In this study, three chemical scrubbing processes were investigated to test their performances for removing odorous compounds in waste gases emitted from feed-drying operations. A gas chromatography with a flame ionization detector (GC/FID) and gas detector tubes were used for the identification of odorous species and their concentrations in the waste gases. Results indicate that toluene, ethyl benzene, xylenes, and styrene could be detected while their concentrations were all below the detection limits. However, trimethylamine, amine, mercaptans, ammonia, hydrogen sulfide, and acetic acid could be detected with the detector tubes. A two-stage scrubbing system with a gas flow rate of 10 L/min was used to test its performance for the odor removal. Results demonstrate that an ¡§acidic chlorination-alkaline sulfite scrubbing¡¨ process was effective. The pertinent operating conditions were found to be: (1) a pH of 6.4-6.6 and total residual chlorine of 55-85 mg/L for the acidic-chlorination solution, and (2) a pH of 8.0-8.5 for the alkaline sulfite solution with a sulfite concentration of 10,000 mg/L (as SO2). By the conditions, ammonia, trimethylamine, and amine in the odorous gas could be completely removed. A sensory measurement indicates that the odor concentration could be reduced from 17,378 to 232 after scrubbing. Cost estimation indicates that it required a chemical cost of NT$ 5.74 for removing the odors from 1,000 m3 of the gas. This study has successfully developed an economical and effective chemical scrubbing technology for the removal of odorous gas emitted from feed-drying operations. It could be applied to the treatment of odorous gases emitted from the manufacture of feeds, seafood, and other related operations.

Rigorous thermodynamic and kinetic models were developed in Aspen Plus® Rate SepTM for 8 m PZ, 5 m PZ, 7 m MDEA/2 m PZ, and 5 m MDEA/5 m PZ. Thermodynamic data was regressed using a sequential regression methodology, and incorporated data for all amine, amine/water, and amine/water/CO₂ systems. The sensitivity of CO₂ absorption rate was determined in a wetted wall column simulation in Aspen Plus®, and the results were used in Microsoft Excel to determine the optimum reaction rates, activation energies, and binary diffusivities. Density, viscosity, and binary diffusivity are calculated using user-supplied FORTRAN subroutines rather than built-in Aspen Plus® correlations. Three absorber configurations were tested: adiabatic, in-and-out intercooling, and pump-around intercooling. The two intercooled configurations demonstrated comparable improvement in capacity and packing area, with the greatest improvement in 8 m PZ occurring between lean loadings of 0.20 and 0.25 mol CO₂/mol alkalinity. The effects of absorber temperature and CO₂ removal were tested in the adiabatic and in-and-out intercooled configurations. For 7 m MDEA/2 m PZ at a lean loading of 0.13 mol CO₂/mol alkalinity reducing the absorber temperature from 40 °C to 20 °C increases capacity by 64% without an appreciable increase in packing area. Increasing CO₂ removal from 90% to 99% does not double the packing area due to favorable reaction rates at the lean end of the absorber. Two stripper configurations were tested: the simple stripper and the advanced flash stripper. For all amines, absorber configurations, and lean loadings the advanced flash stripper demonstrated the better energy performance, with the greatest benefit occurring at low lean loadings. An economic estimation method was developed that converts purchased equipment cost and equivalent work to $/MT CO₂. The method is based on economic factors proposed by DOE-NETL and IEAGHG. The total cost of CO₂ decreases as lean loading decreases for all amines and configurations. Increasing CO₂ removal from 90% to 99% results in a 1% increase in the total cost of CO₂ capture. Decreasing absorber temperature for 7 m MDEA/2 m PZ from 40 °C to 20 °C decreases total cost of CO₂ capture by up to 9.3%. / text

