DE102011076347A1 - System for contactless conductivity measurement of heterogeneous catalysts under reaction conditions - Google Patents

Abstract

A system for measuring the conductivity of heterogeneous catalysts under reaction conditions comprises a resonator (2, 2 ', 20, 20') for microwaves, within which a reactor (3) with a receiving point (4) for a heterogeneous catalyst sample (P) is arranged, wherein the resonator is connected to a microwave source (10) and the reactor can be connected at one end to a source for reaction gases and at the other end to a gas analysis device (GC). A method according to the invention for measuring the conductivity of heterogeneous catalysts under reaction conditions comprises steps in which a catalyst sample is placed in a reactor and this is introduced into a microwave resonator, the catalyst sample is heated to the reaction temperature, the catalyst sample is excited with microwaves, and the resonance spectrum of the catalyst sample is recorded .

Description

The present invention relates to a system for contactless measurement of electrical properties, in particular the complex permittivity and conductivity of heterogeneous catalysts under reaction conditions and the associated method.

State of the art

Electrical conductivity measurements are an important spectroscopic tool in heterogeneous catalysis to study the relationship between catalytic performance and the electronic properties of catalysts. In particular, the rate of reaction for the selective oxidation of hydrocarbons may be limited by electronic and / or ionic transport properties, eg, by the transfer of electrons or holes (in other words, the oxidation or reduction of the transition metal ions of the catalyst), hydrogen or oxygen atoms from the catalyst surface (or even below it) to the substrate (the hydrocarbon) and vice versa. The oxidation of n-butane to maleic anhydride is just one example of a commercially important and already industrial-scale reaction that requires the depletion of eight hydrogen atoms, the incorporation of three oxygen atoms, and the transfer of 14 electrons in a single pass (see for example Millet JMM, Topics on Catalysis 2006, vol. 38, pp. 83 to 92 ) and thus places high demands on the catalyst. If, for example, vanadyl pyrophosphate (VPP) is taken as the oxidation catalyst, current studies assume that the oxygen atoms incorporated into the oxidized product come from the catalyst lattice. Therefore, it has been proposed to describe VPP as a "solid reactant" which can oxidize n-butane to maleic anhydride and must be re-oxidized by molecular oxygen. As a result, this demanding reaction involving adsorption, dissociation (surface reaction), oxygen diffusion, and charge transfer requires a catalyst with good electronic, oxygen, and even proton conductivity. For this reason, it is necessary to precisely determine the electronic and ionic conductivity of the catalysts under reaction conditions (in situ) and to develop an understanding of the influence of these conductivities on the catalytic performance in order to better understand such complex reaction processes.

As noted, in order to determine the actual relationships between the conductivity (or, more generally, any other electronic property of the catalyst) and its catalytic activity and selectivity to a specific product, measurements under reaction conditions must be made. For this purpose, methods known in the art for contact with the catalyst are known in which the conductivity and catalytic power are measured simultaneously under actual operating conditions. That's how it is Hermann, JM et al, Journal of Catalysis 1997, 167, 106-117 a schematic in 1 shown measuring method for determining the electrical conductivity of VPP and VPO catalysts, wherein a powdered catalyst 103 between two platinum electrodes 101 a measuring cell inserted and using a compression spring 102 is compressed to a sufficient electrical contact between the individual particles of the powdered catalyst 103 while ensuring a sufficiently good interaction between the gases (reactants) to be oxidized and the catalyst. The conductivity cell is used to perform the catalytic test in a microreactor 100 flows around the reaction gases under predetermined pressure and temperature conditions and analyzed the reactants and products using gas chromatography. At the same time electrical resistances are recorded by applying a high-frequency AC voltage V _i to the measuring cell. Principle-like measuring methods are also in the already mentioned By Millet and in Sartoni, L. et al., Journal of Materials Chemistry 2006, 16, 4348-4360 described.

From the Article by Rouvet, F. et al., Journal of the Chem. Soc. Faraday Trans., 1994, 90, 1441-1448 a method is known in which the electrical properties of VPP and VPO catalysts are compared with other physicochemical techniques such as in situ Raman spectroscopy or EPR, again for the measurement of conductivity, the heterogeneous catalysts between two platinum electrodes under constant pressure compressed and measuring the temperature and electrical conductivity during various catalytic cycles of the reactant gases.

From the Article by Rihko-Struckmann LK et al, Catalysis Letters, 2006, 109, 89-96 Finally, a comparable method for determining the electrical conductivity of VPO catalysts is known in which first a catalyst precursor powder is pressed into slices and coated by PVD on the two flat surfaces of the slices with thin platinum electrodes. With the aid of electrochemical impedance spectroscopy, the impedance spectra of the VPO disks then become different Measured temperatures and different gas flows in a tubular electrochemical membrane reactor, and the analysis of the gases as in the aforementioned writings by gas chromatography is performed.

The above methods measure the DC or AC resistance up to a frequency of 1 megahertz in samples in contact with two and four electrodes, respectively. While the two-point measurement is widely used in catalysis experiments, it has the serious drawback that it depends very much on the quality of the contact between the electrodes and the catalyst powder, as poor contact and thus high contact resistance significantly distorts the measured data. In addition, especially in the case of the typically used powder catalysts, good and reproducible contact is difficult to achieve, especially if the sample volume and / or geometry changes under the reaction conditions due to sintering, water loss, etc. Four-point measurements could solve this problem in principle, but they are difficult to implement in a catalytic reactor. In addition, electrode polarization would make it difficult to interpret the measured data. Another typical problem with catalysis is the diffusion of catalytically active or inhibiting impurities from the metal electrodes, which are mostly platinum and iron, into the sample.

Presentation of the invention

It is thus an object of the present invention to allow the simultaneous measurement of the conductivity and the catalytic performance of catalysts, in particular semiconducting heterogeneous catalysts under reaction conditions (in situ) at high temperatures, without having to contact the catalyst by attaching electrodes. The reaction conditions are those conditions under which the catalytic reaction shows a detectable efficiency.

