCFs; strong absorbance band at 1730-1630 cm −1 which is attributed to C=O stretching of amide I. Also, the absorption peak at ν 1520-1410 cm −1 is attributed to N-H bending and C-H starching, of amide II of SEDCFs. While, weak band at ν 1240 cm −1 is associated with amide III derived from N-H bending, C-N stretching and some bending from C=O bending and C-C stretching vibration of SEDCFs. Also, a weak peak at ν 700 cm −1 is related to the N-H out-of-plan bending of treated SEDCFs 53 . Notably, strong vibration peak around ν 1730 cm −1 is attributed to C=O of fatty acid ester found in animal skins, and was detected only in cases of treated SEDCFs, which confirms that the treatment of degradation does not affect the main structure of keratin 55 . However, the C-O stretching vibration associated with ester linkage, attributed at around 1230 cm −1 was detected in both CFs control and treated SEDCFs 50,53,55 . Overall, it was shown that there are no effects on the main chemical composition of CFs after degradation treatment, but the chemical composition of SEDCFs becomes more clear and precise.
Wide-angle X-ray diffraction (WAXRD). WAXRD was used to determine the crystal phase of the tested CFs control and SEDCFs samples (Fig. 8). The XRD patterns in Fig. 8 show that both CFs and SEDCFs mainly existed in the semi-crystalline phase, and even after some hydrolysis, they retained the crystallinity. As seen, CFs show diffraction characteristics of α-helix appearing at 2θ = 9.5° and β-sheet at 2θ = 20.8°, however SEDCFs exhibit diffraction characteristics of shifted α-helix appearing at 2θ = 10.6° and β-sheet at 2θ = 21.8°5 0,53,55 . While, the diffraction peaks at 2θ = 13° and 16° of CFs and SEDCFs, respectively are allocated for the amorphous region. Also, diffraction peaks are allocated at 2θ = 29° and 38° were indexed for theβ-sheet crystalline structure of SEDCFs, while the peaks between 2θ = 17°-20° indexed for α-helix diffraction patterns of SEDCFs. Overall, XRD results indicate that partial crystallinity of SEDCFs is retained after the enzymatic degradation process, compared to pristine CFs.
Morphological investigation. The conclusion of untreated and treated CFs morphology was examined via necked eye and using SEM investigation. The outcomes exhibited the presence of all main components of CFs even after degradation treatment with minimum differentiation. However, further morphological investigation and elemental analyses were conducted by SEM and SEM-EDX analyses, as shown in Fig. 9, treated and degraded SEDCFs images were displayed with different magnifications in Fig. 9. As seen, SEDCFs were kept somewhat erection even after degradation treatment as found in pristine CFs, with a lack of woolly-shape structure parts as compared with control CFs (Fig. 9).   www.nature.com/scientificreports/ Mechanical properties for coated HSP. The use of CFs extract as an ecofriendly method for the development of composites and coating agents for the protection of carbon steel from corrosion 10 . This is the first report to use sludge of enzymatically degraded chicken feathers (SEDCFs) as a coating agent for HSP. The HSP was coated to determine its mechanical performance using three different concentrations of SEDCFs (1, 2 and 3%). Table 3 shows the results of mechanical properties, as bulk densities, maximum load, breaking length, tensile index, Young's modulus, work to break, and coating layer. Bulk densities, described as the reciprocal value of density that specifies the compactness or volume of the paper, can be used to conclude the relationship between both the grammage and thickness of paper 56 . The bulk density of blank HSP was more significant than that reported from coated HSP, indicating the opposite the values of the coated HSPs, bulk density shows a reduction upon increasing the content of SEDCFs, thus suggesting that the presence of SEDCFs may affect the cellulosic fibers of HSP, this is compatible with other study 57 . Table 3 shows that with increasing the concentration of SEDCFs, the maximum load and breaking length were increased; this was due to the strong coated HSP formed by the action of coating agent. This result is in agreement with other work 58 . The tensile index is (tensile strength/base weight) a mechanical variable that characterizes tensile strength in relation to material quantity. This variable is influenced by the degree of fiber bond formation 59 . Results represented in Table 3 indicate that the low value of the tensile index was observed in the HSP coated with SEDCFs when compared with blank HSP, so that the tensile index exhibited negligible changes in mechanical properties, this may be due to the large penetration of the coating solutions into the HSP at high coating ratio led to the swelling of the cellulose fibers, which further decreased their mechanical properties 60 .This results are consistent with other research 61 .Young's modulus result of coated HSP refers to the fact that there is no significance was observed due to the action of coated SEDCFs against to blank HSP. The resulting SEDCFs -coated HSP showed increased work at break, this result indicated the flexibility of the coated feather over the HSP, this observation are in agreement with 62 , who coated the HSP with cellulose stearoyl ester, on the other hand, other study reported that with increasing the concentration of coating agent (cellulose nanofiber/chitosan nanoparticles), the value of work at break decreased 63 . In general, a positive effect on the mechanical behavior of coated HSP is induced by the presence of SEDCFs with a relevant increase on both maximum load and work at break.
