Differential conversion

Laboratory operation of equipment with a fixed bed of granules is not highly satisfactory because of difficulty of temperature control and measurement in both radial and axial directions. A short packed bed with extensive recycle, however, can achieve substantially isothermal behavior and measurable differential conversion. 

Third, there may be a concentration gradient of reactants and products along the length of the catalyst bed. If the structure of the catalyst depends upon the composition of the gas phase, then an average of the various structures will be measured. There is little discussion of this topic in the literature of XAFS spectroscopy of working catalysts. An extreme example of structural variations within a sample is discussed in Section 6, where there is a discussion of XAFS spatially resolved spectra recorded to allow direct observation of the axial distribution of phases present. If the XAFS data are not measured with spatial resolution, then it is recommended that XAFS data be measured under differential conversion conditions. However, if the aim of the experiment is to relate the catalyst structure directly to that in some industrial catalytic processes, then differential conversion conditions will only reflect the structure of the catalyst at the inlet of the bed. To learn about the structure of the catalyst near the outlet of the bed, the reaction has to be conducted at high conversions. If it is anticipated that this operation will lead to variations in the catalyst structure along the bed, then the feed to the micro-reactor should be one that mimics the concentration of reactants toward the downstream end of the bed (i.e., products should be added to the reactants). 

A simple mass balance can be obtained for the recycling system when the following conditions are fulfilled (i) the whole system operates under well-stirred conditions, (ii) the ratio Vp/Vj is << 1, and (iii) the recirculating flow rate is high such as to have differential conversion per pass in the photoreactor and, at the same time, improve mixing. Then, it can be shown (Cassano and Alfano, 2000) that the changes in concentration in the tank are related to the reaction rates according to Equations (30) and (31)... 

Employing a high recirculating flow rate in this small laboratory reactor, the following assumptions can be used (i) there is a differential conversion per pass in the reactor, (ii) the system is perfectly stirred, (iii) there are no mass transport limitations. Also, it can be assumed that (iv) the chemical reaction occurs only at the solid-liquid interface (Minero et al., 1992) and (v) direct photolysis is neglected (Satuf et al., 2007a). As a result, the mass balance for the species i in the system takes the following form (Cassano and Alfano, 2000) ... 

In fundamental catalysis studies catalysts are quite often tested under conditions which differ widely fi om the industrial practice of a continuous process, e.g., tests are carried out in batch using model feedstocks, in stirred reactors, with powdered catalyst or single pellets at conversions that are quite different from those in practice (e.g., differential conversions). While such tests can yield valuable... 

Thus, a small tubular reactor that gives differential conversion (i.e., typically below 5 percent) can yield a point value for the reaction rate. In this case, the reaction rate is evaluated at Cf. Actually, the rate could be better calculated with the arithmetic mean of the inlet and outlet concentrations ... 

A study of the kinetics of methanol synthesis at differential conversion was performed in order to eliminate the influence of the products. The only products observed were methanol, CO and water. Data was obtained as initial activity measurements to eliminate the interference from catalyst deactivation in comparisons of the activities of different catalysts. The initial rate of deactivation in the microflow reactor at differential conditions was < 1%/hour for all catalysts. Each data point reported in Figure 2 and Table 2 is the initial activity from a unique experiment using a fresh catalyst charge. 

In order to establish the experimental parameters necessary for differential conditions, we studied the effect of increasing flow rate of COj/Hj through the microflow reactor on methanol yields and production rate over Cu/Zn/Al-1. At flow rates above 650 mol h" gc , where the methanol yields were < 0.33%, the rate of methanol production became independent of flow rate, signalling that true differential reaction conditions were achieved (Figure 2). This constant methanol production rate, 0.45 mol h gc , was the intrinsic forward rate of COj hydrogenation to methanol at the given conditions for these catalysts. The high space velocity necessary to achieve differential conversion under COj/Hj is... 

Table 2. Methanol production from COj/H at differential conversion ...

CO production by the reverse water-gas shift reaction reached differential conversion at relatively low flow rates compared to methanol production. Above a flow rate of 150 mol h gc , the CO production rate became approximately constant at 0.10 mol h gc , corresponding to CO yields < 0.33%. In other words, the intrinsic rate of CO2 hydrogenation to methanol was much faster than the reverse water-gas shift reaction. Similar results were obtained over the three catalysts providing no evidence for Pd promotion of the reverse water-gas shift reaction. 

Table 3. Methanol production from COj/Hj (differential conversion), COj/Hj + H O (differential conversion) and COj/Hj (finite conversion, internal recycle reactor)...

The results at differential conversions with water addition can be compared with methanol production at the finite conversion in the internal recycle reactor where the water concentration as a result of water production was similar (Table 3). The two types of experiment are analogous in that at differential conditions in the microflow reactor the catalyst was uniformly exposed to the feed concentration, whereas at finite conversions in the internal recycle reactor the catalyst was uniformly exposed to the product concentration. The methanol production rate at finite conversion was similar to the methanol production rate from COj/Hj/HjO at differential conditions for both the Cu/Zn/Al-1 and Pd impregnated catalyst. Therefore, the kinetics at the particular finite conversions, well away from equilibrium, can also be described by methanol production by CO2 hydrogenation, and the inhibition of this reaction associated with the presence of the product water. Furthermore, the Pd promotion was similar under the two reaction regimes (Table 3), reinforcing the conclusion that Pd promotion of CO2 hydrogenation is active only in the presence of water. 