To achieve this object, the present invention is based on the insight that the conductivity determination can be measured without contact by excitation of the catalyst by means of microwaves and recording of the corresponding resonance spectrum, such a measurement being easily carried out under reaction conditions. Namely, in polycrystalline low-conductivity semiconductors consisting of conductive and insulating regions, the transport of the carriers is usually limited by the insulating barriers at the grain boundaries of the conductive grains. The DC conductivity is thus determined mainly by the height and width of the barriers at the grain boundaries, but not by the intrinsic electronic properties of the grains per se. In contrast, in AC measurements, and thus in particular microwave measurements, the grain boundaries with increasing frequency are increasingly bridged by capacitive coupling between the conductive grains, so that at high frequencies the actual conductivity of the grains can be determined. Alternatively, charge carrier transport can be described using a hopping mechanism in which the electron conduction is characterized on an atomic scale by an activated hopping of charge carriers between localized states. Both the barrier model and the hopping model describe the frequency dependence of the conductivity at low frequencies with the equation σ _AC (ω) = σ _DC + Aω ^s , (1) where s is in the range of 0 to 2 (normally close to 1), the constant A is slightly dependent on the temperature and ω is the angular velocity. As the conductivity approaches the DC conductivity at low frequencies, and thus the conductivity between the grains, it approaches the bulk conductivity of the conductive grains in the high frequency range. Thus, measurements at microwave frequencies should be ideally suited for determining the intrinsic properties of polycrystalline materials, such as heterogeneous catalysts. Accordingly, it is desirable to measure conductivity over a wide range of frequencies so as to understand the electronic properties of the grains, grain boundaries and polycrystalline semiconductor surfaces. A big advantage of measurements at high frequencies is that the signals can be transmitted wirelessly between transmitter and receiver and are therefore perfectly suited for non-contact methods. In principle, the interaction between radio or microwaves and a material with a measuring setup that uses either a free-field antenna or a resonator ( Chen, L.-F. et al., Microwave Electronics. Measurement and Material Characterization, Wiley 2004, 493-530 ). Since the first-mentioned non-resonant method is suitable only for planar sample geometries, while the resonator technique is characterized by almost no restrictions on the shape of the sample and a very high sensitivity, the present invention focuses on the measurement of the electrical properties of Catalysts using the resonator technique.

While in the so-called "endplate perturbation" the catalyst forms one of the end plates of the resonator, in the present invention it is placed in the center of a cavity of the microwave resonator (usually to the maximum of the electric field, see U.S. Patent Nos. 4,399,459) to avoid contact with the catalyst Chen, L.-F. et al. ). Since it is not possible to replace the Resonatorendplatten with the catalyst in catalytic experiments with heterogeneous catalysts, the inventive system for contactless conductivity measurement of heterogeneous catalysts under reaction conditions based on the so-called Microwave Cavity Enclosed Perturbation Technique (MCPT), in which the catalyst and the surrounding reactor are arranged in the resonator.

The MCPT technique has already been successfully applied over a wide frequency range to conductivity and permittivity measurements of superconductors, superionic conductors, metals, semiconductors and dielectric materials. Also, the oxidation state of automotive three-way catalysts can be qualitatively observed by using the catalyst housing as a microwave resonator and determining its resonant properties in situ ( Moos, R. et al., Topics in Catalysis 2009, 52, 2035-2040 ).

However, it has not hitherto been possible to use the MCPT method for the quantitative determination of the complex permittivity and conductivity of catalysts under operating conditions, that is to say within a reactor at high temperatures in a mixture of reaction gases with simultaneous determination of the reaction products. In fact, there are significant problems in applying the MCPT technique to a catalyst reactor. Namely, while a conventional MCPT structure uses a small sample located in the center of the cavity which induces only a small perturbation of the microwave resonance ratios (ie, a shift of the resonance frequency), the in-situ analysis of a catalyst requires the introduction of a complete one Quartz reactor with the catalytic sample in the cavity. According to the teaching of the prior art of Chen et al. With such a strong perturbation of the electromagnetic field in the resonator, the linear relationship between the properties of the dielectric material and the resonance frequency as well as the quality factor of the filled resonator breaks down.

The theoretical basis for the shift of the resonance frequency of the resonator and thus the determination of the conductivity of the catalyst when introducing a small dielectric material is based on three essential assumptions. On the one hand, for a reliable measurement, the sample must be positioned in a region of uniform field and, on the other hand, the sample must have the shape of an ellipsoid of revolution or another defined geometry. Furthermore, the energy stored in the sample must be small compared to the energy stored in the resonator. In other words, according to the hitherto prevailing teaching of Chen et al. to expect that a complete quartz glass reactor including the catalytic sample in a microwave resonator would result in a significant perturbation of the microwave field and thus the derived approximation for determining the resonance frequency shift is no longer applicable.

A further challenge is the high temperature required for typical oxidation reactions, ie the need to heat the catalyst, whereas the cavity must be kept at room temperature level to avoid aging of the surface of the resonator walls and thus irreversible deterioration of the quality factor of the cavity ,

According to the present invention, there is thus provided a system for measuring the conductivity of heterogeneous catalysts under reaction conditions, comprising a microwave resonator within which is arranged a reactor having a receiving site for the catalyst sample, the resonator being connected to a microwave source and the reactor being connected to a microwave source End connectable to a source of reaction gases and connectable to a gas analyzer at the other end. With this conceivable simple structure, it is possible to excite the heterogeneous catalyst without contact with microwaves whose resonance spectrum can be measured at the resonator. Prior art problems attributable to contact with electrodes, such as high contact resistance, grain boundary configuration, interparticle catalyst spacing, particle size, packing density, anisotropy, etc., which introduce significant non-reproducible errors. are effectively avoided while at the same time enabling high sensitivity. It is also possible to use anisotropic samples and extremely small sample volumes, with a high and reproducible signal at the same time. The heterogeneous catalyst may have a polycrystalline or amorphous structure, and the samples may be in the form of powder or grit as well as shaped articles (eg. Tablets) are used. However, it is also conceivable to investigate monocrystalline catalyst materials.

At the same time, as far as a heatable source of reaction gases is provided, the catalyst sample can be heated by means of the reaction gases, so that only the sample but not the resonator is heated. Alternatively, it is also possible to heat the catalyst sample, rather than by heating the reaction gases, directly through a suitably chosen microwave power input. Since, however, the microwave field acts selectively on the dipoles in the catalyst, this changes the catalyst behavior, which must be taken into account in the evaluation.

The reactor may be a sample tube, such as quartz, in which the receiving site is defined by appropriate means (such as glass wool plug restriction) and the catalyst sample is thereby positioned to lie at the desired position in the resonator. As a reactor but also flat cells (arrays) come into question, which can accommodate not only powdered catalyst samples, but also liquid phases.

