Sugarcane Bagasse Characterization

Sugarcane Bagasse Characterization The major composition of a lignocellulosic biomass is lignin, cellulose and hemicellulose, which is responsible for the structure and rigidity of plant. These components has been reported to have high potential energy and are been widely used as fuel in automobile and industries. The components of the bagasse were chemically characterized by measuring their dry weight. Table # represents the composition of dry sugarcane bagasse analysed in the present study compared with data collected from other research articles. The dissimilarities in composition of lignin and cellulose might be due to genetics variations, growing location, methods of harvesting, growing conditions and analytical procedures. Table 1. Major component of sugarcane bagasse Cellulose (%) Lignin (%) References 46 19.6 Present Study 40.57 25.93 (Zeng, Tong, Wang, Zhu, Ingram, 2014) 25 16.2 (Dhabhai, Jain, Chaurasia, 2012) 40 23 (Irfan, Gulsher, Abbas, Syed, Nadeem, 2011) 45.4 23.4 (Pereira, Jacobus, Cioffi, Mulinari, Luz, 2011) As per the generated data, cellulose content in the bagasse was 46%, which was further reformed into accessible form for the saccharification enzyme. While the lignin constituted 19.6%, thus removal of lignin was carried out by the pre-treatment of bagasse for an efficient enzymatic hydrolysis. Cellulose Unit Activity The AumEnzymes, India generously donated two commercial cellulases, Acid Cellulase and Neutral cellulase. The cellulase activity of Aspergillus terreus, acid cellulase, and neutral cellulase were compared in order to proceed for the optimization of saccharification phase. The International Unit for enzyme activities (IU) of all the three cellulases were based on the total cellulase activity and endoglucanase activity, determined by the CMCase assay and FPU assay respectively. Table# represents the FPU and CMCase activity presented by all the three enzymes. The data in the table# clearly concluded that all the three cellulase have negligible total cellulase activity, while they have a high amount of endoglucanase activity. Table 2. Comparison of Cellulase Activity Cellulase CMCase Activity (IU) FPU Activity (IU) Aspergillus terreus 0.273 0.045 Acid 0.966 0.028 Neutral 0.223 0.000 Which might indicate that all the cellulases has endoglucanase activity but, the negligible exoglucanases activity resulted in considerable reduction in total cellulase activity. Since the Acid cellulase had relatively higher enzyme activity, it was further used as the saccharifying enzyme. The protein content in the Acid cellulose was found using the protein assay and it was found to be 67.67 ÃŽ ¼g/mg of Acid cellulase powder. The specific activity was 14.11 IU/mg of Acid cellulase, indicating that 14.11 ÃŽ ¼mol of sugar is released by 1 mg of Acid cellulase (protein) in one unit. Optimization of alkaline The statistical design used for the microwave assisted alkaline pre-treatment is a four factors (weight of bagasse, power of microwave in wattage, NaOH concentration and the exposure time period) system, the response of the pre-treatment was based on the cellulose composition and reduced lignin after the pre-treatment. The design summary is shown in the Table #. Table 3. Design Summary Study Type: Response Surface Runs: 21 Initial Design: Central Composite Design Model: Quadratic Factor Name Units Low Actual High Actual Low Coded High Coded Mean A Bagasse g% 2.5 10 -1 1 6.25 B Microwave W 100 600 -1 1 350 C NaOH g% 1 5 -1 1 3 D Time minutes 5 10 -1 1 7.5 Response Name Units Analysis Minimum Maximum C.V % R2 Y1 Cellulose g% Polynomial 0 81.2 9.3 0.9679 Y2 Lignin Removal g% Polynomial 0 67.25 8.54 0.9735 The design was a set of 21 runs, combinations of four factor experimental design, based on the RSM and CCD (Tabel#). The RSM is mathematical based system to study the interactions between the factors, while the CCD enables us to deduce an optimal condition for the pre-treatment. Table 4. Test design and results of response surface analysis Factor 1 Factor 2 Factor 3 Factor 4 Response 1 Response 2 Std Run A:Substrate B:Microwave C:NaOH D:Time Cellulose Lignin Removal g W g% minutes g% g% 16 1 6.25 350 3.0 11.7 76.8 48.14 15 2 6.25 350 3.0 3.3 59.2 42.7 6 3 2.50 100 5.0 5.0 55 44.35 21 4 6.25 350 3.0 7.5 72.3 46.7 8 5 2.50 100 1.0 5.0 48.5 35.38 10 6 12.56 350 3.0 7.5 74.6 42.7 13 7 6.25 350 -0.4 7.5 48.25 40 5 8 10.00 100 1.0 10.0 50.6 38.4 9 9 -0.06 350 3.0 7.5 0 0 19 10 6.25 350 3.0 7.5 71.2 46.8 20 11 6.25 350 3.0 7.5 79.5 50.3 2 12 10.00 600 1.0 5.0 56.2 42.7 4 13 2.50 600 1.0 10.0 59.98 48.25 3 14 10.00 100 5.0 10.0 60.6 52.1 