Autocorrelation Test for the Residual Data from the Pseudo-1st Order Kinetic Model of the Brominated Flame Retardant 4-Bromodiphenyl Ether Adsorption onto Biochar-immobilized Sphingomonas sp.

M.Y. Shukor
DOI: https://doi.org/10.54987/jebat.v4i1.583
2021-07-30
Journal of Environmental Bioremediation and Toxicology
Abstract:Because of their fire-retardant properties, polybrominated diphenyl ethers (PBDEs) are frequently used in the manufacturing industry. PBDEs are mixed with polymers as additives and employed in a range of sectors, including plastics and textiles. They are, nevertheless, capable of leaking from the surfaces of these items and into the environment since they are not chemically connected to plastics or textile materials. The adsorption of PBDEs onto biochar-immobilized bacteria is a useful method to remediate PBDEs from the environment. Understanding the kinetics of adsorption can be done by using models such as pseudo-1st or pseudo-2nd. The pseudo-1st order kinetic model was previously found to best fit the data via a nonlinear regression exercise for brominated flame retardant 4-bromodiphenyl ether adsorption onto biochar-immobilized Sphingomonas sp. However, the use of nonlinear regression requires the residual of the fitted curve to be non-autocorrelated. The Durbin–Watson statistic, which is derived from the Durbin–Watson distribution, is one of the most commonly used ways for determining whether or not there is autocorrelation. In this study, the calculated value of the Durbin-Watson statistics was d = 2.260. The Durbin-Watsonâ€TMs lower critical value for dL was 0.700, while the upper critical value dU was 1.252. Since the d value was greater than the upper critical value or dU, this resulted in the null hypothesis not being rejected or indicating that there is no evidence of autocorrelation. This demonstrates that the pseudo-1st model used in the nonlinear regression model is adequate
What problem does this paper attempt to address?