The present invention shows that the changes in resonant frequency and quality factor due to perturbation by the sample quartz reactor continue to depend linearly on the electronic properties, that is, it is possible to determine the complex permittivity and conductivity of the sample being tested, and thus quantitative measurements under reaction conditions, that is, in particular, to carry out high temperatures as required for catalysis reactions. It benefits from the fact that the electric field (E-field) is concentrated in the reactor consisting of a dielectric such as quartz and thus amplified in the region of the sample.

The resonator can be designed according to the mode used. Thus, it is preferable that the resonator is a cylindrical X-band TM _nm0 resonator (n = 0, 1, 2, ...; m = 1, 2, ···), and the sample receiving site of the reactor in the middle (along the central axis) of the resonator cavity is arranged. In this case, n = m = 1 is preferred, the sample receiving point thus being located between the local maxima of the E field. The fact that the sample comes to lie between the counter-nodes (maxima) in a local minimum of the E-field, the overall sensitivity of the system is somewhat reduced, but this can be compensated by using larger sample volumes, which also easier to handle and analyze are. Depending on the size (volume) of the cylindrical resonator, other TM modes may be used, e.g. B. for large resonators (r> 5 cm) TM _0n0 modes (n = 1, 2, 3). On the other hand, it is also conceivable to use rectangular resonators, in which instead of TM modes TE modes are used for excitation, in which case the modes in the resonator center E-field minima may have. However, with the same volume, the quality of the rectangular resonator is lower than that of the cylindrical resonator. Also conceivable is the use of split-ring resonators. Although in principle resonators for frequency ranges from 300 MHz to 300 GHz can be used, in practice for the conductivity measurement, the range of 1 to 30 GHz and in particular that of 8-12 GHz (X-band) is preferred. By suitable choice of the resonance frequency, the influence of certain molecular species on the catalyst efficiency can also be investigated, for example the influence of water by measurement at 20 GHz, at which water molecules or OH groups have their absorption maxima, or at 18 GHz for NH ₃ -Groups. By selecting the frequency but also the intrinsic location of the measurement can be varied, at low frequencies, the intergranular properties, ie the electronic properties between individual grains (at the grain boundaries) are measurable, and with increasing frequency, the contribution to the permittivity and thus conductivity within the grains increases. A corresponding adaptation of the resonant frequency by suitable scaling of the resonator and the reactor is also conceivable.

In a preferred embodiment, the system comprises an evacuated jacketed jacket disposed around the reactor and connectable to a vacuum source, and a heater connected to the reactor for introducing fuel gas therein. In this way, the catalyst sample can be heated separately, that is, without having to be heated by the reaction gases, while the evacuated double-walled jacket ensures that only the samples, but not the resonator are heated. The sheath is expediently permeable to microwaves, and consists for example of quartz or sapphire crystal. A general requirement for the sheath is low dielectric losses, in other words the imaginary part ε _{2 of} its complex permittivity must be very small. One possible design of the casing are the arrangements known as Dewar.

For example, klystrons, GaN diodes or vector network analyzers are used as microwave sources, and these can also detect the microwave spectrum of the excited catalyst.

In a preferred embodiment, the resonator comprises a cooling device, which additionally cools the walls of the resonator cavity and keeps it substantially at room temperature. As a result, a degradation of the resonator walls is particularly effectively prevented, which contributes significantly to the functionality and lifetime of the resonator, since in particular the surfaces of the resonator walls have sensitive effects on the microwave field resulting in the resonator. In other words, the quality of the resonator can be kept constant over very long periods, which allows particularly long measurement cycles for determining the electrical catalyst properties.

The cooling device comprises, for example, a cooling water circuit, which optionally dissipates heat from the resonator via additional cooling plates. Alternatively or additionally, however, an air cooling of the resonator, for example using large heat sinks such as copper masses conceivable. To avoid condensation, an internal ventilation of the resonator is possible. However, the cooling device can be used not only for cooling the resonator at high catalyst temperatures, but also to perform measurements at low temperatures, for which purpose the heating device is regulated accordingly. Thus, by suitable combination of heating and cooling devices, any temperature conditions and temporal temperature profiles can be set, under which the catalytic reaction is to be investigated.

Finally, it is preferred that the system according to the invention further comprises a source of reaction gases as well as a gas analyzer, which is connected to one or the other end of the reactor. This analyzer may be, for example, a gas chromatograph, but also a quadrupole mass spectrograph, or simply solutions for the analysis of reactant and product gases.

In a further embodiment, the system according to the invention further comprises a magnetic field generator which generates a magnetic field in the resonator, wherein the resonator has a bimodal cavity, in which incoming and outgoing waveguides are connected to the resonator at an angle of 90 ° to each other and the sample in Reactor in the cavity is arranged so that it can be excited by both modes. Preferably, the reactor is arranged in the resonator cavity at an angle of 45 ° to both waveguides in the plane thereof. With this structure, it is possible to induce the microwave Hall effect in the catalyst sample. In fact, when the magnetic field is switched off, there is no transmission of microwave power through the resonator due to the orthogonal arrangement of the primary and secondary modes, while coupling of the two orthogonal modes is produced when the magnetic field is switched on (the DC magnetic field is perpendicular to the microwave electric field) and hence microwave power the resonator is transmitted. In this way it is possible to determine the charge carrier mobilities in the catalyst sample.

Brief description of the drawings

In the following, the invention will be described with reference to the accompanying drawings, which describe exemplary and non-limiting embodiments.

1 Fig. 12 is a schematic view of a conventional contact-based system for determining the conductivity of a powdery catalyst in a flow-through reactor;

2 FIG. 2 shows two illustrations of a first embodiment of the system of the invention in which the reactor tube is arranged coaxially with the resonator axis, wherein FIG 2a the details of the cooling system and gas connections are shown and 2 B to illustrate the gas flows through the catalytic sample only schematically shows the sample tube in the resonator;

3 is a schematic view of a second embodiment of the invention, in which the sample tube is arranged perpendicular to the resonator axis, wherein here also the microwave source and analysis device are shown;

4 is a schematic view of the electric field distribution of the cylindrical X-band resonator of 3 , in which 4a the field distribution as a function of the angle Φ shows and 4b represents the field distribution in relation to the sample tube and the sample, while in 4c the maximums of the electric field in the resonator are clarified;

5 shows the results of the calibration of the system 3 with different monocrystalline Permittivitätsstandards, wherein 5a the curve for the calibration of ε ₁ shows, and 5b and c show two curves for calibrating ε ₂ ;

6 shows the results of the calibration of the system 3 with different powdered Permittivitätsstandards different packing density δ, where 6a again the ε ₁ calibration and 6b showing ε ₂ calibration;

7 shows the results of the measurement of complex permittivity ( 7a ) as well as the catalytic efficiency ( 7b ) a powdered Nb ⁵⁺ -doped VPO catalyst in situ at reaction conditions;

8th shows the logarithmic representation of the measurement results of the microwave conductivity of the powdered VPO catalyst in situ at reaction conditions plotted against the inverse temperature to determine the dynamic activation energies;

9 Figure 4 is a schematic representation of a rectangular bimodal resonator cavity for determining charge carrier mobility and density according to another embodiment of the system of the invention; and

10 schematically illustrates the microwave Hall effect, of the embodiment of the 9 is being used.

