EP4648551A1

EP4648551A1 - Efficient induction heating of a reactor

Info

Publication number: EP4648551A1
Application number: EP24174595.9A
Authority: EP
Inventors: designation of the inventor has not yet been filed The
Original assignee: Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek TNO
Current assignee: Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek TNO
Priority date: 2024-05-07
Filing date: 2024-05-07
Publication date: 2025-11-12
Also published as: WO2025233408A1

Abstract

The invention is in the field of inductive heating of reactors and concerns a method for for induction heating of a reactor wherein the reactor comprises one or more heating elements made of a ferromagnetic material and the heating elements are heated above the Curie temperature of the ferromagnetic material. The invention furthermore concerns a method for carrying out an endothermic reaction in a reactor and a method for desorbing an adsorbate from a sorbent in a reactor making use of the method for induction heating.

Description

Introduction

Industrial scale chemical processes are performed in reactors that often require heating. Resistive heating is the process of generating heat in an electrically conductive materials by passing an electric current through the material, wherein the electric current is directly applied to the conductive material. Reactor systems wherein the catalyst is heated using resistive heating are known, for example from WO2019228798 . WO2019228798 describes a structured catalyst comprising a 3D macroscopic structure supporting a ceramic catalyst coating. An electric current is passed through the macroscopic structure, thereby heating the structure by resistive heating and indirectly the catalyst via thermal contact with the structure.
Induction heating (or induction heating) is also a process of generating heat in an electrically conductive materials by passing an induced electric current (or secondary current) through the material (i.e. eddy energy losses). However, induction heating is different from resistive heating in that the current is generated by a rapidly alternating magnetic field penetrating the material (electromagnetic induction). The electric currents thus generated are called eddy currents. A rapidly alternating magnetic field is generated by an induction coil that is typically wound around the reactor. Induction heating may additionally occur by hysteresis effects induced in the electrically conductive elements. Hysteresis occurs in ferromagnetic materials below their Curie temperature. Typically, the heating by hysteresis is relatively small compared to the heating by eddy currents.
Benefits of induction heating compared to other means of heating are numerous. First, the heat is generated inside the reactor which compared to surface heating methods (by steam or electrical) provides for a much more even and rapid heating of the reactor and allowing for larger reactor diameters. Induction heating prevents an unnecessary high heat load as the heat is generated in the area where it is needed. Consequently, catalyst and sorbent lifetime increase, and fouling of catalysts is less. Also, by targeting the heat, the energy consumption is relatively low. In addition, as the heating is by an alternating magnetic field no connections are needed protruding the reactor wall. Moreover, induction heating is possible at modest temperatures but also for very high temperatures. At last, stopping the heating is also instantaneous by turning of the power which allows for a safer reactor operation.
Ideally, the relevant material (catalyst, sorbent etc.) inside the reactor would be made of material susceptible to induction heating, such that it can be targeted directly by induction heating. However, from a materials perspective this is difficult to achieve. Therefore, the relevant material is typically contacted with a structure susceptible to induction heating, for example the wall of the reactor, rods protruding the reactor, or the material is coated on particles that are susceptible to induction heating. The heat is then transferred to the relevant material via thermal conduction.
Latifi et al. (AICHE Journal, vol 61, no 5 p1507-153, 2015) discloses a reactor comprising metal rods for induction heating of a fluidized bed. EP 0363066 A2 discloses a reactor comprising an inductively heatable, fluid permeable reactor element; a monolith. KR 102063414 B1 concerns a reactor stacked with catalytic pellets that may be susceptible to induction heating. In addition, a central metal rod may be present that may be heated by induction heating.
In case ferromagnetic materials are used for induction heating of a reactor, it is often believed that the reactor temperature should be below the Curie temperature of the ferromagnetic material ( WO 2017/186608 , US 2018/0243711 ) as above this temperature magnetic hysteresis losses do not contribute to the heating which would make the heating less efficient.
Whilst induction heating of reactors has many benefits, there is the need to further optimize induction heating of reactors. For example, there is the need for increasing the efficiency of induction heating of a reactor, such as lowering the energy usage of the induction heating of a reactor.