Abstract This study armed to develop a chemical scrubbing process to treat an emitted odorous gas stream from a corn-germ oil-pressing machine. Analytical results demonstrated that the gas contains a total VOC (volatile organic compound) concentration of 323 to 2,100 ppm calibrated as methane. Major VOC ingredients are aldehydes, organic acids, and mercaptans. Ozone, sodium hypochlorite (NaOCl), and hydrogen peroxide (H2O2) were used as oxidants for VOCs other than aldehydes and sodium bisulfite (NaHSO3) was used as a reductant for aldehydes. Experimental setup used in this study consisted of a set of four-stage-in-series scrubbers for VOC oxidation and a set of 18-L bubble column for aldehyde reduction. Each stage of the scrubber was constructed from a 25-cm x 140-cm (OD x H) plastic column packed with 2¡¨-diameter plastic packings to a height of 90 cm. Experiment results showed that suitable conditions for the VOC and odor removal in the oxidation tower are: (1) QL/QG (flowrate ratio of circulating liquid and gas) = 0.0024-0.0035 m3 liquid/m3 gas, (2) dose rate of NaOCl solution (10 % effective chlorine) = 0.0125-0.015 L/m3 gas, (3) pH of the circulating liquid = 7.5-8.0, and (4) supplemental water flowrate = 2.4-5.0 L/m3 gas. Those for the reduction of the residual aldehydes and other odorous compounds are: (1) concentration of NaHSO3 scrubbering solution = 2,156 mg/L as SO2 and (2) pH of the NaHSO3 solution = 5.0-6.0. Under the conditions, experimental data demonstrated that the process could reduce H2S from 0-5.5 to 0 ppm, NH3 from 0.1-0.4 to 0-0.1 ppm, phenols from 0-2.0 to 0 ppm, acids from 15-25 to 0 ppm, acetaldehyde from 10-34 to 0-10 ppm, amines from 0.1 to 0.1 ppm, and mercaptans from 0.5-10.5 ppm to 0 ppm. Overall VOC removal efficacy was 85-100 %. Effluent gas from the oxidation stage had a chlorine smell, while that from the reduction one was odorless. Based on the results from the pilot study, a full-scale plant with a waste gas flow rate 105 m3/min was proposed. It was estimated that the full-scale plant has an equipment cost of US$ 45,000, and an operating cost of around US$ 110/(8-hr day) or US$ 0.75/(1,000 m3 waste gas).

In this study, a set of full-scale two-stage-in-series scrubbing tower was constructed to test its performance for desulfurization of biogas from three anaerobic UASB (upflow anaerobic sludge blanket) reactors of a wastewater plant for treating corn-syrup processing wastewater. Each stage of the absorbing towers was constructed from a 0.6m ¡Ñ 6 m (I.D. ¡Ñ H) plastic column packed with plastic packings (4¡¨ K6 Heilex-Type or 2.5¡¨ Crown-Biopacks with packing heights of 5 m and 3.65 m, respectively). Effluent from the activated sludge sedimentation tank of the wastewater plant was used as a scrubbing liquid and the tower effluent liquid was discharged into the activated sludge tank for oxidation of the absorbed hydrogen sulfide. Results of a continuous operation period of 20 weeks indicate that H2S of 380-19,000 ppm in the biogas could be reduced to 1-2,100 ppm by the scrubbing towers with superficial liquid and gas velocities of GL = 13.2-28.1 and GV = 306-642 m/h, respectively. On an average, the effluent gas contained a H2S concentration of less than 300 ppm and the removal efficacy was more than 96%. Results also indicate that as pH of the scrubbing liquid increased from 7.7 to 8.8, the overall H2S removal efficacy raised from 86 to 98%. About 1/3 of the absorbed H2S-S was detected as H2S-S and SO4-2-S in the tower effluent. The rest was supposed to be as elemental sulfur in a slimy form which sticked to the packing surface. This might be the reason for the increasing pressure drop with the operation time. A set of performance and gas-liquid equilibrium equations were proposed to estimated the influence of liquid pH, gas and liquid flow rates, and packing height on the H2S removal efficacy. Calculation results indicate, at 30oC, the minimum liquid/gas ratios (L/G)min for the absorption are 0.043 and 0.014 m3/m3, respectively, for fresh absorbing liquids with pH = 8.0 and 8.5. Based on the proposed equations and the experimental data, the volumetric mass-transfer coefficient Kya for H2S absorption in the towers could be expressed approximately as Kya =1.15¡Ñ1017GV1.0GL-3.0 with GV =304-644 and GL = 15.1-28.1 m3/m2.h, and liquid pH 7.7-8.6. It was supposed that too much liquid loading leads to a decrease in the specific area a for mass transfer which results in the Kya decrease.

