Factors Affecting Enzymatic Hydrolysis from Pretreated Fibre Pressed Oil Palm Frond Using Sacchariseb C 6

In this study, fibre pressed oil palm frond (FPOPF) was introduced as the raw material for the production of glucose using Sacchariseb C6 to maximise the utilisation of oil palm waste. Sacchariseb C6 is a commercial cellulase blended enzyme from Advanced Enzymes Technology. In order to achieve maximum glucose production, a factorial analysis 25–1 using response surface methodology (RSM) was employed to screen the best enzymatic hydrolysis condition by varying the parameters such as agitation speed, enzyme loading, glucan loading, temperature and hydrolysis time. FPOPF was treated with 4.42% (w/v) of sodium hydroxide at 100°C prior to the enzymatic hydrolysis. Raw FPOPF consists of 40.7% glucan, 26.1% xylan, 26.2% lignin, 1.8% ash and 4.5% extractives. On the other hand, pretreated FPOPF consists of 61.4% glucan, 20.4% xylan, 13.3% lignin, 1.3% ash and 0.3% extractives. From this study, it was found that the best enzymatic hydrolysis condition yielded 33.01 ± 0.73 g/L of glucose when performed at 200 rpm of agitation speed, 60 FPU/mL of enzyme loading, 4% (w/w) of glucan loading, temperature at 55°C and 72 h of reaction time. The model obtained from RSM was significant with p-value <0.0001. It is suggested that this model had a maximum point which is likely to be the optimum point and possible for the optimisation process.


INtrODUCtION
Lignocellulosic biomass (LCB) is one of the most abundant renewable biomasses comprising of cellulose, hemicellulose and lignin. 1,2Commonly, most of the agricultural LCB consists of about 10%-25% lignin, 20%-30% hemicellulose, and 40%-50% cellulose. 3However, the distribution of each component varies significantly between different plants. 4These differences may due to the different types of plant, sources, ages and others.Different geographic locations, climate, and soil conditions can also be the reasons for the variations in the chemical composition among them. 5,6 this study, fibre pressed oil palm frond (FPOPF) was introduced as a raw material where the hemicellulose and cellulose were converted into simple sugars.In order to improve the accessibility of cellulase enzyme on FPOPF, the structure of the lignocellulose must be broken down, i.e., by alkaline pretreatment.In alkaline pretreatment, the alteration of the lignin structure in biomass is achieved by degrading ester and glycosidic side chains of the biomass using alkaline solvent, leading to swelling as well as de-crystallisation of cellulose. 7,8Sacchariseb C6 which is a blended cellulase enzyme was used in enzymatic hydrolysis for the production of glucose by breaking down of cellulose molecule into simple sugar.
Therefore, this study aims to present a systematic study on the effect of simultaneous changes of synthesis conditions on glucose production using Sacchariseb C6 and thus, finding the best enzymatic hydrolysis condition using statistical approach of response surface methodology (RSM).Screening design was used to study potentially affective parameters by identify the dominant and significant factors contribute to the enzymatic hydrolysis.

Materials
Oil palm frond (OPF) was collected from a local palm oil plantation at Kuantan, Pahang, Malaysia.The OPF was pressed by using sugarcane machine to separate juices from the fibre.The FPOPF was dried under the sun for 2-3 days until constant weight.Then, FPOPF was shredded into pieces and sieved into particle size less than 2 mm.The dried FPOPF was stored in sealed plastic bag at room temperature.Enzymatic hydrolysis was carried out using blended cellulase enzyme namely Sacchariseb C6, an industrial grade commercial enzyme obtained from Advanced Enzyme Technology (India).

Compositional Analysis of FPOPF
FPOPF was characterised to determine the composition of glucan, xylan, lignin, extractive and ashes contents.0][11][12] The compositional characterisation analysis was performed on both the untreated FPOPF and pretreated FPOPF.
Moisture content analysis was carried out using A&D MS70 moisture analyser (DSC, UK) while ash content analysis was conducted using a furnace at using muffle furnace at 575°C ± 25°C for 24 ± 6 h.The extractives content was measured using DIONEX ASE 350 (Thermo Scientific, USA) with water and ethanol as solvents for 30 min.The recovered water extract was analysed to determine the soluble sugar content in the FPOPF.Meanwhile, ethanol extracts were used to determine the ethanol extractive that includes chlorophyll, proteins fats and oils.
Analysis on acid insoluble lignin and acid soluble lignin were determined using two-step acid hydrolysis.The acid insoluble material was determined using gravimetric analysis while UV-Vis spectroscopy was used to measure acid soluble lignin in FPOPF. 13

Alkaline Pretreatment
Referring to Sukri and Rahman, 7 the FPOPF sample was soaked in sodium hydroxide (NaOH) solution with concentration 4.42% (w/v). 7The sample was treated at 100°C for about 58.31 min.Then, the treated FPOPF was washed thoroughly with de-ionised water until turned to neutral.It was dried in the oven at 105°C and stored for further analysis.