Summary of the invention

The inventors surprisingly found that the induction heating of a reactor could be improved by making use of heating elements made of a ferromagnetic material and heating these elements above the Curie temperature of those elements. Increasing the power when the heating elements had a temperature just below the Curie temperature resulted in a surprising increase in temperature. In figure 1b this increase was observed when increasing the current from 68A to 77A, in figure 2a this increase was observed when increasing the current from 52A to 53A, in figure 3a this increase was observed when increasing the current from 18A to 19A. By keeping the temperature of the heating elements that are comprised in the reactor above this temperature, a process carried out in a reactor is more efficient, as at a relatively low power input a high temperature can be maintained. The power per degree Celsius needed above the Curie temperature is surprisingly lower above the Curie temperature. An example of a reaction that is carried out in a reactor that is heated by induction heating is in figure 5, in which a reverse water gas shift reaction is carried out at different power inputs and the CO₂ conversion is measured. Clearly when the temperature is increased above the Curie temperature of the heating elements (at about minute 60), the conversion suddenly rises. This further supports the surprising effect that above the Curie temperature of heating elements made of a ferromagnetic material the heating is more efficient, which results in a higher temperature of the heating elements and therewith a higher temperature in the reactor, which results in a higher conversion.
Interestingly, the inventors further found that once the temperature of the heating elements made of a ferromagnetic material was above the Curie temperature, this more efficient heating was maintained, even when lowering the power input to below the power input needed to reach the Curie temperature. It was surprisingly found that for a same power input, a much higher temperature could be maintained once the heating elements made of a ferromagnetic material had experienced the Curie temperature and were kept above this much higher temperature. In the examples this was demonstrated by first increasing the power in steps until the temperature of the heating elements surpasses the Curie temperature, and subsequently reducing the power in steps until the temperature dropped below the Curie temperature.
It is known that physical properties of a ferromagnetic material change considerably when passing the Curie temperature. The specific heat capacity C_p is an example of such property. The inventors believe that higher heating efficiency above the Curie temperature relates to this change in properties. The examples demonstrate that for a non-ferromagnetic material such a copper, which does not have a Curie temperature there is no temperature at which a sudden improvement in heating efficiency occurs. In addition since the phase transition only reverts when the temperature of the ferromagnetic material drops below the Curie temperature, a working area is present in which heating is much more efficient than not having reached the Curie temperature.
The current invention opens the possibility to select a specific ferromagnetic material for the elements depending on the desired temperature in the reactor. In case material is selected that has a Curie temperature below the desired temperature, a more efficient heating can be realized as the process in the reactor is carried out above the Curie temperature of the ferromagnetic material.

Brief description of the drawings

Figure 1a depicts a heating element in the form of an iron mesh. The white areas are open areas, whilst the black areas are iron. The diameter of the mesh is 1.6 cm. Figure 1b shows the current in the heating coil of the example and the resulting temperature of the heating element which is the mesh of figure 1a. The current is altered over time. The temperature in °C is on the primary y-axis on the left side of the figure, the current in ampere is on the secondary y-axis on the right side of the figure. The x-axis is the time in seconds. The dashed line is the current, the solid line is the temperature.
Figure 2a shows the current in the heating coil of the example and the resulting temperature of the heating element which is an iron disk. The current is altered over time. First the current is stepwise increased (which is the ascending cycle), subsequently the current is stepwise decreased (which is the descending cycle). The temperature in °C is on the primary y-axis on the left side of the figure, the current in ampere is on the secondary y-axis on the right side of the figure. The x-axis is the time in minutes. The dashed lower line is the current, the solid upper line is the temperature. Figure 2b is a different representation of figure 2a. The temperature in °C is on the y-axis whilst the current in ampere is on the x-axis. The open circles concern the ascending cycle whilst the closed circles concern the descending cycle.
Figure 3a shows the current in the heating coil of the example and the resulting temperature of the heating element which is a nickel disk. The current is altered over time. First the current is stepwise increased (which is the ascending cycle), subsequently the current is stepwise decreased (which is the descending cycle). The temperature in °C is on the primary y-axis on the left side of the figure, the current in ampere is on the secondary y-axis on the right side of the figure. The x-axis is the time in minutes. The dotted line is the current, the solid line is the temperature. Figure 3b is a different representation of figure 3a. The temperature in °C is on the y-axis whilst the current in ampere is on the x-axis. The open squares concern the ascending cycle whilst the closed circles concern the descending cycle.
Figure 4 shows the current in the heating coil of the example and the resulting temperature of the heating element which is an iron disk. The current is altered over time. First the current is set so that the temperature of the heating element is below the Curie temperature. Subsequently the current is increased so that the temperature surpasses the Curie temperature of the heating element. In a next step the current is reduced to the previous current and kept at this current for about 10 hours. During these 10 hours the temperature remains constant above the Curie temperature of the heating element. The temperature in °C is on the primary y-axis on the left side of the figure, the current in ampere is on the secondary y-axis on the right side of the figure. The x-axis is the time in hours. The dotted line is the current, the solid line is the temperature.
Figure 5 show the CO₂ conversion of a RWGS reaction carried out in a reactor that is heated by induction heating. The power provided to the heating coil is altered over time. First the power is stepwise increased (which is the ascending cycle), subsequently the power is stepwise decreased (which is the descending cycle). The conversion in percentage is on the primary y-axis on the left side of the figure, the power in Watt is on the secondary y-axis on the right side of the figure. The x-axis is the time in minutes. The dotted line is the power, the solid line is the conversion.