Water vapor permeability (WVP). The WVP of HSP was defined as the mass of water vapor passing through the HSP per unit area and time under defined conditions. The low WVP values are desirable for industrial packaging like food, drugs, and instruments 62 .WVP of the coated HSP was estimated as well and compared with the uncoated HSP as presented in Fig. 10. It is evident that coated HSP exhibits higher resistance to WVP compared to uncoated HSP. Although SEDCFs have a strong hydrophilic performance, the usage of SEDCFs in paper coating results in lower water vapor penetration through the coated HSP and consequently improves hydrophobicity. In particular, the efficiency of the SEDCFs in reducing the WVP is evident even at low concentration (1%), reaching an improvement in WVP up to 18%. This could be due to the low concentration of feather extract blocking the pour size of HSP.On the other hand, as the concentrations of SEDCFs increased the WVP of coated HSP increased. This explained by the aggregation and coagulation of SEDCFs at high concentration on the surface of HSP. These findings are consistent with previous research, which found that a low concentration of sodium alginate/nanocellulose/Ag-NPs nanocomposite decreased the WVP of coated, whereas a higher concentration of cellulose nanocrystal/Ag-NPs increased the WVP of coated paper 64 .  Table 4. The results showed that HSP coated by SEDCFs had remarkable oil resistance when olive oil was evaluated. Herein, HSP and coated HSP were tested for their oil-resistance using the olive oil assay, where the time required for penetration of olive oil through the sample is observed. The results of the test showed that coated HSP at different concentrations from SEDCFs showed excellent grease proof property since the time needed for penetration of olive oil was > 600 s, which can be classified as high grease-proof materials. As well as, with the SEDCFs concentration increase, the oil resistance of HSP increased. The coated HSP exhibited a good ORB (130-600 s) compared with the uncoated HSP (80 s). Coatings based on SEDCFs are highly lipophilic materials, such that hygroscopic oil does not dissolve in its HSP, leading to excellent ORB 65 . The oil resistance of paper coated with different SEDCFs concentration exhibited good oil resistance as the SEDCFs concentration increased from (1-3%) due to the high hydrophobicity of it, which meets the demands for food packaging. Several studies were reported, the addition of active materials as coated part enhance their ORB as cellulose nanofibers/chitosan nanoparticles 65 , HSP/ZnO/SiO2 58 and HSP/Microcrystalline Wax Emulsion 66 .
Removal of heavy metals using SEDCFs. Heavy metals accumulation represents a great environmental challenge attributed to the adverse effects on human and animal health 33 . Several approaches are currently reported for heavy metal removal. However, the cost implemented, energy consumption, and impact of the environmental impacts of the metal removal method are still challenging 31 . On the other hand, CF is a main waste in the poultry industry with high environmental impacts 31 . In this regard, the efficiency of the SEDCFs on heavy metal removal was evaluated toward four model metal ions, namely, Cu ++ , Fe ++ , Cr ++ and Co ++ . The results (Fig. 11) indicate a significant affinity of SEDCFs toward Cu ++ , Fe ++ and Cr ++ ,which were completely removed from metal solutions (Fig. 11). The SEDCFs affinity ability toward the three metals was in the following order: Cu > Fe > Cr as indicated in the sorption capacity results: 3.29, 2.017 and 1.46 mg/g for the three metals, respectively. The high potency of feather waste for Cr ++ and Cu ++ has been reported in other studies 67 .
