Band 10, Heft 4 -

In this work, process modelling, thermodynamic analysis and optimization of stand-alone dry and partial oxidation reforming of methane as well as, the auto-thermal reforming processes were investigated. Firstly, flowsheet models were developed for both the stand-alone systems and auto-thermal reforming process using ASPEN HYSYS ® . Furthermore, thermodynamic studies were conducted for the stand-alone and auto-thermal reforming processes for temperatures range of 200–1000°C and pressure range of 1–3 bar using Gibbs free energy minimization methods which was also performed using ASPEN HYSYS ® . The simulation of the auto-thermal reforming process was also performed at 20 bar to mimic industrial process. Process parameters were optimized in the combined reforming process for hydrogen production using desirability function. The simulation results show that 84.60 kg/h, 62.08 kg/h and 154.7 kg/h of syngas were produced from 144 kg/h, 113 kg/h and 211 kg/h of the gas fed into the Gibbs reactor at CH 4 /CO 2 /O 2 ratio 1:1:1 for the stand-alone dry reforming, partial oxidation reforming and auto-thermal processes respectively. Equilibrium conversion of CH 4 , CO 2 , O 2 were thermodynamically favoured between 400 and 800°C with highest conversions of 100%, 95.9% and 86.7% for O 2 , CO 2 and CH 4 respectively. Highest yield of 99% for H 2 and 40% for CO at 800°C was obtained. The optimum conditions for hydrogen production were obtained at CH 4 /CO 2 , CH 4 /O 2 ratios of 0.634, 0.454 and temperature of 800°C respectively. The results obtained in this study corroborate experimental studies conducted on auto-thermal reforming of methane for hydrogen and syngas production.

In this study, functionalized silica supported ionic liquids phases (SILPs) were synthesized and characterized using Fourier Transform infrared (FTIR) spectroscopy, CHNS elemental analysis, thermo gravimetric analysis (TGA) and scanning electron microscopy (SEM). The adsorption of model naphthenic acid such as hexanoic acid, benzoic acid and commercial naphthenic acid from dodecane/kerosene was also investigated. The silica supported ionic liquids are good adsorbent for the removal of naphthenic acid from highly acidic model oil. The regeneration of supported ionic liquid phases as well as recovery of naphthenic acids from the supported ionic liquid phases (SILPs) was investigated.

A mathematical description is presented for the leaching behavior of cobalt and chromium from their solidified/stabilized forms under the attack from biofilms commonly developed by acid forming bacteria such as Thiobacillus thiooxidans or Thiobacillus ferooxidans. The proposed model predicts the metals diffusion and subsequent removal from cylindrical cement waste forms using a Michaelis-Menten-type kinetics, as a special case of the widely accepted Langmuir-Hinshelwood mechanism at the surface of the encapsulating cylinder. The resulting nonlinear model is solved by applying boundary perturbation to reduce the nonlinear problem to an infinite series of linear problems that are solvable by Laplace transform methods. Model predictions compares well with published experimental data and confirms that a Michaelis-Menten-type kinetics is most probably the dominant mechanism for the leaching of heavy metals from cement based waste forms.

A mathematical model for an ultrasound-assisted vapor liquid equilibrium (VLE) system developed in Aspen Custom Modeler is exported to Aspen Plus flowsheeting environment to form a block in the software library to represent a single stage flash separation unit. The block is then arranged in series to mimic an ultrasonically intensified distillation process. As a case study, an ethanol/ethyl acetate separation process is examined. The results show that a 99 mol% ethyl acetate recovery is established when a series of 27 flash VLE devices are used. The system is also evaluated for other minimum boiling azeotropic mixtures, and consistent results are obtained, thus revealing the potentials of ultrasound-assisted system as an alternative strategy for azeotropic mixture separations.

The non-catalytic supercritical water oxidation (SCWO) of phenol was modeled using Gopalan-Savage and Thornton-Savage global and network rates. Comparison of experimental data for the phenol conversion with the numerical predictions of this study indicated very close compatibility. Applying the validated model, the phenol conversion and selectivity of various products were studied as a function of effective parameters such as feed phenol concentration, reactor residence time, feed temperature, and feed oxygen concentration. The results of modeling analysis show that an appropriate elevated temperature range (460°C < T <500°C) and long residence time (≈90 s) reduce the concentration of hazardous products (i.e., dimers, dibenzofuran, dibenzo-p-dioxin) and maximize the selectivity of environmental benign products such as water and carbon dioxide. Also, high oxygen concentration (≈0.01 mol/L) increase water and carbon dioxide yield. Moreover, high feed phenol concentrations cause a shortcoming for the SCWO system in terms of phenol conversion and selectivity of desirable environmental products. As a consequence, the feed phenol concentration of ≤2 × 10 −3 mol/L is recommended as the appropriate condition.

In this paper, heuristic algorithm based optimization approach is proposed to design the PID controller in Double Feedback Control Loop (DFCL) for a class of stable and unstable Single Input Single Output (SISO) process models. In this work, a three dimensional search is attempted and the heuristic algorithm is employed to find the best possible values for inner loop proportional gain ( K p1 ), outer loop integral gain ( K i ), and derivative gain ( K d ). A weighted sum of Objective Function (OF) is considered to guide the optimization search in order to attain the global best values. A comparative analysis is presented between heuristic algorithms, such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Bacterial Foraging Optimization (BFO) and Firefly Algorithm (FA). From this experimental study, it is confirmed that, FA based DFCL offers better performance in reference tracking and disturbance rejection operations with reduced error values for most of the considered process models compared with the alternatives. The statistical significance of the FA is also verified using Wilcoxon’s rank test.