Results at true differential conversions have demonstrated that the intrinsic rate of COj hydrogenation is unaffected by the presence of Pd, whether it is incorporated by impregnation or added in a physical mixture, in the absence of water. For catalytic conversion of CO2 to methanol, a limiting factor at industrially useful yields is the high concentration of product water the effect of which can be ameliorated to some extent by Pd probably due to hydrogen spillover. 

As stated earlier, the product distribution during butadiene epoxidation over an unpromoted catalyst indicated that epoxybutene was strongly bound to the Ag surface and that the CsCI promoter lowered the desorption energy of epoxybutene. These observations should be reflected in the steady-state kinetics of the reaction. The data summarized in Table 5 list the steady-state reaction conditions used to determine the reaction orders for the reactants C4H6 and O2 as well the reaction products epoxybutene, CO2, and H2O. In all these experiments differential conversions of C4H6 and O2 were maintained and the data fitted to the typical power rate law expression for epoxybutene formation... 

The two rate constants of Equation 19, klt and k21, are, respectively, proportional to the net dehydrogenation rate of step 1 and that of step 2. Thus, if experimental conditions are chosen so that the isomerization reaction rate between 1-butene and 2-butenes is slow compared with the two dehydrogenation rates of n-butane, then the concentration distributions of 1-butene and 2-butenes in the reaction products provide information concerning the relative value of kit and k2t. To fulfill the above experimental conditions, we performed several experimental runs with a very low partial pressure of n-butane at differential conversion levels. Analyses of these experimental data indicated that the value of klt may be approximately 10 times larger than that of k2ty provided that the reaction scheme shown by Equation 23 is correct. From Table III the ratio of kit/k2t can be calculated to be 15. The separation of these reactions will be studied further. 

Rates were calculated using a plug-flow reactor design equation and the differential conversion of n-butane to each of the products. Activity decreases quickly within ten minutes and deactivates significantly more slowly for the remainder of the reaction up to one hour. Rapid deactivation followed by much slower deactivation on sulfated zirconia has been reported by several authors (9). 

This article presents a discussion of substrate transport in polymer-immobilized catalyst systems. The mathematical formalism necessary to model a reaction system is presented and forms the basis for qualitative comments on substrate transport. The mathematical formalism also provides a common point for published reaction data and its interpretation. The mathematical models are developed from the perspective of a small scale experimental study and therefore are written for the most common research reactor for polymer-immobilized studies, a batch reactor, and for a fixed-bed reactor operating at differential conversion. 

Rates are established in fixed-bed reactors operating at differential conversion (generally X 0.05) by means of... 

For the reactor used in these studies [2-5] only low conversions could be utilized to obtain differential conversion levels (< 2% based on CO). This limitation on conversion was imposed because of the exotherm of the FTS reaction. The FTS reaction usually takes days to stabalize to make a correct assessment of the steady-state activity and selectivity of the catalyst. This stabalization period was not practiced in the work in references 2-5. 

This is illustrated in the top half of Figure 7.14. A difierential reactor, on the other hand, uses a differential amount of catalyst (usually less than 1 g) in which a differential conversion (less than 1-2%) occurs, so that the rate may be obtained directly as... 

Figure 3.40 shows a schematic diagram of a continuous recycle reactor, CRR. With high values of F, the whole reactor (balance line 2) approximates a CSTR, working at differential conversions with rj 2 according to Equ. 3.90. Writing the balance for line 1, the equation for is... 

In all the studies reported here for CO oxidation, GC analysis of the efQuent stream was carried out using a HayeSep Q column at room temperature (RT) in a GC (Varian 3300) equipped with a TC detector. The conversion data were reproducible within 5% accuracy. Reaction rates were calculated at various temperatures at less than 5% conversion to fulfill differential conversion operation in the 10-well reactor. In the recycle reactor, reaction rates were calculated in the whole conversion range due to the perfect mixing approximation obtained via the use of a large recycle ratio (> 20). Turnover frequencies were calculated from the rate and the dispersion values obtained for the freshly reduced catalysts and plotted versus 1/T to obtain activation energies. 

We have presented earher differential kinetics data for the NO oxidation and standard SCR reactions. Here we report on a similar set of experiments involving a feed with different amounts of NO, NO2, and NH3. Metkar et al. [42] showed for standard SCR that a space velocity of 285,000 h was needed to keep the NO conversion below 15 % in the temperature range of 200-300 °C. In contrast, when the SCR reaction was carried out with an equimolar NO/NO2 feed, a space velocity of 2 x 10 h was needed to ensure differential conversion for temperatures below 245 °C. This high space velocity was achieved by reducing the... 

Figure 6.1.6 Influence of temperature and particle size on pore effectiveness factor (a) 30 bar, particle diameter 2.6 mm [data from Bokhoven and Raayen (1954)] (b) 100 bar, 500°C, differential conversion, feed gas with 4% NH3, H2/N2 = 3 [data from Jennings and Ward (1989)].

---