In the same regard, the metal removal was reported for CF in Pb ++ removal from wastewater by de la Rosa et al. 68 , and also for CFs-ash for Cd ++ removal from waste water 39 . On the other hand, the SEDCFs revealed a low affinity toward cobalt ions, revealing only 18% Co ++ removal from aqueous solution with a sorption capacity of about 0.42 mg/g. The results are in accordant with Zhang et al. 67 who reported the low affinity of CFs in addition to three other natural keratin sources, toward cobalt ions, and contrary to Chakraborty et al. 35 , reporting the opposite. The low affinity toward cobalt ions, in the current study, asserts the selectivity of biologically prepared  Figure 11. The affinity of SEDCFs toward Cu ++ , Fe ++ , Cr ++ and Co ++ in metal solutions (A) and its sorption capacity q (mg/g) (B). www.nature.com/scientificreports/ SEDCFs toward different metals that could be attributed to the diverse functional groups on the SEDCFs sludge particles' surfaces 2,67 . As per literature, the CF was hypothesized to remediate heavy metals through surface precipitation in addition to ion exchange between the CF surface's Ca ++ ions and the targeted heavy metal 39 .

Conclusion
This study highlights the production of biocellulose by Komagataeibacter hansenii AS by using a new alternative carbon/nitrogen source namely, CFs lysate in submerged fermentation. A complete green sustainable bioprocess was applied successfully and the optimal conditions were attained. The results indicate that the EDCFs waste product macromolecules could be used as a good nutritional ingredient of culture media not only for BC but for any other metabolites; this is implying a potential economic and environmental benefits. Several properties and structural morphology of BC produced from EDCFs waste by-products were compared to that from synthetic medium, where they are compatible. On the other hand the developed sludge upon enzymatic hydrolysis of CFs (SEDCFs) was applied for coating HSP as a green composite. The newly developed bio-composite showed an improvement in paper flexibility, less water permeability and remarkable oil resistance compared to uncoated paper. Additionally, the efficiency of the SEDCFs on heavy metal ions removal was evaluated toward four model metal ions namely, Cu ++ , Fe ++ , Cr ++ and Co ++ . A significant high affinity of SEDCFs was recorded toward Cu ++ , Fe ++ , Cr ++ and low affinity towards Co ++ in metal solution.

Material and methods
Gathering and processing of feathers. CFs was purchased from a nearby market. Feathers were washed with detergent and then washed again with tap water to remove the detergent. Waste feathers were washed and dried at 50 °C for 6 h before being subjected to enzymatic breakdown 67 .
Inoculum preparation. Cells of Laceyella sacchari YNDH from a freshly produced plate (starch/nitrateagar medium) were allowed to grow in a 50 ml aliquot of starch/nitrate broth medium that was dispensed in a 250 ml Erlenmeyer flask as the inoculum. The incubation process was conducted for roughly 48 h at 45 °C and 200 rpm 69 .

Preparation and manufacturing of protease/keratinase. Protease/keratinase enzyme produced by
Laceyella sacchari YNDH was generated in the optimized production medium 67 , after 48 h fermentation period; the broth was collected, centrifuged for 20 min, and the supernatant filtered using 0.2-filters (MDI, India) using a vacuum pump (WATSON-MARLOW-101 U/R).
Enzymatically degradable chicken feathers (EDCFs) and sludge of enzymatically degradable chicken feathers (SEDCFs) preparations.. As previously reported by Doaa et al. 70 , the micro filtered supernatant was employed as the crude enzyme for feather breakdown and feather meal synthesis by adding it directly to the feather wastes. Feather lysate in total included both liquid and sludge has been used in biosynthesis of bacterial cellulose while, for sludge separation, this lysate has been centrifuged for 20 min at 5000 rpm, washed with 70% ethanol followed by distilled water and dried for 8 h at 40 °C for further applications in heavy metal removal and HSP coating.

Biocellulose production. Microorganism. An Egyptian local isolate has been previously identified as
Komagataeibacter hansenii AS 20 , deposited in GenBank at accession number MH109871 and known as a producer of biocellulose was applied in this study. The culture (slant form) was kept at 4 °C using Hestrin and Schramm medium (HS) agar for short term strain preservation. The HS broth medium composed of (g/L): 20 D-glucose, 5 peptone, 5 yeast extract, 2.7 disodium hydrogen phosphate, 1.15 citric acid and ethanol 5 ml (pH 5.5) was used for preinoclume preparation upon incubation the inoculated broth medium under shaking at 30 °C for 2 days for culture activation as reported by Saleh et al. 20 .
Statistical optimization for production of biocellulose using EDCFs. In two steps, physicochemical variables for the synthesis of biocellulose from Komagataeibacter hansenii AS5 using EDCFs as carbon/ nitrogen sources were optimized. The first was the use of PBD to filter physicochemical factors. The second was to use BBD to optimize the most important factors that influence the synthesis of biocellulose processes depending on EDCFs a medium backbone.