Detailed description

A first embodiment of the system according to the invention for determining the conductivity of a heterogeneous catalyst under reaction conditions is shown in FIG 2a ) and b). The system comprises a cylindrical resonator 2 which comprises two disk-shaped end plates closing the cylinder and forming the resonator cavity 2a . 2 B includes. Along the longitudinal axis of the cylinder is a reactor tube 3 arranged one in the middle between the end plates 2a . 2 B positioned sample receiving point 4 having. In the 2 not shown is that the end plates 2a . 2 B are connected to a microwave waveguide and further with suitable coaxial cables to a microwave source and a frequency analyzer.

The sampling site 4 is limited for example by glass wool plug (not shown) from above and below, on the one hand, the powdery catalyst sample P in the middle of the resonator 2 and on the other hand the passage of reaction gases through the sample tube 3 allow. The sample tube 3 is inside a double-walled quartz dewar 5 , on the one hand, a preheated fuel gas (N ₂ ) is passed around the reactor tube through the inner volume and on the other hand, by connecting to a vacuum pump for thermal insulation of the heated interior of the dewar of the Resonatorkavität. For measuring the reaction temperature is still a thermocouple 6 provided by the sample tube 3 can be introduced to the region of the catalyst sample P.

Since it is advantageous for the function of the system to keep the resonator cavity at a constant temperature and thus to prevent temperature-induced changes in the resonance conditions in the cavity, is a cooling water circuit 7 provided, which in thermal contact with the end plates 2a . 2 B or the resonator housing and dissipates heat emitted by the sample or the heating gas despite Dewar to the resonator (mainly radiatively).

In operation, reaction gases (in 2 B from below) through the sample tube 3 introduced, flow around the preheated to the reaction temperature catalyst P and thereby undergo the catalytic reaction. The product gases as well as remaining reactants are then (in 2 B above) from the tube into a gas chromatograph (not shown) which analyzes the various molecular species, from which the efficiency of the catalytic reaction can then be calculated. At the same time, depending on the temperature of the sample (measured by the thermocouple 6 ) Microwave resonance spectra of the catalyst sample P recorded and related to the efficiency data.

3 schematically shows a second embodiment of the present invention, in which a cylindrical X-band resonator 2 ' with a TM ₁₁₀ cavity and a height of 19.5 mm and a diameter of 38.5 mm is used. Such a resonator is available, for example, from Bruker BioSpin. In the middle between the end plates 2a ' and 2 B' of the cylindrical resonator 2 ' is, this time in the radial direction, a similar to the first embodiment reactor tube 3 through the Resonatorkavität performed, the reactor tube 3 again from a quartz dewar 5 is surrounded, which, as in the first embodiment by corresponding terminals 5a . 5b connected to a vacuum suction (outer dewar volume) and a source of heating gas (inner dewar volume). At the Heizgaszuleitung is accordingly a heat source (oven) 8th is provided, which brings the introduced nitrogen gas N ₂ and thereby held in the nitrogen gas stream sample P to reaction temperature.

The reactor tube 3 has an inner diameter of 3 mm and an outer diameter of 4 mm and contains at its sample receiving point 4 the powdered catalyst sample P, which in turn is held in position on the longitudinal axis of the resonator by two glass wool plugs (not shown). The height of the powdery sample is about 10 mm in this embodiment. The quartz dewar, more specifically its tubular member inserted into the resonator, has an outer diameter of 10 mm and is positioned, along with the sample tube in its interior, perpendicular to the waveguide connected to the resonator cavity.

The reactor tube is upstream of a reaction gas supply port 5c with mass flow controllers (Bronkhorst El-Flow) and downstream to an on-line gas chromatograph GC (Agilent 7890A). For quantitative analysis, the gas stream is split into a first gas stream for analysis of CO ₂ , H ₂ O, n-butane, N ₂ , O ₂ and CO and a second gas stream for analysis of maleic anhydride (product). The separation of the individual gases in the first gas stream is carried out by a Poraplot column (CO ₂ , H ₂ O, C ₄ H ₁₀ ) and a molecular sieve (N ₂ , O ₂ and CO), the measurement using a thermal conductivity sensor. The maleic anhydride in the second gas stream was separated from the reaction mixture by a DB1 column and detected by a flame ionization detector.

The oven 8th is designed as a resistance furnace (Sylvania Tungsten Series I) and provided eight liters per minute preheated nitrogen to heat the catalyst sample P to temperatures between room temperature and 500 ° C. Other measures such as avoiding condensate and evacuating the quartz dewar to 10 ^-7 mbar (using a Pfeiffer HiCube 80 Eco pump) and the previously described water cooling (using a two-circuit cooling system via copper plates attached to the resonator endplates) were also implemented.

The resonator cavity was via an aperture with a waveguide 11 coupled via coaxial cable 10 to a vector network analyzer 9 (VNA, Agilent PNA-L N5230C-225 with operating range between 10 MHz and 20 GHz) is connected to record the resonance spectra of the so-called S11 parameters in reflection mode (reflected power as a function of frequency). The microwave output power was 11 dBm and the resonant frequency was measured directly by determining the frequency at the minimum of the S11 parameter spectrum. The Q factor (Q) was determined by measuring the frequency difference of the resonance peaks at a 3 dB power absorption level (FWHM) and multiplying the determined reciprocal bandwidth by the resonance frequency. The critical coupling of the microwave into the resonator was by phase shifters 12 set and over the Smith chart representation of the VNA 9 controlled. For independent checking of the frequency of the microwave source in the VNA 9 is also a reference resonator 13 to the waveguide 11 connected.

