Discussion of Modeling Results

Before collecting data, at least two lean/rich cycles of 15-min lean and 5-min rich were completed for the given reaction condition. These cycle times were chosen so as the effluent from all reactors reached steady state. After the initial lean/rich cycles were completed, IR spectra were collected continuously during the switch from fuel rich to fuel lean and then back again to fuel rich. The collection time in the fuel lean and fuel rich phases was maintained at 15 and 5 min, respectively. The catalyst was tested for SNS at all the different reaction conditions and the qualitative discussion of the results can be found in [75], Quantitative analysis of the data required the application of statistical methods to separate the effects of the six factors and their interactions from the inherent noise in the data. Table 11.5 presents the coefficient for all the normalized parameters which were statistically significant. It includes the estimated coefficients for the linear model, similar to Eqn (2), of how SNS is affected by the reaction conditions. 

As Figure 11.26 undoubtedly demonstrates, the deviation between the same catalytic material under practically identical reaction conditions is in the range of 2% conversion (if appropriate measures are taken this error can be reduced to 0.5%). These experimental data points lead to the important verification of the above-discussed CFD modeling results and confirm the assumption of realizing identical reaction conditions over the whole reactor system independent from the position of a catalyst to be tested. By testing inert carrier material in reactor column number 8, the inertness and catalytic inactivity of the reactor steel can be proven. 

The following discussion begins by presenting an in-depth view of the mechanism for the photochemical reduction of benzophenone by N, iV-dimethyl-aniline. This discussion is followed by a presentation of the theoretical models describing the parameters controlling the dynamics of proton-transfer processes. A survey of our experimental studies is then presented, followed by a discussion of these results within the context of other proton-transfer studies. 

As in the recent QCCD study by Head-Gordon et al. (28, 128), we tested the ECCSD, LECCSD, and QECCSD methods, based on eqs (52)-(59), using the minimum basis set STO-3G (145) model of N2. In all correlated calculations, the lowest two core orbitals were kept frozen. As in the earlier section, our discussion of the results focuses on the bond breaking region, where the standard CCSD approach displays, using a phrase borrowed from ref 128, a colossal failure (see Table II and Figure 2). 

After briefly presenting some important milestones of MP2//HF studies in the quantum chemistry description of DNA base pairs, we turn to a more extensive discussion of DPT results for extended DNA-base aggregates, including model stacks and real molecular fragments. 

The hat-curved-harmonic oscillator model, unlike other descriptions of the complex permittivity available now for us [17, 55, 56, 64], gives some insight into the mechanisms governing the experimental spectra. Namely, the estimated relaxation time of a nonrigid dipole (xovib 0.2 ps) is close to that determined in the course of very accurate experimental investigations and of their statistical treatment [17, 54-56]. The reduced parameters presented in Tables XIVA and XIVB and the form of the hat-curved potential well (determined by the parameters u, (3, f) do not show marked dependence on the temperature, while the spectra themselves vary with T in greater extent. We shall continue discussion of these results in Section X.A. 

Multilevel prognostic models may be divided into two groups with respect to the mode of consideration of the vertical coordinate. One group uses the traditional Cartesian vertical coordinate with horizontal levels the other considers a vertical coordinate normalized with respect to the sea depth at the site (the so-called a-coordinate). Let us start the discussion of the results with the first group of models, because they prevail in the studies of the BSGC. 

Before entering into a discussion of our results, it is appropriate to point out the essential differences in approach. The Lehigh group measured chemisorption on powdered samples and deduced the electronic interaction with the surface from a plausible model. We, on the other hand, measured the conductivity and lifetime... 

The discussion of the results shows that all such model calculations are extremely dependent on various assumptions (e.g. the ratio of species in the initial composition). Another source for uncertainties lies in the f-values and rate constants which are in many cases not very well known. The calculations give, however, a valuable survey of the processes which may lead to the formation of the observed o>ma constituents and their relative importances. Also Huebner and Giguere emphasize again the importance to search for other primary molecules, especially to get more reliable estimates whether or in which abundances ammonia and methane are stored in the nucleus. 

Closed-loop identification has been addressed extensively in a linear stochastic control setting (Astrom and Wittenmark, 1989). Good discussions of early results from a stochastic control viewpoint are presented by Box (1976) and Gustavsson et al (1977). Landau and Karimi (1997) provide an evaluation of recursive algorithms for closed-loop identification. Van den Hof and Schrama (1994), Gevers (1993), and Bayard and Mettler (1992) review research on new criteria for closed-loop identification of state space or input-output models for control purposes. 

As mentioned in the introduction, the following discussion on modeling results takes as a lead that distinction should be made between steady-state models, unsteady-state models, and dynamic models. The results mentioned focus mainly on automotive exhaust gas treatment, which application has been widely studied, with major emphasis on the oxidation of carbon monoxide. 

The same approaches that were successful in linear chromatography—the use of either one of several possible liunped kinetic models or of the general rate model — have been applied to the study of nonlinear chromatography. The basic difference results from the replacement of a linear isotherm by a nonlinear one and from the coupling this isotiienn provides between the mass balance equations of the different components of the mixture. This complicates considerably the mathematical problem. Analytical solutions are possible only in a few simple cases. These cases are limited to the band profile of a pure component in frontal analysis and elution, in the case of the reaction-kinetic model (Section 14.2), and to the frontal analysis of a pure component or a binary mixture, if one can assume constant pattern. In all other cases, the use of numerical solutions is necessary. Furthermore, in most studies found in the literature, the diffusion coefficient and the rate constant or coefficient of mass transfer are assumed to be constant and independent of the concentration. Actually, these parameters are often concentration dependent and coupled, which makes the solution of the problem as well as the discussion of experimental results still more complicated. 

Discuss limitations to the potential accuracy and applicability of modeling results obtained in the previous problem. 

XPS (ESCA) and UPS are powerful tools to determine electronic energy levels by irradiating a sample with monochromatic X-rays and UV light, respectively, and measuring the energy distribution of the emitted electrons. With alkali metal suboxides UPS provides a quantitative proof of the bond model. Furthermore, the UPS results offer an explanation for the low energy photoemission process with oxidized Cs and thus (unexpectedly) open up a field of applied research with infrared sensitive photocathodes. The discussion of the results will be focussed on the chemical bonding. 

---