PBD. The design was utilized to identify the important factors that had a substantial impact on the amount of biocellulose produced when EDCFs debris was employed among medium components. By using EDCFs waste as a carbon/nitrogn source, a PB experimental design with a set of 12 experiments (trials) was utilized to determine the relative significance of nine factors or variables (glucose, yeast extract, peptone, NaHPO 4 , citric acid, pH, temp. time & EDCFs) that influenced biocellulose formation. PBD is based on the first-order model Y = β 0 + ∑ β i x i , where in this model, the response is represented by Y, the model intercept is represented by β 0 , the variable estimate is represented by βi and the variable is represented by xi. The p value was calculated using standard regression analysis to determine the weight of the studied variables. Table 1 shows the investigated factors, as well as the values of each component in the experimental design and the measured response (cellulose conc. g/L). A high (+ 1) and low (− 1) concentration was evaluated for each component. All trials were done in triplicate (using 50 mL medium in 250 mL Erlenmeyer flasks), and the average value for the measured response was computed. Using BBD, response surface methodology (RSM) was employed to www.nature.com/scientificreports/ optimize the screened components for better biocellulose production. After calculating the relative significance of independent variables, the three major significant factors were chosen for further evaluation in terms of cellulose conc. (g/L) as a response after 10 days of incubation time under static condition.
BBD. BBD was applied for optimizing the production of biocellulose using EDCFs lysate fraction as carbon and nitrogen sources, where the conducting statistically planned trials, estimating the coefficients of the constructed mathematical model, anticipating the response, and judging the model's appropriateness were all part of the optimization approach 71 . Table 2 shows the design matrix (containing 12 trials), three levels (high, medium and low) for the chosen variables (denoted by + 1, 0, and − 1) and 2 central trials to find faults in handling as well as the measured response (biocellulose conc.) 72 . The coefficient results of each variable were used to apply the model 73 . The following second-order polynomial structured model was used to predict biocellulose conc. (Y) as a function of cultivation conditions (X) for three variables: Y is the anticipated response; β 0 is the model intercept; X 1 , X 2 and X 3 are the independent variables; β 1 , β 2 and β 3 are linear coefficients; β 12 , β 13 and β 23 are cross product coefficient s; and β 11 , β 22 and β 33 are the quadratic coefficient.
Quantification of BC concentration/yield gravimetrically. At the end of the cultivation time, the observed BC pellicle at the air-liquid interface was collected and washed several times with distilled water to get rid of the excess medium components. Afterwards, the BC sample was then soacked three times in 0.5% sodium hydroxide at 90 °C for 30 min, to remove bacterial contaminants and other impurities immobilized on the BC, and then washed with distilled water until neutralization. Finally, the purified BC sample was dried in the oven at 50 °C until a constant weight was recorded 74 .
Data analysis using statistical techniques. Multiple linear regressions were used to estimate the t values, p values, and confidence levels for biocellulose production yield using data analysis JMP software. The Student t test was used to evaluate the significance level (p value). The t test for any individual effect allows for an assessment of the likelihood of discovering the observed effect by chance. It will be accepted if the probability of the variable under test is low enough. The confidence level is a percentage representation of the p value. Using the JMP software, the optimal value of the biocellulose conc. yield was calculated. A three-dimensional graph was created using STATISTICA 7.0 software 70,71 , in order to display the simultaneous impact of the three most significant independent factors on each response.
Instrumental characterization of bacterial cellulose. FTIR: The chemical structures of BC membranes were analyzed by FTIR (IR, 8400 s Shimadzu, Japan), with the IR fingerprints recorded between 4000 and 400 cm −1 using transmittance modes. XRD: The overall crystalline phases of BC membranes were determined by XRD measurement on an (X Ray Diffractometer, Malvern Panalytical Empyrean, France). Radial scans of intensity were recorded at ambient conditions over scattering 2 angles from 5° to 80° with a step increment of 0.02°/s. SEM: The surface structure of BC membranes was investigated by scanning electron microscopy (SEM, Joel GSM-6610LV, Japan). The average diameter of bacterial cellulose fibers was measured by software Image-J.