As already mentioned, the MCPT method is based on a slight perturbation of the resonance conditions of a cavity when a sample with a given complex permittivity and / or electrical conductivity is introduced into the cavity. In such a cavity, standing waves can form whose modes are characterized by a characteristic resonance frequency ν ₀ (depending on the geometry of the cavity) and the resonator Q ₀ (depending on the geometry and the conductivity of the resonator). The quality is related to the resonant frequency ν ₀ by the so-called half-width Γ ₀ in combination.

After introducing a small sample into the cavity, the resonance frequency and the quality of the cavity change to the values ν ₁ and Q ₁ , and the associated shift of the values can be related to the permittivity and the volume of the sample and the cavity be set:

wobei A und B Resonatorkonstanten darstellen, die von der Resonatorgeometrie und der Mode abhängen und V_s und V_c das Volumen der Probe beziehungsweise der Kavität darstellen. Zwar können die Konstanten A und B prinzipiell direkt berechnet werden, wenn die Feldverteilung in der Kavität bekannt ist und sich durch die Einführung der Probe nicht wesentlich ändert. Aus praktischen Gründen werden die Resonatorkonstanten jedoch bevorzugt durch Messung geeigneter Kalibrationssubstanzen mit bekannten komplexen Permittivitäten bestimmt, die im Folgenden näher besprochen werden. Zu beachten ist noch, dass die quasi-statische Näherung für isolierende beziehungsweise schlecht leitende (halbleitende) Materialien, wie sie für die vorliegend interessierenden Katalysatoren verwendet werden, eine sehr gute Näherung ist, für metallische Leiter jedoch die Effekte an der Oberfläche im sogenannten „Skin-Depth”-Regime berücksichtigt werden müssen.In the quasi-static approximation, in which the penetration depth of the microwaves into the surface of the sample is substantially greater than the dimensions of the sample, the shift of the frequency and the quality directly with the complex permittivity ε = ε ₁ - iε _{2 of} the sample in Be related.

where A and B represent resonator constants that depend on the resonator geometry and the mode and V _s and V _{c represent} the volume of the sample and the cavity, respectively. Although the constants A and B can be calculated directly in principle, if the field distribution in the cavity is known and does not change significantly by the introduction of the sample. For practical reasons, however, the resonator constants are preferably determined by measuring suitable calibration substances with known complex permittivities, which are discussed in more detail below. It should also be noted that the quasi-static approximation for insulating or poorly conducting (semiconducting) materials, as used for the present catalysts of interest, is a very good approximation, for metallic conductors, however, the effects on the surface in the so-called "Skin -Depth "regime must be considered.

As already mentioned, it has hitherto been assumed that a conductivity measurement in which a sample is introduced into the electric field in a reactor (for example to the counter nodes of an X-band resonator of a TE ₀₁₁ cavity) is difficult if the reactor is not strong can be miniaturized and gas flow rates of milliliters per minute (instead of microliters per minute) can be used. In the present embodiment of the invention, however, it is surprisingly clear that a reproducible and sensitive measurement of the conductivity can also be carried out using a cylindrical resonator with a diameter of 38.5 mm and a height of 19.5 mm working with a TM ₁₁₀ mode and while the aforementioned approximation can be maintained. In the case of the TM ₁₁₀ mode used in the present embodiment, only the longitudinal component E _{z of} the electric field is different from 0 and thereby independent of the resonator length L. The corresponding electric field distribution for a resonator having a radius of R = 2 cm is in 4a (as a function of angle), while from the axial representation of the 4b it can be seen that the sample is placed directly between the counter nodes of the electric field. The counter nodes are in 4c again schematically shown within the resonator in relation to the quartz reactor tube and the sample P.

The catalyst sample P consisted of a powdery heterogeneous Nb-doped vanadyl pyrophosphate catalyst (VNbPO with Nb / V = 0.08), which was milled to a powder of 100 to 200 μm particles. The permittivity standards used were single crystals of sapphire (0001), sapphire (11-20) rutile (001), rutile (100), lanthanum aluminate (100) and strontium titanate (100), available from Crystal GmbH Berlin. The permittivity standards were both ground into powdered samples and cut into 1.2 mm and 5 mm long 3 mm diameter cylinders in their monocrystalline form. The shape of the cut single crystal samples largely corresponds to that of the later analyzed catalyst sample.

The calibration was performed on the single crystals both parallel and perpendicular to their optical axes (c-axes), that is, the optical axes were alternately parallel and perpendicular to the vector of the electric field in the resonator. The ones mentioned above in equations (5) and (6) Frequency and quality shifts were measured and plotted against the known real and imaginary parts of the permittivity. A linear fitting of this calibration is in 5a ) for ε ₁ . It turned out, however, that in particular the quality depends strongly on the sample geometry, which is due to the non-homogeneous fields along the long cylinder axes of the samples. Therefore, two calibration constants B were determined for the differently long cylindrical permittivity standards ( 5b and c) and using these calibration constants, calculate the complex permittivity of each single crystal sample using the resonant frequency and quality shifts actually obtained. The values obtained were compared with values found in the literature and measured under similar conditions (at about 10 GHz and room temperature) and the values given by the manufacturers of single crystals. Thus, it could be found that the MCPT reactor of the system according to the invention actually provides a linear relationship between the complex permittivities and the shift of the resonance frequency and the quality.

wobei ε_s und ε_p die komplexe Permittivität des Einkristalls und des Pulvers bei einer gegebenen Packungsdichte δ darstellen. Mithilfe dieser Gleichung können dann die Gleichungen (2) und (3) wie folgt umgeschrieben werden:

Based on this linear dependence, a calibration curve for powdered samples was then determined. The relationship between the dielectric properties of single crystals and powdered samples was determined using Landau and Lifshitz ( Landau / Lifshitz, Electrodynamics of Continuous Media; Pergamon Press, London, 1960 ) derived formula:

where ε _s and ε _{p represent} the complex permittivity of the single crystal and the powder at a given packing density δ. Using this equation, equations (2) and (3) can then be rewritten as follows:

The validity of these two equations (8) and (9) was also confirmed by measuring different packing densities of the calibration samples and thus a greater range of permittivities. The results are in 6a ) and 6b ) are respectively shown for the shift of the resonance frequency (calibration constant A _p = -0.115 ± 0.005) and the shift of the quality (calibration constant B _p = 0.067 ± 0.003).

With the exception of sapphire, in which in particular the measured imaginary permittivity deviates by up to 50% from the reference value, the permittivities can be determined with a margin of error of 5 to 20%, so that the system according to the invention can in fact be used for the quantitative determination of the complex permittivities of Single crystals as well as powders is suitable.

