Reactor micro-channel

Reactor type Ceramic platelet reactor Micro channel width length (type A + B) 500 pm 25 mm... 

Figure3.67 Steep increaseofpropaneand oxygen conversions with increasing reactorinlettemperature. ( ) Fixed-bed reactor ( ) micro-channel reactor. C3Hg/02/Ne=0.3/0.15/0.55, 50mlmin a,b) ...

Fig. 2 shows a schematic diagram of a micro-channel of reformer section to be examined in this study. A multi-physics computer-aided numerical model framework integrating kinetics, mass transport, and flow dynamics in micro-channel reactors has been established. 

The numbering-up concept demands achieving absolutely uniform flow equi-partition by placing special headers in front of the parallel micro channels [5]. Although this has been solved for tubes in conventional multi-tubular reactors, the expenditure for equipartition in micro channel stacks is assumed to be higher, since the small channels may have more relative differences in structural preci-... 

As examples of micro-channel process intensification and the respective equipment, in particular gas/liquid micro reactors and their application to toluene and various other fluorinations and also to carbon dioxide absorption can be mentioned [5]. Generally, reactions may be amenable to process intensification, when performed via high-temperature, high-pressure, and high-concentration routes and also when using aggressive reactants [5]. 

Micro Total Analysis Systems (pTAS) are chip-based micro-channel systems that serve for complete analytics. The word Total refers to the monolithic system character of the devices, integrating a multitude of miniature functional elements with minimal dead volumes. The main fields of application are related to biology, pharmacology, and analytical chemistry. Detailed applications of pTAS systems are given in Section 1.9.8. Recently, pTAS developments have strongly influenced the performance of organic syntheses by micro flow (see, e.g., [29]). By this, an overlap with the micro-reactor world was made, which probably will increase more and more. 

Other types of non-micro-channel, non-micro-flow micro reactors were used for catalyst development and testing [51, 52]. A computer-based micro-reactor system was described for investigating heterogeneously catalyzed gas-phase reactions [52]. The micro reactor is a Pyrex glass tube of 8 mm inner diameter and can be operated up to 500 °C and 1 bar. The reactor inner volume is 5-10 ml, the loop cycle is 0.9 ml, and the pump volume adds a further 9 ml. The reactor was used for isomerization of neopentane and n-pentane and the hydrogenolysis of isobutane, n-butane, propane, ethane, and methane at Pt with a catalyst. 

The two different concepts are depicted schematically in Figure 1.15. The fixed bed is assumed to have a cross-section Sp and a height Hp and is fiUed with non-deformable spherical particles with diameter dp, where the density of the packing (particles per m ) is denoted by n-p. The micro-channel reactor has a cross-section and a height and comprises channels of diameter d with a specific density (number of channels per m ) n-j. ... 

In order to compare the micro-channel and the fixed-bed reactor, the design and operation parameters should be adjusted in such a way that certain key quantities are the same for both reactors. One of those key quantities is the porosity s, defined as the void fraction in the reactor volume, i.e. the fraction of space which is not occupied by catalyst pellets or channel walls. The second quantity is the specific... 

In order to verify that the fixed bed and the micro-channel reactor are equivalent concerning chemical conversion, an irreversible first-order reaction A —) B with kinetic constant was considered. For simplicity, the reaction was assumed to occur at the channel surface or at the surface of the catalyst pellets, respectively. Diffusive mass transfer to the surface of the catalyst pellets was described by a correlation given by Villermaux [115]. 

For fast reactions Da becomes large. Based on that assumption and standard correlations for mass transfer inside the micro channels, both the model for the micro-channel reactor and the model for the fixed bed can be reformulated in terms of pseudo-homogeneous reaction kinetics. Finally, the concentration profile along the axial direction can be obtained as the solution of an ordinary differential equation. 

In continuous flow systems, the expenditure in mechanical energy necessary to run a process is directly proportional to the pressure drop over the system. Hence the pressure drop is an important figure determining the operating costs of a device. After having verified the chemical equivalence of the two reactor types introduced above, the question arises of whether using a micro-channel reactor instead of a fixed-bed reactor allows a decrease in the pressure drop. In order to estimate the pressure drop in the fixed-bed reactor, the Carman-Kozeney hydraulic diameter model (see, e.g., [116]) was used ... 

For a fixed bed of spherical particles, the Kozeney coefficient lies between 4.5 and 5. Hence the pressure drop in the fixed bed is up to 2.5 times larger than that in the micro-channel reactor, for otherwise comparable conditions. This suggests... 

For axial dispersion in the micro-channel reactor, the usual relationships from Taylor-Aris theory were employed. In order to assess the performance of both reactor types, the widths of two initially delta-like concentration tracers are compared after they have passed through the flow domain. The results of this comparison are displayed in Figure 1.16. 

The figure shows the ratio of the widths of initially delta-like concentration tracers at the reactor exits as a function of the micro-channel Peclet number for different values of the porosity. Taking a value of = 0.4 as standard, it becomes apparent that the dispersion in the micro-channel reactor is smaller than that in the fixed-bed reactor in a Peclet number range from 3 to 100. Minimum dispersion is achieved at a Peclet number of about 14, where the tracer width in the micro-channel reactor is reduced by about 40% compared with its fixed-bed counterpart. Hence the conclusion may be drawn that micro-channel reactors bear the potential of a narrower residence time than fixed-bed reactors, where again it should be stressed that reactors with equivalent chemical conversion were chosen for the comparison. 