This thesis seeks to improve the economic viability of carbon capture process by reducing the energy requirement of amine scrubbing technology. High steam requirement for solvent regeneration in this technology can be reduced by improvements in the regeneration process. Solvent models based on experimental results have been created by previous researchers and are available for simulation and process modeling in Aspen Plus®. Standard process modeling specifications are developed and multiple regeneration processes are compared for piperazine (a cyclic diamine) in Chapter 2. The configurations were optimized to identify optimal operating conditions for energy performance. These processes utilize methods of better heat recovery and effective separation and show 2 to 8% improvement in energy requirement as compared to conventional absorber-stripper configuration. The best configuration is the interheated stripper which requires equivalent work of 29.9 kJ/mol CO₂ compared to 32.6 kJ/mol CO₂ for the simple stripper. The Fawkes and Independence solvent models were used for modeling and simulation. A new regeneration configuration called the advanced flash stripper (patent pending) was developed and simulated using the Independence model. Multiple complex levels of the process were simulated and results show more than 10% improvement in energy performance. Multiple cases of operating conditions and process specifications were simulated and the best case requires equivalent work of 29 kJ/mol CO₂. This work also includes modeling and simulation of pilot plant campaigns carried out for demonstration of a piperazine with a 2-stage flash on at 1 tpd CO₂. Reconciliation of data was done in Aspen Plus for solvent model validation. The solvent model predicted results consistent with the measured values. A systematic error of approximately +5% was found in the rich CO₂, that can be attributed to laboratory measurement errors, instrument measurement errors, and standard deviation in solvent model data. Stripper Modeling for CO₂ capture from natural gas combustion was done under a project by TOTAL through the Process Science and Technology Center. Two configurations were simulated for each of three flue gas conditions (corresponding to 3%, 6% and 9% CO₂). Best cases for the three conditions of flue gas require 34.9, 33.1 and 31.6 kJ/mol CO₂. / text

This dissertation presents a series of studies relating to pad-wafer and brush-wafer contact characterization in planarization and post-planarization processes. These are also evaluated with the purposes of minimizing environmental impact and reducting cost of ownership.Firstly, a new method using spectral analysis based on real-time raw friction data is developed to quantify the total amount of mechanical interaction in the brush-fluid-wafer interface in terms of stick-slip phenomena in post-planarization scrubbing. This new method is remarkable from the standpoint of its potential to eliminate having to perform a multitude of experiments needed for constructing and interpreting Stribeck curves, and its application to processes where Stribeck curves fail to yield any useful data. Moreover, this method is applied to investigate the effect of brush roller design on scrubbing process and to analyze behaviors of eccentric brushes.In order to study pad-wafer contact in planarization processes, a mechanical characterization method (incremental loading test) is developed and applied to analyze different types of pads and pad surfaces subjected to various treatments. Along with optical interferometry and theoretical analysis, flow resistance due to pad land area topography can be estimated.The greatest contribution of this dissertation involves development of real pad-wafer contact area measurement technique using confocal microscopy. The real pad-wafer contact area is a difficult property to measure in planarization, yet it is a key feature to further understand the process. A custom-made sample holder with a sapphire window and a miniature load cell is used to collect confocal images at controlled values of down force.At last, the two newly developed techniques (incremental loading test and real pad-wafer contact area measurement using confocal microscopy) together with dual emission UV enhanced fluorescence imaging are utilized to investigate conditioning effects in planarization process.

The Versal Adaptive Compute Acceleration Platform (Versal ACAP) is a system-on-chip (SoC) developed by AMD Xilinx. To help protect the programmable logic from soft errors, the configuration needs to be constantly checked and repaired through a process called scrubbing. This thesis provides a methodology for scrubbing the configuration over JTAG. The scrubber uses two platform device image (PDI) files, one to read the configuration and one to send corrected frames. The methodology is characterized to determine the time it takes to completely scrub the configuration. The designed scrubber was able to scrub the VM1802 in 11.5 seconds, or 41.6 Mbits/second, when the JTAG interface was operated at 50MHz.

No description available.