Instrumental characterization of SEDCFs. FTIR: The chemical structure of degraded CFs waste (SED-CFs) was analyzed by FTIR (IR, 8400 s Shimadzu, Japan) with the IR fingerprints recorded between 4000-400 cm −1 using transmittance modes. XRD: The overall crystalline phases of SEDCFs were determined by XRD measurement on a (X Ray Diffractometer, Malvern Panalytical Empyrean, France). Radial scans of intensity were recorded at ambient conditions over scattering two angles from 5° to 80° with a step increment of 0.02°/s. SEM: The surface structure of the purified SEDCFs was investigated by scanning electron microscopy (SEM, Joel GSM-6610LV, Japan).
Preparation of HSP. Sugarcane bagasse was treated by two chemical treatments for pulping to remove the lignin content and then bleaching to obtain the pure cellulose which was used for HSP making according to the S.C.A standard, using a sheet former (S.C.A model-AB Lorentzen and Wettre). The HSP was prepared as reported by Atykyan et al. 56 with minor modifications. Briefly, about 1.8 g of bleached sugarcane bagasse was homogenized with 5-7 L of tap water. After a homogenizing step, the suspension is separated through a screen. A sheet of 214 cm 2 surface area and 165 mm in diameter is formed in the appliance, and then it is pressed for 4 min using a hydraulic press. The wet sheet is then collected on blotting paper, protected between two sheets. Drying of the prepared sheets is finished by using a rotary drum dryer for 2 h at 105 °C.
Coating of HSP. The coating of HSP was carried out according to Vaithanomsat et al. 75  Lloyd Instruments, Fareham, UK) with a 100-N load cell at a constant crosshead speed of 2.5 cm/min in line with TAPPI (T494-06) standard method was used to determine the mechanical properties of coated HSP with different concentrations of SEDCFs. The gauge length is set at 10 cm and strips of 15 mm in width and 15 cm in length are used for the analysis. Each HSP thickness was determined by an electronic digital micrometer before the examination. Each test was performed by using 3 specimens, and the average of the results was recorded.
Physical properties. WVP (g.m −1 .KPa −1 .hr −1 ). The WVP of blank and coated HSP was evaluated according to the standard method ASTM E96-E80 with minor modifications. About 1.5 cm of blank and coated HSP with different concentrations of SEDCFs was sealed on the falcon tube (15 ml) containing 1 g of anhydrous calcium chloride. Plastic adhesive film was used to maintain the sample with the wide rim of the falcon tube to avoid air penetration. They were weighed and then placed in desiccators containing saturated potassium sulphate solution to get 95% relative humidity (RH) at 25 °C throughout the experiment. The weight of the falcon tube covered with film was monitored every day for a period of 10 days. The WVP of paper samples was calculated using the following Eq. (1): (W/t) = the slope of the plot between weight gain and time, x = the average thickness of the coated HSP, A = the permeation area, and ΔP = the partial water vapor pressure difference of the atmosphere in the cup and saturated sodium chloride solution corresponding to 0-95% RH.
Oil resistance barrier (ORB). The oil resistance was measured according to the TAPPI (T-454 om-06) method. In this study, olive oil was used. the tested areas of the uncoated and coated HSP samples were placed under defined conditions of 25 °C and 50% RH; then, samples were placed in contact with a piece of white blank paper and 5 g of sand with a specified particle size (Sieve No. 30) on top of coated paper sheets; about (1.1 ml) of oil with soluble red dye saturated the sand sample. The time required for the oil to penetrate the samples was recorded to the nearest 10 s. The test of oil penetration was calculated as the average value of three measurements, where the resistance % was calculated as (the time required for the oil to penetrate the SEDCFs -coated HSP/the time required for the oil to penetrate the uncoated HSP) *100. , then the SEDCFs residue was separated by centrifugation and the residual metals concentrations were determined in the supernatant using atomic absorption spectrometry (Perkin Elmer A Annalist 100 spectrophotometer equipped with an air-acetylene burner). All the experiments were conducted in triplicate, and the average results are reported and compared to positive controls containing 0.1 g/L final concentration of each metal. The concentrations of the four metals were also evaluated in the SEDCFs fraction suspended in 100 ml of water and cultivated at the same experiment conditions (negative control). The sorption capacity q (mg/g) of the CFs sludge was calculated using the following equation: q = C0−Ct W V where C0 is the initial metal concentration (mg/L) and Ct is the remaining metal concentration (mg/L) at the time t (h), V is the volume of solution (L), and W is the sorbent amount (g).

Data availability
All data produced during this study are included in this published article. www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.