11 15 6.25 -70 3.0 7.5 61 48.53 18 16 6.25 350 3.0 7.5 77.1 44.23 7 17 2.50 600 5.0 10.0 75.6 62.5 12 18 6.25 770 3.0 7.5 76.3 67.25 17 19 6.25 350 3.0 7.5 69.7 48.9 1 20 10.00 600 5.0 5.0 71.85 57.23 14 21 6.25 350 6.4 7.5 81.2 60.56 According to the table#, runs #17, #18 and # 21 had maximum lignin removals while the #1, #1 and#21 showed maximum retained cellulose. The quadratic polynomial equations describes the correlation between the significant coefficients i.e. p-value (Prob>F) less than 0.05 and is used to obtain the regression values of coefficients where only significant coefficients are considered. But since this model supports hierarchy, the insignificant coefficients were not omitted. This equation was used to derive the predicted responses for cellulose (equation 1) and lignin removal (equation 2) Equation1 Equation 2 The adequacy of the quadratic model for the experimental responses (cellulose Y1 and lignin removal Y2) was checked using the Analysis of Variance (ANOVA), which was verified using the Fisher’s statistical model (F-value). The table# shows the ANOVA for Y2 response. Table 5. ANOVA result of quadratic regression model for lignin removal Source Sum of Squares Mean squares F-value p-value (Prob > F) Model 3411.23 14 243.66 15.74 0.0014 significant A-Bagasse 911.65 1 911.65 58.88 0.0003 B-Microwave 175.22 1 175.22 11.32 0.0152 C-NaOH 541.91 1 541.91 35.00 0.001 D-Time 14.80 1 14.80 0.96 0.366 AB 3.88 1 3.88 0.25 0.6347 AC 3.14 1 3.14 0.20 0.6684 AD 0.86 1 0.86 0.06 0.8216 BC 4.67 1 4.67 0.30 0.6028 BD 534.56 1 534.56 34.52 0.0011 CD 2.48 1 2.48 0.16 0.7031 A2 955.51 1 955.51 61.71 0.0002 B2 362.14 1 362.14 23.39 0.0029 C2 74.46 1 74.46 4.81 0.0708 D2 3.95 1 3.95 0.25 0.6317 Residual 92.90 6 15.48 Lack of Fit 71.34 2 35.67 6.62 0.0539 not significant Pure Error 21.56 4 5.39 Cor Total 3504.13 20 ANOVA of the regression model for lignin removal had 15.74 â€Å"F-value† which described that the model is significant and also defined that there is only 0.14% chance that a â€Å"Model F-value† this large could arise due to noise. Since the â€Å"p-value† 0.0014, lesser than 0.005, it indicates that the lignin removal is sensitive to the coefficients/factors in the model. In other words weight of bagasse (A), microwave exposure (B), NaOH (C), BD, A2 and B2 have strong influence on the lignin removal. The p-value 0.0011 for BD (B-coded for microwave, D-coded for time), indicates the strong mutual interaction between B and D in removal of lignin. The â€Å"Lack of Fit F-value† of 6.62 justifies that there are 5.39% chances that such large values of â€Å"Lack of Fit F-value† might occur due to noise, where lack of fit is an error that would occur when one of the factor is omitted from the process model. Another statistical measurement that is a sig nal to noise is the ‘‘Adequate precision’’. The desirable ratio is greater than 4, as such the Adeq Precision value is 20.22, this model can be used to navigate design space and further optimization. â€Å"Multiple correlation corfficient or R2† value denotes the correlation between observed and predicted values, i.e. if the value is closer to 1, it means better correlation. In this case the R2 value is 0.9735, indicating better agreement between experimental values and predicted values. The â€Å"coefficient of variation (CV)† indicates the degree of precision to which the experiments are compared. The lower reliability of the experiment is usually indicated by a high value of CV. In the present case the CV value is low (8.5%) indicates a good precision and reliability of the experiment. At the same time, Adjusted determination coefficient (Adj R2)† was high specifies improved precision and reliability of the conducted experiments. The 3D surface plot illustrated below (Figure#) shows co-operative effect of microwaves and NaOH on the removal of lignin. From the plot, it can be predicted that with rise the concentration of NaOH and high powered microwaves exposure a increased degradation of lignin was observed, maximum lignin removal is observed with 5% NaOH concentration and microwave irradiation with power of 600W. But the low power microwaves and NaOH concentrations had no substantial removal of lignin. Figure 1. Co-operative effect of Microwaves and NaOH on lignin removal The second response considered in the pre-treatment was the amount of cellulose retained (Y1) after the process. The ANOVA of quadratic regression model for cellulose retained after pre-treatment illustrated in Table # is a significant model as evident from the Fishers F-test value (12.91) with a very low probability value [(Prob > F) = 0.0165]. This also