Charles E. Holland

Elucidation of the ion binding mechanism in heterogeneous carbon-composite adsorbents

Abstract A two-stage analysis of the mechanism of metal ion binding by a heterogeneous adsorbent was developed. In the first stage, a continuous proton affinity distribution was calculated from potentiometric titration data using the CONTIN method with a Langmuir kernel. Electrostatic effects were accounted for using a diffuse layer model. In the second stage, the parameters obtained from the continuous distribution function (i.e. the number of different types of surface sites, site densities and their protonation constants) were utilized in a discrete distribution to represent the adsorbent in surface complexation and double layer models using the GRFIT speciation code. This information on the surface groups was applied to metal ion potentiometric titration experiments to calculate the surface complexation equilibrium constants of the metal ions and hence elucidate the mechanism of ion binding to these sites. The proposed method was applied successfully to the adsorption of Sr and Cu ions on carbon-composite adsorbents, KAU-mod and SCN-mod. The continuous distribution method (CONTIN) revealed three types of surface sites within these carbon-composite adsorbents with pK values ranging between 3.5–4.1, 5.3–6.3 and 7.7–8.2. The analysis of the metal ion adsorption data using the GRFIT speciation code showed that only the first two surface sites were capable of forming surface complexes with the Sr ions, and that only the first site governed the adsorption of the Cu ions.

Solvent Vapor Recovery by Pressure Swing Adsorption. III. Comparison of Simulation with Experiment for the Butane—Activated Carbon System

Abstract A fully predictive (no adjustable parameters), nonisothermal, multicomponent mathematical model was developed and used to simulate a pressure swing adsorption (PSA) process designed for the separation and recovery of concentrated butane vapor from nitrogen using BAX activated carbon. Nearly quantitative agreement with experiment was realized with this model over a wide range of process conditions, and for both the transient and periodic state process dynamics and the periodic state process performance. The model also verified some unique characteristics of this PSA process, and it revealed some of the subtleties associated with accurately simulating a PSA-solvent vapor recovery (SVR) process. These subtleties included the need to account for the adsorbate heat capacity and the temperature dependence of the gas-phase physical properties. No PSA models in the literature have included both of these features, which were critical to the accurate prediction of the heat effects in this PSA-SVR process.

Electrochemical Removal of Carbon Monoxide in Reformate Hydrogen for Fueling Proton Exchange Membrane Fuel Cells

A twin-cell electrochemical filter is demonstrated to reduce the CO concentration in reformate hydrogen. In this design, the potential and gas flow are switched between the two filter cells so that alternative CO adsorption and oxidation occur in each cell while providing a continuous flow of H 2 to a fuel cell. The effects of filter switching time and applied potential on the CO concentration of gas exiting the filter are presented here for a CO concentration of 1000 ppm in nitrogen flowing at 100 cm 3 /min. The parasitic loss of hydrogen from a corresponding reformate stream was estimated to be 1.5%.

Solvent Vapor Recovery by Pressure Swing Adsorption. I. Experimental Transient and Periodic Dynamics of the Butane-Activated Carbon System

ABSTRACT An experimental investigation was carried out for the separation and recovery of butane vapor (10 to 40 vol%) from nitrogen using Westvaco BAX activated carbon in a twin-bed pressure swing adsorption (PSA) system utilizing a 4-step Skarstrom- type cycle. Twenty-four runs, covering a broad range of process and initial column conditions, were performed to investigate the transient and periodic process dynamics. In all cases the approach to the periodic state was very slow, taking up to 160 cycles depending on the initial condition of the beds; and peak bed temperatures of up to 105°C were observed depending on both the initial condition of the beds and the process conditions. Also, the periodic state of each run was unique when approaching a new periodic state from less contaminated beds. The uniqueness of the periodic states, together with the exceedingly high peak temperatures, inferred much about the practice of preconditioning beds to avoid high temperature excursions. The periodic enriched but...

Solvent Vapor Recovery by Pressure Swing Adsorption. II. Experimental Periodic Performance of the Butane-Activated Carbon System

ABSTRACT An experimental investigation was carried out for the separation and recovery of butane vapor (10 to 40 vol%) from nitrogen using Westvaco BAX activated carbon and a unique pressure swing adsorption (PSA)-solvent vapor recovery (SVR) system. The effects of six important process and operating parameters on the periodic process performance were obtained, i.e., the purge-to-feed ratio, purge pressure, volumetric feed flow rate, feed concentration, cycle time, and pressurization/blowdown step time. Overall, the experimental results were consistent with theoretical results in the literature for the effects of most of these parameters; however, some opposite and unique trends were observed. The experimental results verified that the concentration wave front may be contained within the bed even when the purge-to-feed ratio is less than unity, and that the process performance may be very sensitive to minor changes in the purge pressure. Moderate temperature swings (18 to 54°C) were exhibited in all cases...

A flexible multivariable experimental air tank system for process control education

This paper presents an industrially relevant multivariable experimental air pressure tank system which has been developed at the University of South Carolina for process control education. Inspired by experimental systems for liquid level modeling and control of a four-tank system, this pressure control apparatus is quite flexible. It offers a wide variety of uses for both educational and research purposes, and it does so at moderate expense. As opposed to liquid level systems, pressure differences in the system drive the flow, removing limitations in system flexibility associated with gravity driven liquid systems. The four tank system can be configured to exhibit a multivariable right-half plane zero, demonstrating advanced concepts of input directionality and control limitations. A detailed description of the system and a model based on fundamental principles is provided. The current system allows for a computer interface to both MATLAB/Simulink and LabView.