With the successful calibration, an in-situ measurement of the permittivity and the electrical conductivity of a vanadium-phosphorus-oxide catalyst (VPO catalyst) can now be carried out. VPO catalysts are known to be extremely good industrial catalysts for the selective oxidation of n-butane to maleic anhydride (see, for example Millet, op. Cit. ): C ₄ H ₁₀ + 7 / 2O ₂ → C ₄ H ₂ O ₃ + 4H ₂ O.

The only (unwanted) byproducts that occur in significant amounts are CO and CO ₂ . Usually, the active phase of VPO, viz. Vanadyl pyrophosphate (VPP), is formed only after activation under reaction conditions (for example, 2% n-butane, 20% O ₂ , balance N ₂ at 400 ° C), thus the catalyst material for the present measurement Activated for 60 hours in the gas stream. X-ray diffractometry was used to ensure that the main crystalline phase contained in the sample is VPP.

For the MCPT measurements, 75 mg of a seven-pass fraction of particles of the VPO catalyst of 100 to 200 μm in size were introduced into the quartz reactor tube up to a sample fill height of 10 mm. Then, the temperature dependence of the complex permittivity was measured by heating the catalyst from room temperature to 416 ° C at a heating rate of 10 Kelvin per minute while mixing the sample with 5 ml / min reaction gas mixture (2% n-butane, 20% O ₂ and Residual N ₂ ) was lapped. At the same time, at various temperatures after an isothermal hold time of 10 minutes each, the resonant frequency and quality as well as the catalytic performance were determined by analyzing the effluent gases using a multidimensional online gas chromatograph. The results of this catalytic in situ MCPT / GC assay are in 7a (complex permittivity as a function of temperature) and 7b (Concentration of the reactant and product gases or the butane conversion as a function of temperature) shown.

With increasing temperature both an increase of the real and the imaginary part of the permittivity could be observed ( 7a ). ε ₁ increases only slightly from a value of about 9 at room temperature to 11 at 416 ° C. Therefore, no significant phase changes of the first order can be observed in the temperature range investigated, which was expected in view of the 60-hour activation and preparation of the sample under similar conditions. At ε ₂ , on the other hand, a substantial increase from about 0.2 at room temperature to 0.77 at 416 ° C was observed.

wobei σ₀ eine Konstante ist, E_c die Aktivierungsenergie der elektrischen Leitfähigkeit ist und k_B die Boltzmann-Konstante ist, kann die Aktivierungsenergie E_c aus der Steigung der linearen Regression der 8a berechnet werden (vgl. Herrmann J. M., Catalysis Today, 2006, 112, 73–77 ). Ein linearer Fit im Bereich zwischen 200 bis 416°C gibt ein Aktivierungsenergie E_c von 140 meV (14 kJ/mol). Dieser Wert liegt unterhalb der Gleichstrom-Aktivierungsenergien, die für Vanadiumphosphate (38,1 bis 82,1 kJ/mol, siehe Rouvet, F. et al., op. cit. ) und Vanadylpyrophosphate (73,2 bis 138,1 kJ/mol; siehe Hermann, J. M. et al., Journal of Catalysis 1997, 167, 106–117 ) in der Literatur gemessen wurden. Da Gleichstrommessungen hauptsächlich durch die Berührung zwischen den Körnern bestimmt werden, erscheint es vernünftig, dass die von der Leitfähigkeit innerhalb der Körner bestimmte Mikrowellenaktivierungsenergie niedriger ist.In order to deduce a mechanism for determining the conductivity from the temperature dependence of the data, the logarithm of the microwave conductivity σ · (σ = ωε ₀ ε ₂ ) was plotted as an Arrhenius plot against the reciprocal of the temperature ( 8a ). From this analysis, two areas with different slopes can be identified. By considering the exponential dependence on the reciprocal of the temperature for the electrical conductivity,

where σ _{0 is} a constant, E _{c is} the activation energy of the electrical conductivity and k _{B is} the Boltzmann constant, the activation energy E _c can be calculated from the slope of the linear regression of the 8a be calculated (see. Herrmann JM, Catalysis Today, 2006, 112, 73-77 ). A linear fit in the range of 200 to 416 ° C gives an activation energy E _c of 140 meV (14 kJ / mol). This value is below the dc activation energies found for vanadium phosphates (38.1 to 82.1 kJ / mol, see Rouvet, F. et al., Op. Cit. ) and vanadyl pyrophosphates (73.2 to 138.1 kJ / mol; see Hermann, JM et al., Journal of Catalysis 1997, 167, 106-117 ) were measured in the literature. Since DC measurements are mainly determined by the contact between the grains, it seems reasonable that the microwave activation energy determined by the conductivity within the grains be lower.

wobei D die Diffusionskonstante, c die Konzentration der Ionen mit Ladung z, und e die Ladung des Elektrons ist. Wenn sowohl D als auch c thermisch aktiviert sind, kann die Temperaturabhängigkeit der Leitfähigkeit geschrieben werden als:

wobei E_σ die Aktivierungsenergie für die Ionen-(Protonen-)Leitfähigkeit ist. Die graphische Darstellung von ln(σT) in Funktion von 1/T ergibt wieder zwei Bereiche (8b), wobei nun im unteren Temperaturbereich eine positive Aktivierungsenergie von 50 meV ermittelt wurde, die durch Protonenleitung an der Oberfläche und/oder durch die elektrische Relaxation von an der Oberfläche adsorbiertem Wasser oder Hydroxylgruppen erklärt werden kann. Der Beitrag von physisorbiertem Wasser, Hydroxylgruppen und/oder Protonen zur Mikrowellenadsorption bei niedrigeren Temperaturen wird durch thermogravimetrische/massenspektroskopische Messungen gestützt. Es konnte nämlich von den Erfindern gezeigt werden, dass die untersuchte Probe ungefähr 0,5 Gewichtsprozent Wasser enthält, welches unterhalb von 200°C größtenteils desorbiert ist. Im Bereich hoher Temperatur konnte eine sehr viel höhere Aktivierungsenergie von 190 meV ermittelt werden. In diesem Bereich sollte die Leitung durch Elektronen überwiegen. Wenn die Elektronenleitung durch das Hüpfen von Elektronen beziehungsweise Löchern zwischen den Positionen unterschiedlicher Valenz der Vanadium Atome bedingt ist (hauptsächlich V⁴⁺ und V⁵⁺ in VPO), wie es durch die Theorie kleiner Polaronen beschrieben wird, sollte sich dieselbe lineare Abhängigkeit von in (σT) von 1/T gemäß Gleichung (12) ergeben ( Mott, N. F.; Davis, E. A., Electronic Processes in Non-Crystalline Materials, 2. Auflage, Clarendon Press, Oxford, 1979 ). Die genaue Zuweisung der verschiedenen möglichen Leitungsmechanismen und der Einfluss der Gasphase (hauptsächlich Sauerstoff, n-Butan, Dampf, Wasserstoff) auf die Leitfähigkeit können somit in zukünftigen Studien mithilfe des erfindungsgemäßen Systems und Verfahrens untersucht werden.However, almost no temperature dependence in the range below about 200 ° C was observed in the Arrhenius plot. A similar trend has already been made by Herrmann et al. (op. cit.) observed in the study of the DC conductivity of vanadyl pyrophosphate, namely a weak temperature dependence at low temperatures. The determined "transition temperatures" between the two regimes range from 121 to 348 ° C depending on the pretreatment. Further, it has been reported that vanadium phosphates, such as VOPO ₄ · 2H ₂ O, which can be found as a minor phase in the precursors of our VPO catalyst, is a mixed proton-electron conductor and that protons are the dominant charge carriers at room temperature. The ionic conductivity can be generally described by the Nernst-Einstein equation:

where D is the diffusion constant, c is the concentration of ions with charge z, and e is the charge of the electron. When both D and c are thermally activated, the temperature dependence of the conductivity can be written as:

where E _{σ is} the activation energy for the ion (proton) conductivity. The plot of ln (σT) in function of 1 / T again gives two ranges ( 8b ), wherein now in the lower temperature range, a positive activation energy of 50 meV was determined, which can be explained by proton conduction at the surface and / or by the electrical relaxation of adsorbed on the surface of water or hydroxyl groups. The contribution of physisorbed water, hydroxyl groups and / or protons to microwave adsorption at lower temperatures is supported by thermogravimetric / mass spectroscopic measurements. In fact, it has been shown by the inventors that the sample studied contains about 0.5% by weight of water, which is largely desorbed below 200 ° C. In the high temperature range, a much higher activation energy of 190 meV could be determined. In this area the conduction should predominate by electrons. When the electron conduction is due to the hopping of electrons or holes between the positions of different valences of the vanadium atoms (mainly V ⁴⁺ and V ⁵⁺ in VPO), as described by the theory of small polarons, the same linear dependence of in (σT) of 1 / T according to equation (12) ( Mott, NF; Davis, EA, Electronic Processes in Non-Crystalline Materials, 2nd Ed., Clarendon Press, Oxford, 1979 ). The precise assignment of the various possible conduction mechanisms and the influence of the gas phase (mainly oxygen, n-butane, steam, hydrogen) on the conductivity can thus be investigated in future studies by means of the system and method according to the invention.

In a further embodiment of the invention, which is schematically illustrated in FIGS 9 and 10 As shown, the system of the present invention may also utilize the microwave Hall effect to determine the charge carrier mobilities and densities in the catalyst sample. For this purpose, the system of the invention comprises a magnetic field generator (not shown) and a microwave resonator 20 with bimodal cavity, with the waveguides connected at their ends via iris apertures 21A . 21B are arranged at 90 ° to each other. The resonator can be both rectangular ( 9 ) as well as cylindrical ( 10a , Reference numeral 20 ' ) be configured. The resonators used here have already been used by Na, BK et al., Meas. Sci. Technol. 1991, 2, 770-779 described. The reactor 3 is disposed in the resonator cavity at an angle of 45 ° to both waveguides and is otherwise constructed as in the previously described embodiments. The receiving point with the sample P is at the intersection of the axes of the two mutually perpendicular waveguide connections 21a . 21b positioned. The magnetic field B (DC magnetic field, DC field) generated by the generator is perpendicular to the plane of the resonator (ie the plane formed by the two waveguide connections and the sample). The impedance of the bimodal cavity 20 can by using iris screws 22a . 22b , which the iris apertures 21a . 21b selectively zoom in / out to the waveguides 21A . 21B be adjusted. The resonator cavity also includes brass tuning screws, which, as in 9 are shown shown. Using the screws 23a and 23b It is possible to set the mode that excites the cavity while the resonant frequency of each mode is adjusted using the tuning screws 24a and 24b adjusted and adjusted.

The bimodal cavity is designed with mutually perpendicular primary and secondary modes. This means that when the magnetic field is switched off (B = 0) no microwave power through the resonator 3 should flow. In other words, ideally all microwave power is reflected in the primary mode and at the port 21a detected while the transmitted microwave power of the secondary mode disappears (Anschuss 21b ). However, due to the non-ideal cavity (geometric aberrations, impurities in the wall material, and iris-related aberrations), a low transmitted microwave power in the secondary mode, which is on the order of magnitude comparable to the Hall signal sought, is still measured. For this reason, an extinguishing channel is additionally connected in the measuring system between the two waveguide connections, which at B = 0 downshifts the power transmitted by the secondary mode to almost zero (coupling of the primary and secondary modes of the same amplitude, but with a phase shifter) 180 ° shifted phase).

wobei Q₀ und Q_L die Gütefaktoren des leeren und mit Probe beladenen Resonators sind und B das magnetische Gleichfeld ist.Switching on the DC magnetic field B now leads to easily detectable changes in the microwave power of the secondary mode at the terminal 21b due to the mode locking due to the standing perpendicular to the electric microwave field DC magnetic field. As in 10 is shown, the stationary magnetic field B induced due to the Lorentz force a Hall voltage in the sample material P, which changes its sign according to the oscillating electric field E _x ("driving mode") of the exciting microwaves and so the so-called MHE mode E _H (" Microwave Hall Effect mode ") generated vertically to the excitation mode E _x . The superposition of the two modes causes an ellipsoidal movement of the respective charge carriers, which in 10 is drawn. The Hall mobility of the charge carriers can now be determined as follows from the microwave power P ₁ fed into the primary mode and the (transmitted) microwave power P _H measured in the secondary mode ( Chen, LF et al., Microwave Electronics 2004, John Wiley & Sons ):

where Q ₀ and Q _{L are} the merit factors of the empty and sample laden resonator and B is the DC magnetic field.