indicates that there is only 0.24% chance that the F-value occurs due to errors during the experiments. Among model terms A, C, BD and A2 are also significant with probability of 99%. The interaction between B and D significant effect on increase in cellulose retaining response. The goodness of fit of the model was checked by determination coefficient (R2). In this case, the value of the R2 (0.9676) indicates that only 3.24% of the total variation between experimental values and predicted values are not explained by the model. The value of the adjusted determination coefficient (Adj. R2=0.8929) was also high, at the same time a rela tively lower value of the coefficient of variation (C.V. = 9.3%) which indicates model is significant and the conducted experiment is consistent and has a good precision. The level of noise that affected the model is also very low, i.e. 11.16% determined using the Lack of Fit F-value (3.99). The Adequate Precision (15.608) for this model is greater than 4, this suggests the model can be used for navigating the design space and optimizing the experiment. Table 6. ANOVA result of quadratic regression model for cellulose concentration after pre-treatment Source Sum of Squares df Mean Squares F-value p-value (Prob > F) Model 6226.99 14 444.79 12.91 0.0024 significant A-Bagasse 2782.58 1 2782.58 80.76 0.0001 B-Microwave 117.05 1 117.05 3.40 0.1149 C-NaOH 779.62 1 779.62 22.63 0.0031 D-Time 154.88 1 154.88 4.49 0.0783 AB 36.72 1 36.72 1.07 0.3417 AC 1.56 1 1.56 0.05 0.8387 AD 8.14 1 8.14 0.24 0.6441 BC 27.27 1 27.27 0.79 0.4079 BD 1626.88 1 1626.88 47.21 0.0005 CD 1.51 1 1.51 0.04 0.8414 A^2 2013.06 1 2013.06 58.42 0.0003 B^2 4.08 1 4.08 0.12 0.7426 C^2 54.52 1 54.52 1.58 0.2552 D^2 8.46 1 8.46 0.25 0.6379 Residual 206.74 6 34.46 Lack of Fit 137.67 2 68.83 3.99 0.1116 not significant Pure Error 69.07 4 17.27 Cor Total 6433.73 20 Figure # is a 3D response surface plot generated for 6.25 g of bagasse and 7.5 minutes of treatment by the regression mode, illustrates the effect of microwave irradiation (B) and NaOH (C) variables and the interactive effects of each on the cellulose concentration. It can be observed that by increasing both factors B and C results in increased cellulose concentration. The shading on the graph indicates the NaOH concentration from 3% to 5% is adequate for increasing the cellulose concentration to 75% and above along with the microwave irradiation within range of 350 W to 600W. Which indicates that higher microwave irradiation favours lignin removal. This results in high power consumptions and charring of cellulose. To avoid the destruction of cellulose to an inaccessible substance, the treatment can be carried at lower power microwave irradiations under high pressures. The two response models of microwave assisted alkaline pre-treatment have shown positive influence on the removal of lignin and increased cellulose in bagasse. Thus the statistical analysis is reliable to generate the optimal conditions required for pre-treatment, the optimum condition was predicted using numerical optimization. The optimal values selected were, 6.37 g of bagasse irradiated at 350 W in 5% NaOH solution for 8.87 minutes. The predicted cellulose concentration was 81.94% and 56.6% lignin removal. The figure # represents the graph obtained using the numerical optimization methods. Figure 2. Co-operative effect of Microwave and NaOH on cellulose concentration Figure 3. Counter plot for predicted values of Lignin removal and cellulose concentration at optimized condition There was 48% loss in dry weight of the bagasse after pre-treatment at optimized conditions, which might be either due to removal of lignin or lost during the washing process after pre-treatment bagasse. The result was similar to the work done by (Farid, Noor El-Deen, Shata, 2014). Optimization of Saccharification The pre-treated bagasse was washed and further used for saccharification using the Acid cellulase. The efficiency of saccharification is evaluated by the saccharification%, it is the ratio of sugar released and the amount of polysaccharide present in the bagasse. Thus the saccharification% was used as the response factor for the statistical design used to optimize saccharification. The saccharification% response was assessed as a function of pre-treated bagasse loading (A), Acid cellulase loading (B) and time of incubation (C). The design developed using RSM and CCD is summarized in the Table # below. Table 7. Design Summary Study Type: Response Surface Runs: 20 Initial Design: Central Composite Design Model: Quadratic Factor Name Units Low Actual High Actual Low Coded High