Optimization of batch conditions for COD and ammonia nitrogen removal using cockle shells through response surface methodology

The optimal conditions for the reduction of COD and NH3-N using cockle shells (CS) from a stabilised landfill effluent were analyzed. The influence of two variables (adsorbent dosage and pH) were analysed through the application of response surface methodology (RSM) and central composite design (CCD...

Full description

Saved in:
Bibliographic Details
Main Authors: Daud, Zawawi, Abubakar, Mahmoud Hijab, Awang, Halizah, Ahmed, Zainab Belel, Rosli, Mohd Arif, Ridzuan, Mohd Baharudin, Aliyu, Ruwaida, Tajarudin, Husnul Azan
Format: Article
Language:English
Published: Penerbit UTHM 2018
Subjects:
Online Access:http://eprints.uthm.edu.my/5949/1/AJ%202018%20%28947%29%20Optimization%20of%20batch%20conditions%20for%20COD%20and%20ammonia%20nitrogen%20removal%20using%20cockle%20shells%20through%20response%20surface%20methodology.pdf
http://eprints.uthm.edu.my/5949/
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Tun Hussein Onn Malaysia
Language: English
Description
Summary:The optimal conditions for the reduction of COD and NH3-N using cockle shells (CS) from a stabilised landfill effluent were analyzed. The influence of two variables (adsorbent dosage and pH) were analysed through the application of response surface methodology (RSM) and central composite design (CCD). Quadratic models were developed for the removals of COD and NH3-N parameters. The optimum conditions for removal of 65.6% and 53.6% for COD and NH3-N respectively was achieved at pH 6.34, adsorbent dosage of 20.21 g having 0.888 desirability value. The model F-value obtained for NH3-N removal Prob. > F value of 0.0001 with F-value of 104.21 was obtained. Similarly the Prob. > F value of < 0.0001 for COD with F-value of 82.74 was obtained, these P-values confirmed the significance of the model. The predicted response versus the experimental response depicted that the experimental data were relatively close to the predicted data. Thus, the generated models significantly enclosed the correlation between the process variables and the response.