Chemical oxygen demand and ammonia nitrogen mycoremediation from 1 domestic wastewater effluent using G anoderma lucidum in a batch system 2

17 The fungi-based technology, wild-Serbian Ganoderma lucidum (WSGL) as myco-alternative to existing 18 conventional microbial-based wastewater treatment is introduced in this study as a potential alternative 19 treatment. The mycoremediation is highly persistent for its capability to oxidatively breakdown pollutant 20 substrates and widely researched for its medicinal properties. Utilizing the non-hazardous properties and 21 high degradation performance of WSGL, this research aims to find optimum conditions and model the 22 mycoremediation treatment design for Chemical Oxygen Demand (COD) and Ammonia Nitrogen (AN) 23 removal in domestic wastewater via response surface methodology (RSM). Combined process variables were temperature (⁰C) (Model 1) and the volume of mycelial pellets (%) (Model 2) against treatment time 25 (hour). Response variables for these two sets of central composite design (CCD) were the removal 26 efficiencies of COD (%) and AN (%). The regression line fitted well with the data with R 2 values of 0.9840 27 (Model 1-COD), 0.9477 (Model 1-AN), 0.9988 (Model 2-COD) and 0.9990 (Model 2-AN). The lack of fit 28 test gives the highest value of Sum of Squares equal to 9494.91 (Model 1- COD), 9701.68 (Model 1-AN), 29 23786.55 (Model 2-COD) and 13357.02 (Model 2-AN), with probability F values less than 0.05 showing 30 significant models. The optimum conditions were established corresponding to the percentage of COD and 31 AN removal obtained were 95.1% and 96.3%, accordingly at the optimum temperature 25 ° C at the 32 treatment time of 24 h, meanwhile 0.25% of mycelial pellet with 76.0% and 78.4% COD and AN removal, 33 respectively. The high performance achieved demonstrates that the mycoremediation of G. lucidum is 34 highly potential as part of the wastewater treatment system in treating domestic wastewater of high organic 35 loadings.


Introduction 41
The wastewater influent in Malaysia has a low concentration of COD between 117 mg/L to 612 mg/L, 42 lower than the influent designed STP with COD concentration equivalent to 500 mg/L (Hanafiah et al., 43 2019a). Factors influencing the biological processes in wastewater treatment include short Solids Retention The effect of process parameters (i.e. independent variables) on the COD and removal from synthetic 155 domestic wastewater is discussed. The Model 1 and Model 2 for the degradation of COD and AN were 156 optimized based on the RSM. The central composite design (CCD) produced 12 sets of experimental and 157 predicted values of percentage removal for COD and AN for both models, as shown in Table 2. 158 159

161
The following regression quadratic equations (Eq.  As the final output of RSM, the degradation of COD, AN as well as the integrated kinetic rates were 173 conducted based on the RSM optimized treatment design. Figure 4 shows the graph for the coefficient of 174 determination (R 2 ) of the models and model terms as described in Equations (2 -5). This presentation is an 175 efficient approach to assess the variations of each experimental factor. This explains the accuracy of the 176 models that fit according to each factor, where the response is valid for optimization. The results were analyzed using analysis of variance (ANOVA) and the results of the established quadratic 183 models for the percentage of COD and AN removal were shown in Table 3

191
The fit summary determines the nonlinearity on the outliers observed (as shown in Table 4). 192 Overall, the quadratic model is significant for optimization purposes. The R 2 for model 1-COD (0.9840), 193 Furthermore, Figure 5 shows the scatter plot of predicted against actual values of responses, COD and AN. 202 Based on the scatterplots pattern, the regression assumptions of the proposed models were satisfactory, 203 showing a linear relationship and that the constant variance assumption was established. 204  Based on RSM analysis, the optimized condition for the WSGL in treating domestic wastewater for Model 1 was in the condition of temperature 25ºC, meanwhile for Model 2 with the inoculum volume of 0.25%. Therefore, the experimental results for both conditions were then tried to fit into the first order and secondorder kinetic to find which model is the most suitable to present the degradation using WSGL.
The first-order model is described as

Conclusion
The treatment of synthetic domestic wastewater via mycoremediation of WSGL has efficiently degraded the COD and AN. From the batch experiments, WSGL favors acidic conditions to obtain high performance in reducing the concentration of pollutants. RSM-based optimization provides the optimum condition of WSGL wastewater treatment. A maximum reduction of up to 95% and 96% of COD and AN, respectively were obtained at the optimized temperature of 25ºC and 24 h of treatment time. Meanwhile, the 0.25% inoculum volume was sufficient for the WSGL where 76.0% and 78.4% of COD and AN were removed, respectively. Moreover, the experimental data followed the first-order kinetics model, and simultaneously able to determine the rate of removal for both COD and AN. Results here showed that the WSGL is promising and could be utilized as an attractive alternative method, or as a tertiary treatment in a Sewage Treatment Plant.

 Inoculum Preparation
The wild-Serbian Ganoderma lucidum (WSGL) strain BGF4A1 stock culture was obtained from Serbia   Table 1. To determine the relationship between the factors and response variables, the data collected were Then, to visualize the relationship between experimental variables and the responses, the response surface and contour plot was generated from the models and the optimum values of the process variables were extracted from the plot. Through the ANOVA analysis, the adequacy of the models was further justified. The adequacy of a model was then compared to pure errors using lack-of-fit, based on the replicate measurements to the other lack-of-fit and model performance. The calculated ratio of the lack-of-fit mean square and the pure error mean square represents as F value, is to determine whether the lack-of-fit is significant or not, at a significant level.