Bioremediation of estrone from water matrices using the enzyme laccase combined with mathematical modelling

Jenidi, Youla (2017) Bioremediation of estrone from water matrices using the enzyme laccase combined with mathematical modelling. PhD thesis, University of Nottingham.

[img]
Preview
PDF (BIOREMEDIATION OF ESTRONE FROM WATER MATRICES USING THE ENZYME LACCASE COMBINED WITH MATHEMATICAL MODELLING) (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (3MB) | Preview

Abstract

The presence and impact of steroid estrogens in natural water matrices has driven development and evaluation of wastewater treatment technologies that may reduce the steroid load entering water environments. This work was undertaken to assess and predict the ability of Trametes versicolor laccase to degrade estrone (E1) in water matrices under realistic conditions to wastewater treatment plants (WWTPs) and with consideration of the complex and variable nature of the wastewater matrix. A robust experimental procedure was developed to ensure the efficiency of the enzyme laccase to degrade E1 in water matrices was not overestimated due to errors arising from poor experimental design. These experiments demonstrated that commercially-obtained laccase in concentrations above >1 mg/ml are inhomogeneous requiring centrifugation prior to use to reduce error and provide more accurate evaluation of laccase capability. Sample filtration, which is necessary for chromatographic analysis, identified regenerated cellulose (RC) membrane filters as the optimum filters for particulates removal from E1 solutions due to their low affinity toward E1 (3.2 ±1.72 %). An optimum enzyme inactivation procedure using hydrochloric acid was also developed to ensure that the enzyme laccase was instantly inactivated without affecting the target steroid E1 itself.

Using the established experimental procedure, bench-scale studies evaluating the efficiency of laccase-based treatment in a ‘clean’ water matrix were investigated. Experiments in deionised water provided a proof of concept of laccase ability to degrade E1 in water under realistic ranges of temperature [6˚C - 25˚C] and contact time [0.5 hr – 8 hrs] to the WWTP and evaluate the use of models to fit experimental data and predict within that system. Box Behnken Design (BBD) was applied to determine the number and the conditions of the performed experiments. The experimental data was then utilised to build two different models to predict E1 removal efficiency under any set of conditions and optimise the performance of laccase-based treatment system. The goodness of the fit for each model was tested using statistical indices such as coefficient of determination (R2), mean squared error (MSE) and absolute average deviation (AAD). The artificial neural network (ANN) model showed a better fit to the experimental data than the response surface methodology (RSM) model (RSM and ANN of R2 = 0.9908 and R2 = 0.9992 respectively. In addition, the predictive capabilities of RSM and ANN were tested using a set of statistically designed unseen data that was not previously used in models’ training. Both models showed limited predictive capabilities.

The ability of laccase-based treatment to remove E1 in real-world wastewater was studied at bench scale. To account for the complexity and variability of the wastewater matrix, effluent samples during the period December 2014 - June 2015 were characterised for standard water quality parameters, where the temporal variation in wastewater chemical oxygen demand (COD), total suspended solids (TSS) and pH, were observed. A new water quality parameter, “Benchmark” was also developed and applied to quantify the impact of wastewater variability on laccase performance for E1 removal. The average benchmark value in the period between December 2014 and June 2015 was 79.8±3.7%. In addition, the impact of laccase inhibitors, which are likely to be present within the wastewater matrix, such as chloride, copper, iron and zinc, on laccase activity was investigated. The inhibitory effect of chloride ions increased with increasing chloride concentration above 200 mg/l. Copper and zinc ions exhibited negative effects on the enzymatic degradation of E1 at concentrations equal or above 10 mg/l and 200 mg/l.

The impact of water matrix temperature, contact time and laccase concentration were studied in wastewater effluent and the experimental data was used to build RSM and ANN models. The predictive capability of the generated RSM model was relatively poor (R2 = 0.863) and even lower than the achieved predictive capability in clean matrix when tested using unseen data, this was partially attributed to the variability of wastewater matrix that could have not been addressed in this type of models. Whilst the improved ANN model showed a better predictive capability than RSM (R2=0.991) An advantage of the ANN model compared to the RSM model and reported for the first time, was the ability to include the impact of matrix complexity and variability on laccase performance, assessed via the benchmark data added as a forth factor in the ANN model. The final ANN model incorporating the matrix variability observed temporally during the sampling period had extremely high predictive capabilities (R2 > 0.99). This model approach holds the potential to help researchers evaluate and optimise laccase-based treatment (as well as other treatment technologies) and predict the removal efficiency of various bioactive chemicals under a wide range of conditions. Performing laccase-based treatment in a continuous reactor, utilising actual wastewater effluent and under realistic conditions to WWTPs, is the next stage that should be investigated in detail.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Gomes, R.L.
Stephens, G.
Keywords: steroids, enzyme, laccase, wastewater treatment, artificial neural networks (ANN), response surface methodology (RSM), Modelling, bioactive chemicals, estrone
Subjects: T Technology > TD Environmental technology. Sanitary engineering
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 42868
Depositing User: Jenidi, Youla
Date Deposited: 13 Jul 2017 04:40
Last Modified: 07 May 2020 12:32
URI: https://eprints.nottingham.ac.uk/id/eprint/42868

Actions (Archive Staff Only)

Edit View Edit View