Chemical Engineering

Permanent URI for this community

https://hdl.handle.net/10352/175

Browse

Browsing Chemical Engineering by Author "Banza, Musamba Jean Claude"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Production of adsorptive material from modified nanocrystals cellulose for industrial application
    (Vaal University of Technology, 2022-05) Banza, Musamba Jean Claude; Seodigeng, Tumisang, Prof.; Rutto, Hilary Limo, Prof.
    Water is the essence of life, yet it is progressively polluted by dyes, heavy metal ions, food additives, medicines, detergents, agrochemicals, and other toxins from industrial, municipal, and agricultural sources. Among the different wastewater treatment technologies, adsorption is a technique that, when used in conjunction with a welldesigned system, produces high-quality treated water at a reasonable cost. For water treatment, activated carbon is the most often employed adsorbent. Its manufacture, on the other hand, is energy demanding, costly, and creates greenhouse emissions. As a result, finding low-cost alternative adsorbents from industrial and agricultural waste and biomass has attracted a lot of interest. In this context, developing sustainable platforms for wastewater treatment using sustainable nanomaterials such as cellulose nanocrystals (CNCs) is a unique technique with a low carbon footprint. CNCs, which are made by hydrolyzing pulp fibers in sulfuric acid, are rod-like nanomaterials with a lot of remarkable qualities including high specific surface area, high specific strength, hydrophilicity, biodegradability, and surface functionalization. These characteristics, as well as their accessibility, make them suitable candidates for water treatment applications. However, because of their great colloidal stability and nano-dimensions, extracting these CNCs after usage in water treatment is difficult. To overcome this problem, including these CNCs into nanocomposite systems that can be readily separated after usage in batch and continuous water treatment processes is a great concept. Furthermore, pure CNCs have low selectivity towards a wide range of water pollutants, necessitating surface functionalization to provide this selectivity. As a result, this thesis investigates the extraction of CNCs from millet husk waste and waste papers, the development of CNC-incorporated nanocomposites and evaluation of their adsorption characteristics using batch and fixed bed column adsorption studies, and (ii) the evaluation of the selective adsorption characteristics of surface functionalized CNCs and their ability to tailor the nanocomposites' characteristics for use in water treatment applications. The response surface methodology, artificial neural network, and adaptive neuro-fuzzy inference systems were also applied to model the removal of heavy metal ions. The first part of the research (cellulose nanocrystals extraction and optimization) The cellulose nanocrystals were extracted from millet husk residue waste using a homogenized acid hydrolysis method. The effects of the process variables homogenization speed (A), acid concentration (B), and acid to cellulose ratio (C) on the yield and swelling capacity were investigated and optimized using the Box Behnken design (BBD) method in response surface methodology. The numerical optimization analysis results showed that the maximum yield of CNCs and swelling capacity from cellulose was 93.12 % and 2.81 % obtained at homogenization speed, acid concentration, and acid to cellulose ratio of 7464.0 rpm, 63.40 wt %, and 18.83 wt %, respectively. ANOVA revealed that the most influential parameter in the model was homogenization speed for Yield and acid concentration for swelling capacity. The TGA revealed that cellulose had greater heat stability than CNCs. The functional groups of CNCs and cellulose were identical according to the FTIR data. When compared to cellulose, the SEM picture of CNCs is porous and shows narrow particle size with needle-like shape. The XRD pattern revealed an increase in CNC intensity. The second part of the research (CNCs modification for selective removal) A novel type of recyclable adsorbents with outstanding adsorption capability was produced using CNCs with succinic anhydride and EDTA. and their adsorption properties were studied in detail utilizing batch adsorption experiments of Chromium (VI) in aqueous solution. The effects of several factors on Cr (VI) adsorption were examined, including contact duration, adsorbent dose, starting concentration, pH, and temperature. The cellulose nanocrystals treated with succinic anhydride and EDTA possessed a needle-like form, high porosity, and a narrow particle size distribution. The carboxylate transition of the carboxyl group of cellulose was verified by FTIR. XRD analysis of particles after modification revealed the presence of additional phases, which were attributed to succinic anhydride and EDTA modification. A spontaneous exothermic adsorption process was validated by the observed