Figure 1.16 Ratio of the width of concentration tracers at the exit of the micro-channel reactor to the corresponding quantity in the fixed-bed reactor as a function of the micro-channel Peclet number udf /D, as obtained in [114. ...

In order to show how specific guidelines for the reactor layout can be derived, the maximum allowable micro-channel radius giving a temperature rise of less than 10 K was computed for different values of the adiabatic temperature rise and different reaction times. For this purpose, properties of nitrogen at 300 °C and 1 atm and a Nusselt number of 3.66 were assumed. The Nusselt number is a dimensionless heat transfer coefficient, defined as... 

In order to exemplify the potential of micro-channel reactors for thermal control, consider the oxidation of citraconic anhydride, which, for a specific catalyst material, has a pseudo-homogeneous reaction rate of 1.62 s at a temperature of 300 °C, corresponding to a reaction time-scale of 0.61 s. In a micro channel of 300 pm diameter filled with a mixture composed of N2/02/anhydride (79.9 20 0.1), the characteristic time-scale for heat exchange is 1.4 lO" s. In spite of an adiabatic temperature rise of 60 K related to such a reaction, the temperature increases by less than 0.5 K in the micro channel. Examples such as this show that micro reactors allow one to define temperature conditions very precisely due to fast removal and, in the case of endothermic reactions, addition of heat. On the one hand, this results in an increase in process safety, as discussed above. On the other hand, it allows a better definition of reaction conditions than with macroscopic equipment, thus allowing for a higher selectivity in chemical processes. 

On shrinking the size of micro-channel reactors by reducing the channel dimensions, a number of characteristic quantities such as pressure drop and the degree of chemical conversion are affected. In order to permit a meaningful comparison of the reactor geometry with a scaled geometry, it is important to keep one or a few... 

Table 1.5 Dependence of the number of micro channels N, their length L, the cross-sectional area of the reactor S and the pressure drop AP on the micro-channel diameter, when the efficiency (i.e. a fixed number of transfer units) and at least one specific characteristic quantity are kept fixed in each line. Three cases with operation time-scales varying as (c/m)°. are considered [114],...

Worz et al. give a numerical example to illustrate the much better heat transfer in micro reactors [110-112]. Their treatment referred to the increase in surface area per unit volume, i.e. the specific surface area, which was accompanied by miniaturization. The specific surface area drops by a factor of 30 on changing from a 11 laboratory reactor to a 30 m stirred vessel (Table 1.7). In contrast, this quantity increases by a factor of 3000 if a 30 pm micro channel is used instead. The change in specific surface area is 100 times higher compared with the first example, which refers to a typical change of scale from laboratory to production. 

One of the most interesting theorems of Worz et al. is that they see a serious potential for micro reactors to permit small-scale production of some different sort [110-112]. Micro channels serve as an ultra-precise measuring tool, whereas production is done in channels about 10 to 100 times larger, i.e. miUimeter-sized channels. The limit of tube diameter of industrial production reactors is reported to be 2 cm hence any new reactor of smaller characteristic dimensions bears some potential for improvement. Worz et al. conclude with the remark that the above strategy could be the most important result of their studies [110-112]. 

Calculations predict that improved heat transfer for reacting systems in micro-channel heat-exchanger reactors could lead to considerable size reduction of the equipment, by enhancing the degree of product formation per micro channel (see Figrue 1.24) [140, 141]. This was exemplarily shown for a fast, high-temperature... 

When the space-time yield is referred to the total reactor volume (and not only to the micro-channel volume), the large share of inactive construction material has to be taken into account. Consequently, the space-time yields per micro-channel volume have to differ by orders of magnitude, e.g. more than a factor of 1000, from... 

Keeping in mind the controversial discussion on new physics in micro reactors [198], we certainly have to be at least as careful when introducing or claiming essentially novel chemical processes. A thorough scientific consideration is required for an exact definition and differentiation here that is beyond the scope of this book. So far, no deep-rooted scientific work has been published analyzing the origin of the novelty of chemistry under micro-channel processing conditions. 

A growing number of research groups are active in the field. The activity of reforming catalysts has been improved and a number of test reactors for fuel partial oxidation, reforming, water-gas shift, and selective oxidation reactions were described however, hardly any commercial micro-channel reformers have been reported. Obviously, the developments are still inhibited by a multitude of technical problems, before coming to commercialization. Concerning reformer developments with small-scale, but not micro-channel-based reformers, the first companies have been formed in the meantime (see, e.g., ) and reformers of large capacity for non-stationary household applications are on the market. 

Steinfeldt, N., Buyevskaya, O. V, Wole, D., Baerns, M., Comparative studies of the oxidative dehydrogenation of propane in micro-channels reactor module and fixed-bed reactor, in Spivey, J. J., Iglesia, E., Fleisch, T. H. (Eds.), Stud. Surf. Sci. Catal., pp. 185-190, Elsevier Science, Amsterdam (2001). 

---