Thus, based on the Microwave Cavity Perturbation Technique (MCPT), the present invention provides both a non-contact measurement method for determining conductivity and charge carrier mobility and density, which can be calibrated for testing powder catalysts under absolutely realistic working conditions. The MCPT is based on the adiabatic changes in the properties (resonant frequency, quality) of a microwave resonator (cavity) when the sample is introduced, allowing a direct calculation of the complex permittivity and microwave conductivity of the sample. The system _{according to the} invention comprises a resonator for microwaves, which _operates preferably in the X-band and with TM _nm0 modes (n, m> 0), within which a reactor is arranged with a receiving point for the heterogeneous catalyst sample. The resonator is connected to a microwave source and the reactor can be connected at one end to a source of reaction gases and connected at the other end to a gas analyzer. The device was calibrated with various single crystals and powders of known complex permittivities and the method was successfully used to study the temperature dependence of the complex permittivity and electrical conductivity of a Nb ⁵⁺ -doped vanadyl pyrophosphate catalyst for the selective oxidation of n-butane to maleic anhydride in a reaction gas mixture , As a result, two conductivity regimes with a temperature limit of 200 ° C between the two regimes could be determined, with no or very little temperature dependence at low temperatures and Arrhenius dependence at high temperatures in one range.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited non-patent literature

• Millet JMM, Topics on Catalysis 2006, vol. 38, pp. 83 to 92 [0002]
• Hermann, JM et al, Journal of Catalysis 1997, 167, 106-117 [0003]
• Specialist article by Millet and in Sartoni, L. et al., Journal of Materials Chemistry 2006, 16, 4348-4360 [0003]
• Article by Rouvet, F. et al., Journal of the Chem. Soc. Faraday Trans., 1994, 90, 1441-1448 [0004]
• Article by Rihko-Struckmann LK et al, Catalysis Letters, 2006, 109, 89-96 [0005]
• Chen, L.-F. et al., Microwave Electronics. Measurement and Material Characterization, Wiley 2004, 493-530 [0008]
• Chen, L.-F. et al. [0009]
• Moos, R. et al., Topics in Catalysis 2009, 52, 2035-2040 [0010]
• Chen et al. [0011]
• Chen et al. [0012]
• Landau / Lifshitz, Electrodynamics of Continuous Media; Pergamon Press, London, 1960 [0051]
• Millet, op. Cit. [0054]
• Herrmann JM, Catalysis Today, 2006, 112, 73-77 [0058]
• Rouvet, F. et al., Op. Cit. [0058]
• JM et al., Journal of Catalysis 1997, 167, 106-117 [0058]
• Herrmann et al. (op. cit.) [0059]
• Mott, NF; Davis, EA, Electronic Processes in Non-Crystalline Materials, 2nd Ed., Clarendon Press, Oxford, 1979 [0059]
• BK et al., Meas. Sci. Technol. 1991, 2, 770-779 [0060]
• Chen, LF et al., Microwave Electronics 2004, John Wiley & Sons [0062]

Claims (13)

System for measuring the conductivity of heterogeneous catalysts under reaction conditions, comprising a resonator ( 2 . 2 ' . 20 . 20 ' ) for microwaves within which a reactor ( 3 ) with a receiving location ( 4 ) is arranged for a heterogeneous catalyst sample (P), wherein the resonator with a microwave source ( 10 ) and the reactor is connectable at one end to a source of reaction gases and connectable at the other end to a gas analyzer (GC). System according to claim 1, wherein the resonator ( 2 ' ) is a cylindrical X-band TM _nm0 resonator (n = 0, 1, 2, ..., m = 1, 2, ...), and the sample receiving site ( 4 ) of the reactor ( 3 ) is arranged along the central axis of the resonator cavity, wherein preferably n = m = 1, so that the sample receiving point ( 4 ) is between the local maxima of the E-field. A system according to claim 1 or 2, wherein the system further comprises an evacuatable double-walled jacket ( 5 ) which surround the reactor ( 3 ) is arranged around and is connectable to a vacuum source, and a heating device ( 8th ) which is connected to the reactor and introduces heating gas (N ₂ ) into it. A system according to claim 1, 2 or 3, wherein the resonator comprises a cooling device ( 7 ), which surrounds the walls of the resonator cavity ( 2 . 2 ' . 20 . 20 ' ) and substantially kept at room temperature. A system according to any one of the preceding claims, wherein the system further comprises a source of reaction gases and a gas analyzer (GC), each at one end of the reactor (s) ( 3 ), wherein the gas analysis device is in particular a gas chromatograph (GC) for determining the reactant and product gases. System according to one of the preceding claims, wherein the system further comprises a magnetic field generator, which in the resonator ( 20 . 20 ' ) generates a magnetic field (B), wherein the resonator has a bimodal cavity, in which incoming and outgoing microwave conductors ( 21A . 21B ) are connected at an angle of 90 ° to each other to the resonator and the sample in the reactor ( 3 ) is arranged in the cavity so that it can be excited by both modes. Method for measuring the conductivity of heterogeneous catalysts under reaction conditions, wherein placing a catalyst sample in a reactor and placing it in a microwave resonator, the catalyst sample is heated to reaction temperature, the catalyst sample is excited with microwaves, and the resonance spectrum of the catalyst sample is detected. The method of claim 7 wherein the mode of microwave radiation is a TM _nm0 mode (n = 0, 1, 2, ...; m = 1, 2, ...) and the catalyst sample of the reactor along the central axis of the resonator cavity is preferably, where n = m = 1, so that the catalyst sample is between the local maxima of the E field. A method according to claim 7 or 8, wherein the catalyst sample flows around upon reaching the reaction temperature in the reactor with reaction gases and the reactants and product gases are led out of the reactor and analyzed. A method according to claim 7, 8 or 9, wherein the catalyst sample is heated to reaction temperature by proper selection of the microwave power, by heating the reaction gases or by heating gas from a heater. The method of any one of claims 7 to 10, wherein prior to introduction of the catalyst sample, the reactor is introduced into the microwave resonator empty, heated to reaction temperature, excited with microwaves and the detected resonance signal of the empty reactor is subtracted from the resonance spectrum of the catalyst sample. The method of any of claims 7 to 11, wherein the resonant signal of the catalyst sample is maximized by appropriate phase adjustment of the exciting microwave field. Method according to one of claims 7 to 12, wherein perpendicular to the microwave electric field, a DC magnetic field is applied and the resonance spectra of the catalyst sample are detected at 90 ° to the excitation direction of the microwave field.

Images