thermodynamic characteristics. The best model for describing adsorption kinetics and mechanism was a pseudo-second order kinetic and intra-particle diffusion model. The Langmuir adsorption isotherm was seen in equilibrium adsorption data, with a maximum adsorption capacity (qmax) of 387.25± 0.88 mgL-1. We showed that the removal effectiveness of Cr (VI) maintained at 220± 0.78 mg.g-1 after 6 adsorption-desorption cycles, and that the CNC-ALG hydrogel beads are excellent adsorbents for the selective removal of Cr (VI) from wastewaters. The third part of the research (modeling of removal of heavy metal ions using RSM, ANN and quantum mechanism studies) The effects of contact time , pH, nanoparticle dose, and beginning Cd2+ concentration on Cd2+ removal were examined using the central composite design (CCD) technique. The performance and prediction capabilities of Response Surface Methodology (RSM) and Artificial Neural Network (ANN) modelling methodologies were explored, as well as their performance and prediction capacities of the response (adsorption capacity). The process was also described using the adsorption isotherm and kinetic models. Statistical data, on the other hand, revealed that the RSM-CCD model beat the ANN model technique. The optimum conditions were determined to be a pH of 5.73, a contact time of 310 minutes, an initial Cd2+ concentration of 323.04 mg/L, a sorbent dosage of 16.36 mg, and an adsorption capacity of 440.01 mg/g. The spontaneous adsorption process was well characterized by the Langmuir model, and chemisorption was the dominant regulator. The binding energy gaps HOMO-LUMO were used to find the preferred adsorption sites. The fourth part of research (optimization of removal using ANN and ANFIS) An artificial neural network and an adaptive neuro-fuzzy inference system were utilized to predict the adsorption capability of mix hydrogels in the removal of nickel (II) from aqueous solution. Four operational variables were evaluated in the ANFIS model to determine their influence on the adsorption study, including starting Ni (II) concentration (mg/L), pH, contact time (min), and adsorbent dosage (mg/L) as inputs and removal percentage (percent) as the single output. In contrast, 70% of the data was employed to develop the ANN model, while 15% of the data was used in testing and validation. To train the network, feedforward propagation with the Levenberg-Marquardt algorithm was used. To optimise, design, and develop prediction models for Ni (II) adsorption using blend hydrogels, (ANN) and (ANFIS) models were employed for trials. The results demonstrate that the ANN and ANFIS models are viable prediction techniques for metal ion adsorption. The fourth part of research (mechanistic modeling and optimization of removal using ANN, RSM and ANFIS) An artificial neural network, response surface methodology and an adaptive neuro-fuzzy inference system were utilized to predict the adsorption capability of modified cellulose nanocrystals and sodium alginate for the removal of Cr (VI) from aqueous solution. Four variables such as time, pH, concentration, and adsorbent dose were evaluated to determine their influence on the adsorption study. To examine the viability of the models, five statistical functions ( RMSE, ARE, SSE, MSE, and MPSD) were applied. absorption mechanism was described via four mechanistic models (Film diffusion, Weber and Morris, Bangham, and Dummwald-Wagner models. Further statistical indices supported ANFIS as the best prediction model for adsorption compared to ANN and RSM. Film diffusion was identified as the rate-limiting process via mechanistic modelling. The sixth part of research (continuous fixed-bed column study) The hydrogel's technical feasibility for adsorption of Cu2+, Ni2+, +Cd2+, and Zn2+ ions from the packed bed column's produced AMD was assessed. The hydrogel was considered to have a high potential for significant interactions with dangerous metal ions. This characteristic, together with the adsorbent's availability, low cost, and efficient regeneration of the spent adsorbent, distinguishes it from the many other adsorbents described in the literature by other researchers. With a bed height of 25 cm, an influent metal ion concentration of 10 mg/l, and a flow velocity of 10 ml/min, the bed performed better. As a consequence, the breakthrough curve for the packed bed experiment shows that the breakthrough points were approached sooner by increasing the flow rate and influent concentration, and later by increasing the bed height. The experimental results were satisfactorily described by the BDST, Yoon–Nelson, and Thomas models. The hydrogels had a net-work structure and more homogeneous porosity, according to the SEM, TGA, XRD, and FTIR results for CNCs. The hydrogels revealed varied degrees of opacity and heavy metal ions absorption capacity depending on the temperature of the analysis. Diffraction confirmed the existence of crystalline structures and the presence of carboxyl and amide groups.