Enzymatic Hydrolysis
Prior to the enzymatic hydrolysis, the moisture content of pretreated FPOPF should be less than 5%.The pretreated FPOPF (1%-4% w/v) was weighed using analytical balance and added into 20 mL scintillation vial containing 0.02% (w/v) sodium azide and 0.05 M citrate buffer at pH 4.8.Sodium azide was added to prevent microbial growth.Then, the mixture was pre-incubated at certain temperature (35°C-55°C) prior to the addition of enzymes.The enzymatic hydrolysis was then initiated by adding Sacchariseb C6 (20-60 FPU/ml) and Novozyme 188 (64 pNPGU/mL).
The incubator shaker started to agitate at ranges 50 to 200 rpm.At the end of the hydrolysis (3-72 h), the samples were filtered for further analysis.Each experiment was carried out in triplicate.

Experimental Design
The experimental design for factorial analysis was performed using Design Expert 7.0.0(Stat-Ease Inc., USA) software.Five independent factors as shown in Table 1 were analysed using RSM.5][16][17] The factors were constructed in half level factorial designs of 2 5-1 to screen their effect on the response of glucose production.The validation run for factorial analysis was carried by comparing the experimental values with the predicted model generated by Design Expert software.The condition for the validation run was obtained from the predicted best condition developed from 2 5-1 factorial design.The percentage of error was calculated using Equation 1.

HPLC Analysis
The hydrolysate was determined using Agilent 1200 high performance liquid chromatography (HPLC) system equipped with refractive index (RI) detector.The column used was RHM Monosaccharide H+ column.Pure water was used as the mobile phase prepared using Milli-Q ultrapure water (Millipore, USA).The column temperature was maintained at 60°C.The flow rate of 0.4 mL/min and 5µL injection volume was used.Hydrolysate samples and standards were filtered using 0.22 µm syringe filter before HPLC analysis.The calibration curve was prepared with the ranges of 1 g/L to 40 g/L.

3.
rESULtS AND DISCUSSION

Composition of FPOPF
The characterisation of FPOPF was carried out for untreated and pretreated FPOPF according to NREL's methods.Untreated FPOPF refers to the biomass without any pretreatment applied on it.Whereas, pretreated FPOPF refers to the biomass that undergoes alkaline pretreatment.The compositional analysis of both FPOPF were analysed in terms of glucan, xylan, lignin, ashes and extractives as shown in Table 2.The total structural carbohydrate content for untreated FPOPF was found to be 66.8% with 40.7% of glucan as the major structural carbohydrate followed 26.1% of xylan.Meanwhile, pretreated FPOPF shown 81.8% of total carbohydrates comprise of 61.4% glucan and 20.4% xylan.An increase of glucan but a reduction of xylan could be seen because of the outer layer was disrupted during alkaline pretreatment caused by partial removal of lignin and hemicellulose resulting exposure of cellulose fibers.Lower yield of lignin was observed after alkaline pretreatment which proved that alkaline pretreatment caused the delignification to occur.

Screening of Enzymatic Hydrolysis
The experimental design of half level factorial analysis was carried out to determine the factors affecting production of glucose during enzymatic hydrolysis.These five factors include agitation speed, enzyme loading, glucan loading, temperature, and hydrolysis time.Table 3 clearly shows that the highest production of glucose was obtained at 33.01 ± 0.73 g/L where the conditions at 200 rpm of agitation speed with temperature of 55°C, 4% of glucan loading, and 60 FPU/mL of enzyme loading for 72 h in hydrolysis time.The relative size of effects is shown in Pareto chart, where the bar length is proportional to the absolute value of estimated effect.Effects of t-value limit (black line) are considered statistically significant at 95% confidence level whereas the effects below t-value limit are not likely to be significant.Effect above Bonferroni's corrected t-value limit (red line) is almost certainly significant.A quick analysis was performed on the selected effects using Pareto chart to statistically check for significance of the selected effects at 95% confidence level.All the selected effects (A, B, C, D, E, AC, CD, and CE) shown to be significant at both t-value limit and Bonferroni's corrected t-value limit.

Analysis of Variance (ANOVA)
The analysis of variance (ANOVA) was carried out to determine the significant effect of the model in this enzymatic hydrolysis process.The model obtained was significant with p-value <0.0001 as shown in Table 4.A good fitting model can be determined by the value of the coefficient of determination (R 2 ) more than 0.80. 18In this study, R 2 value obtained in this model was 0.9959, which is in good agreement with the adjusted R 2 value of 0.9912.The high R 2 value of 0.9959 indicates that the model was well adapted to the response.
where, Y = concentration of glucose (g/L), A = agitation speed (rpm), B = enzyme loading (FPU/ ml), C = glucan loading (% w/v), D = temperature (°C), and E = hydrolysis time (h) The unknowns A, B, C, D, and E were referred to the main effects while AC, CD and CE were the interaction effects contributed in the enzymatic hydrolysis process.Based on the quadratic model, coefficients of A to E are small compared to constant.This gives an indicator that the model equation is good with small error and can be used for further analysis.