The solvent 7 molal (m) methyldiethanolamine (MDEA)/2 m piperazine (PZ) presents an attractive option to industry standard solvents including monoethanolamine (MEA) for carbon dioxide (CO₂) capture in coal-fired power plant flue gas scrubbing applications. The solvent was tested under thermal and oxidizing conditions, including temperature cycling in the Integrated Solvent Degradation Apparatus (ISDA), to measure rates of degradation for comparison to other solvents. Unloaded 7 m MDEA/2 m PZ was generally thermally stable up to 150 °C, exhibiting very low loss rates. However, at a loading of 0.25 mol CO2/mol alkalinity, loss rates of 0.17 ± 0.21 and 0.24 ± 0.06 mM/hr, respectively, for MDEA and PZ were measured. No amine loss was observed in the unloaded blend. Thermal degradation was modeled as first-order in [MDEAH⁺], and a universal Ea for amine loss was estimated at 104 kJ/mol. An oxidative degradation model for 7 m MDEA was developed based on the ISDA data. From the model, the rate of amine loss in 7 m MDEA/2 m PZ was estimated at 1.3 X 10⁵ kg/yr, based on a 500 MW power plant and 90% CO₂ capture. In terms of amine loss, the solvent can be ranked with other cycled solvents from greatest to least as follows: 7 m MDEA>7 m MDEA/2 m PZ>8 m PZ. Thermal degradation pathways and mechanisms for 7 m MDEA/2 m PZ include SN2 substitution reactions to form diethanolamine (DEA), methylaminoethanol (MAE), 1-methylpiperazine (1-MPZ), and 1,4-dimethylpiperazine (1,4-DMPZ). The formation of the amino acids bicine and hydroxyethyl sarcosine (HES) has been directly tied to the formation of DEA and MAE, respectively, through oxidation. As a result of the construction and operation of the ISDA for cycling of solvents from an oxidative reactor to a thermal reactor, several practical findings related to solvent degradation were made. The ISDA results demonstrated that increasing dissolved oxygen in solvents leaving the absorber will increase the rate of oxidation. A simple N2 gas stripping method was tested and resulted in a reduction to 1/5th the high temperature oxidation rate associated with dissolved oxygen present in the higher temperature regions of an absorber/stripper system. The ISDA experiments also demonstrated the need to minimize entrained gas bubbles in absorber/stripper systems to control oxidation. When the ISDA was modified to intercept entrained gas bubbles, the oxidation rate was reduced 2 to 3X. / text

O objetivo deste trabalho foi investigar, sob o enfoque da modelagem matemática, a operação de escrubagem da bauxita proveniente da jazida Miltônia 3, da Mina de Bauxita de Paragominas-PA, com vistas a fornecer parâmetros para previsão de seu desempenho e otimização. Para isso, foi realizada uma campanha de experimentos de escrubagem, em um tambor desagregador de laboratório, baseada em planejamento fatorial. Os parâmetros operacionais avaliados foram os seguintes: grau de enchimento, tempo de residência da polpa e velocidade de rotação. A variável de resposta selecionada foi a quantidade de finos (partículas menores que 0,037 mm) no produto desagregado. O programa de experimentos permitiu a análise da influência de cada variável operacional selecionada na desagregação da bauxita. De acordo com os resultados obtidos, o parâmetro operacional que produziu o efeito mais significativo na variável de resposta foi o grau de enchimento. O modelo desenvolvido foi validado através da comparação entre os valores obtidos em ensaios de escrubagem em uma unidade piloto e aqueles previstos pelo modelo. Os valores da quantidade de finos, no produto desagregado, previstos pelo modelo apresentaram uma excelente aproximação com os dados experimentais da operação em escala piloto. / The aim of this study was to investigate and model the bauxite scrubbing of bauxite samples from Miltonia 3, a Vale operation at Pará state, Brazil. The experimental program included the design of a standard laboratory test, from which parameters were derived for predicting the operation of a scrubber in steady state conditions. Three main variables were selected for the laboratory experimental program using the factorial design technique. These were load fraction, residence time and rotation speed. The amount of fines was determined through screening both feed and product of the scrubbing test. The former was considered as a material characteristic while the second was the dependent variable, i.e. the result of the scrubbing process. An empirical model was developed according to which the load fraction was found the most important variable to the scrubbing process. Residence time was also included in the model due to its importance in designing scrubbers for industrial plants. To validate the model a comprehensive pilot plant program was carried out with the same bauxite sample from Miltonia 3 deposit used in the laboratory investigations. The comparison between experimental data and model calculated values indicated a good agreement, as most values were within ±10% deviation range.