Kenneth B. Bischoff

CO2 absorption in aqueous solutions of hindered amines

Sterically hindered amines constitute a new class of chemicals which have recently come into industrial use in a variety of gas-treating processes: chemical solvents in both aqueous and non-aqueous solutions, rate promotion additives for the hot carbonate process, and chemical solvents for the selective removal of hydrogen sulfide. The scant published data on the behavior of hindered amines do not allow one to estimate the actual values of the apparent kinetic constants or equilibrium constants, and even less to establish the chemical steps involved. In this paper, the results of an investigation of the behavior of one particular amine, 2-amino-2-methyl-1-propanol (AMP), as a chemical solvent for CO2 in aqueous solutions are reported. The equilibrium behavior of a hindered monoamine like AMP in aqueous solutions is dominated by the values of two equilibrium constants: the protonation constant Kp and the carbamate stability constant Kc. The value of Kp at infinite dilution has been determined experimentally, and is large enough to neglect formation of the carbonate ion. The value of Kc has been found experimentally to be significantly less than 10−1 1/gmol, as is expected for a hindered amine. A thermodynamic model has been developed and tested against experimental equilibrium data. Preliminary kinetic data seem to indicate that the reaction with CO2 is first order with respect to both CO2 and AMP. A first step in the elucidation of the chemical steps involved is presented.

Species rank in reaction pathways: Application of Delplot analysis

Abstract A new method of reaction pathway analysis that consists of plotting (molefraction( y p )/ (conversion) r ( x r )) versus conversion ( r is the rank of the Delplot) was recently proposed. In this paper, two species rank are identified, the Delplot rank D ( P ) and the network rank N ( P ). The network rank is the minimum number of slow steps between the reactant and the product. The Delplot rank is r, where the plot of y / x r vs x shows a finite intercept. The network rank is the information sought after in reaction pathway analysis, and the Delplot rank is the information obtained from Delplot analysis. A simple algebraic recursive equation is proposed to find the Delplot rank of a product in a given reaction network. This equation illustrates the coupling between the order of reaction and the rank species. Furthermore, this equation differentiates between network rank and Delplot ranks. Also, an asymptotic analysis that utilizes the network information in the slope of the Delplot is presented. This method is illustrated for sequential reaction networks and for series-parallel reaction networks.

A physiological model for the pharmacokinetics of methylene chloride in B6C3F1 mice following i.v. administrations

A physiologic mathematical model was developed to describe the time course of14C-methylene chloride (14CH2Cl2) distribution and elimination in mice following single i.v. administrations of 10 and 50mg/kg. A whole-body model was used to simulate14CH2Cl2 concentrations in blood and tissues, pulmonary clearance of unchanged14CH2Cl2, and metabolic conversion to14CO2 and14CO as monitored by the appearances of these metabolites in expired breath. This diffusion-limited model was identified via a sequential optimization scheme using hybrid models for each compartment. Pulmonary elimination of unchanged14CH2Cl2 was modeled as a linear process while hepatic metabolism of14CH2Cl2 to the compounds14CO2and14CO was described by a saturable metabolic rate term. The model adequately described the dose dependence in methylene chloride distribution and metabolism when simulations were compared to experimental data.

Lumping analysis in the presence of measurement error

A review of the analysis of our earlier paper is presented, where systems mathematics is utilized to define in a direct way the conditions for exact lumping of monomolecular first-order kinetic systems. A new extension is to show that a system that is exactly lumpable in a batch reactor will follow the same lumping in any reactor type. In addition, our new approach shows how cluster analysis provides a useful way to analyse approximate lumping. This is especially important when measurement error is present, since the implementation of a successful lumping scheme must be viewed in the context of the errors.