Three-dimensional computational fluid dynamics and Lagrangian particle deposition models were developed to compare the deposition of aerosolized Bacillus anthracis spores in the respiratory airways of a human with that of the rabbit, a species commonly used in the study of anthrax disease. The respiratory airway geometries for each species were derived respectively from computed tomography (CT) and µCT images. Both models encompassed airways that extended from the external nose to the lung with a total of 272 outlets in the human model and 2878 outlets in the rabbit model. All simulations of spore deposition were conducted under transient, inhalation–exhalation breathing conditions using average species-specific minute volumes. Two different exposure scenarios were modeled in the rabbit based upon experimental inhalation studies. For comparison, human simulations were conducted at the highest exposure concentration used during the rabbit experimental exposures. Results demonstrated that regional spore deposition patterns were sensitive to airway geometry and ventilation profiles. Due to the complex airway geometries in the rabbit nose, higher spore deposition efficiency was predicted in the nasal sinus compared to the human at the same air concentration of anthrax spores. In contrast, higher spore deposition was predicted in the lower conducting airways of the human compared to the rabbit lung due to differences in airway branching pattern. This information can be used to refine published and ongoing biokinetic models of inhalation anthrax spore exposures, which currently estimate deposited spore concentrations based solely upon exposure concentrations and inhaled doses that do not factor in species-specific anatomy and physiology for deposition.

A method is presented for optimizing a desired electrochemical reaction in the presence of undesired reactions by the optimal choice of a time varying control variable, such as the current or voltage waveform. The method is illustrated by a PEM fuel cell example: computing the maximum power production from the oxidation of H-2 on Pt/Ru in the presence of 1 per cent CO at 50 C and 1 atm. Dynamic programming, in combination with a surface coverage anode model, is used to show that the optimal solution yields a series of pulse-like overvoltage waveforms - a generalization of previous heuristic pulsing approaches to increasing power in PEM fuel cells in H-2 with CO. Essentially, the solution indicates how to optimally allocate electrochemical resources between two competing tasks: investing electrical energy in freeing sites for the desired reaction and producing electrical energy from the desired reaction. Because dynamic programming is computationally intensive, an efficient technique is presented for implementing the method using a priori predictions of the family of optimal performance paths and using experimental feedback from estimates of surface coverage based upon observer theory. The observers are shown to converge to a limited set of experimental data. (C) 2011 The Electrochemical Society. [DOI: 10.1149/2.063202jes] All rights reserved.

Improved CO tolerance in PEM fuel cells was achieved by periodically varying the anodic overvoltage to convert CO to CO2 directly on the anode. The conversion was controlled by a feedback control algorithm, which used current pulsing and time- varying flow rate parameters as the control variables. Single cell performance data was obtained with 1 and 3 percent CO in a synthetic reformate mixture at 50 C using conventional catalysts and a Nafion 115 membrane. Favorable comparisons are made to 20 ppm CO in H2 for conventional operation and to representative DMFC performance. Durability remains a challenge, but preliminary data with less degradation is presented.

This report was prepared as an account of work sponsored by ARTI under its HVAC&R Research for the 21st Century Program. It presents the results of the first phase of a multi-phase research program that is intended to develop an understanding of re-entrainment of flue gases and building-air exhaust in packaged HVAC equipment. The objectives of Phase I were to determine the design factors that affect the amount of air that is re-entrained in packaged HVAC equipment, and to identify the relationships between the percentage of exhaust air re-entrained and separation distances, direction of airflow, and air velocities. This report can be downloaded from: www.arti-21cr.org/research/completed/index.html.