Comparison of Actual versus Predicted Graph
A regression model can be used to predict expected new observations on the glucose production corresponding to experimental values of the factors.Meanwhile, the data that extrapolate beyond the straight line generated by Design Expert is highly possible that a model is no longer fit well in the regression model.The experimental data for the production of glucose from the empirical model is in good agreement with the observed ones in the range of the operating factors as shown in Figure 2.

Effect of Interfacial Polymerisation Factors on the Glucose Production
The interaction effect plot was generated to represent the results of the regression analysis.It was represented the deviations of the average between the high and low levels for each factors.

Interaction between time and glucan loading
The interaction between glucan loading and time (CE) gives highest contribution of 7.92% to the enzymatic hydrolysis process, as shown in Figure 3.The amount of glucose after enzymatic hydrolysis was higher at 72 h compared to 3 h.In this study, having longer reaction time with high glucan loading was more beneficial because it has huge positive effect on glucose production and allowed the enzyme to hydrolyse more cellulose in FPOPF will be converted into monomer.Thus, prolong the hydrolysis time will increase the glucose production.Similar trend was obtained by Tan and Lee 16 and Zheng et al. 13 A comparison of results based on two different hydrolysis time reveals glucan and time are interrelated in increasing the glucose production.

Interaction between glucan loading and agitation speed
Guo et al. explained that increase of agitation speed from 0 rpm to 150 rpm will increase the initial hydrolysis rates and final sugar concentration. 15However, Champagne and Li stated that mixing rate above 200 rpm resulted in decreased enzymatic hydrolysis due to the shear-induced deactivation of cellulose occurred. 18igure 4 shows the second interaction between agitation speed and glucan loading (AC) that gives 0.43% of contribution in enzymatic hydrolysis.In this present study, it was found that the glucan loading highly affected the enzymatic hydrolysis from FPOPF.There was a huge difference between low and high glucan loading at different agitation speed in the production of glucose.At lower glucan loading (1% w/v), no significant effect to the production of glucose can be observed either at low or high agitation speed, but not the case at high glucan loading (4% w/v).
An increase of glucose production under agitation condition at high solid loading was due to the enhancement of mass transfer and cellulase diffusion. 20Thus, at high glucan loading (4% w/v) with high mixing rate of 200 rpm can promote more cellulose to be hydrolysed to glucose because of the sufficient mixing between FPOPF and cellulase.

Interaction between temperature and glucan loading
Another interaction can be seen between glucan loading and temperature (CD) as plotted in Figure 5.The temperatures at 35°C and 55°C do not give any obvious effect on the glucose production.However, an increase in temperature of the incubator shaker during enzymatic hydrolysis at high glucan loading (4% w/v) drastically improves the glucose production.This might due to the positive relationship between adsorption and hydrolysis of FPOPF occurred excellently at high temperatures (55°C).The enzyme-substrate interaction performed the best activity at 55°C resulting in increase of glucose production.Overall, high glucan loading (4% w/v) with temperature either at 35°C or 55°C resulted almost same level of hydrolysis efficiency in order to produce high amount of glucose.Similar results reported by Tan and Lee and Nieves et al.where the glucose yield was gradually increased as the substrate loading increased. 16,20igure 5: Effect of glucan loading and temperature to the glucose production.

Validation run of Factorial Analysis
The validation experiments were conducted in triplicate based on suggested best condition by Design Expert 7.0.The experiments were carried out at 160 rpm of agitation speed, 20 FPU/mL of Sacchariseb C6, glucan loading at 4%, temperature at 56°C and hydrolysis time at 72 h.The error from these validations runs was in between 1.85% to 4.70% as presented in Table 5.The model was found to be in good agreement with the experimental values with error less than 10%.

CONCLUSION
FPOPF is a promising feedstock for the production of fermentable sugar due to its high biomass yield and potential fermentable sugar yield from bagasse.Sacchariseb C6 was performed excellently well in enzymatic hydrolysis of pretreated FPOPF at high glucan loading (4% w/v) and moderate temperature (55°C).The results obtained from this study showed the best condition for the enzymatic hydrolysis was at 33.01 ± 0.73 g/L of glucose, 200 rpm of agitation speed, 60 FPU/ mL of enzyme loading, 4% (w/v) of glucan loading, temperature at 55°C and 72 h of hydrolysis time.Based on the quadratic model, coefficients of X 1 to X 5 are small compared to the constant.This gives an indicator that the model equation obtained in this study was good with small error and might had a maximum point which is likely to be the optimum point and possible for the optimisation process later.The model was found to be in good agreement with the experimental values with the error obtained from validations runs was less than 5%.

Figure 3 :
Figure 3: Effect of glucan loading and time to the glucose production.

Figure 4 :
Figure 4: Effect of agitation speed and glucan loading to the glucose production.

Table 1 :
Parameters and their designated low and high value.

Table 2 :
Differences in composition of untreated and pretreated FPOPF.

Table 3 :
Experimental design for factorial analysis with its response.

Table 4 :
ANOVA for factorial analysis.Equation 2 shows the response surface quadratic model for glucose production which can be presented in terms of coded factors as in the following equation:

Table 5 :
Validation run for factorial analysis.