Detailed Description

In a first embodiment the invention concerns a method for induction of a reactor wherein the reactor comprises one or more heating elements made of a ferromagnetic material, wherein

in a first step the one or more heating elements are heated by induction heating to a first temperature above the Curie temperature of the ferromagnetic material, and subsequently the temperature of the one or more heating elements is kept constant at this first temperature; or wherein
in a first step the one or more heating elements are heated by induction heating to a first temperature above the Curie temperature of the ferromagnetic material, and in a second step the temperature of the one of more heating elements is lowered to a second temperature by lowering the induction heating power, and subsequently the temperature of the one or more heating elements is kept constant at this second temperature, wherein the second temperature is below the first temperature but above the Curie temperature of the ferromagnetic material.

Furthermore the invention concerns a method for carrying out an endothermic reaction in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to the invention wherein a feed provided to the reactor is converted to a reaction product that exits the reactor, preferably wherein the reactor comprises a catalyst in the form of a particulate solid and/or wherein the heating elements are coated with a catalyst.
The invention also concerns a method for desorbing an adsorbate from a sorbent in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to the invention, wherein the reactor comprises a sorbent preferably in the form of a particulate solid and wherein the adsorbate that is desorbed from the sorbent exits the reactor, preferably wherein a feed is provided in the form of a carrier gas.

Induction heating

Induction heating of reactors is known in the art. Induction heating is a process that uses electromagnetic induction to heat electrically conductive materials without direct contact. Typically, the reactor is part of a reactor system comprising the reactor and an induction heating coil surrounding the reactor and a power source suitable for supplying an alternating current to the induction heating coil. The reactor comprises elements susceptible for induction heating. These are heating elements made of electrically conductive material and/or made of a ferromagnetic material. Elements made of a ferromagnetic material are electrically conductive. The alternating current that is generated by the power source goes through the induction heating coil which subsequently generates an alternating magnetic field in the reactor. The alternating magnetic field subsequently induce eddy currents in the heating elements. These currents lead to energy dissipation and heating up of the elements. The heat of the elements is subsequently transferred to the rest of the reactor. In addition, when the heating elements are ferromagnetic and below their Curie temperature, a second heating mechanism occurs, which is caused by hysteresis losses in the ferromagnetic material. Hysteresis heating is known in the art.
Ferromagnetic materials are known in the art. Ferromagnetic materials are magnetic in the absence of an applied magnetic field. Ferromagnetic materials are characterized in that have a relative magnetic permeability µ_r higher than 0.5, preferably higher than 1 more preferably higher than 10 for temperatures below their Curie temperature, preferably for temperatures above 0°C and below their Curie temperature. Preferred ferromagnetic materials are selected from the list of iron, iron alloys, nickel, nickel alloys, cobalt, cobalt alloys, manganese, manganese alloys, bismuth, bismuth alloys, antimony, antimony alloys, neodymium, neodymium alloys and mixtures thereof, more preferably selected from the list of iron, iron alloys, nickel, nickel alloys, cobalt, cobalt alloys and mixtures thereof. A preferred iron alloy is steel. Ferromagnetic materials are electrically conductive. The electrical resistivity ρ (Ω·m) is the direct inverse of the electrical conductivity σ (S/m). Consequently, the electrical resistivity of a material is a relevant quantity for the electrical conductivity of that material. The higher the resistivity, the lower the conductivity. In the context of induction heating, so for the purpose of inductively heating of reactors a material is electrically conductive when having an electrical resistivity below 1000 µΩ·m, preferably below 100 µΩ·m, more preferably below 10 µΩ·m.
The Curie temperature T_C is the temperature above which a ferromagnetic material does no longer exhibit ferromagnetic behaviour. The Curie temperature of iron is 770 °C, Curie temperature of cobalt is 1130 °C, the Curie temperature of nickel is 354 °C. For alloys the Curie temperature depends on the composition of the alloy. Curie temperatures of materials are known in the art. The Curie temperature of a material may be measured by a thermogravimetric analysis (TGA) whilst applying a constant magnetic field to that material. In more detail, a constant magnetic field is applied to a sample while heating the sample and whilst the mass and temperature of the sample are measured. The Curie temperature is the temperature at which the mass of the sample suddenly drops or increases. This drop or increase is due to that the magnetic forces exercised on the sample disappears at the Curie temperatures and therewith the force exercised on the microbalance changes. The magnetic forces exercised on the sample disappears because the ferromagnetic properties disappear at the Curie temperature. Such method is explained by Hasier et al. in EPJ Techniques and Instrumentation (2017) 4:5; DOI 10.1140/epjti/s40485-017-0040-y.
Above the Curie temperature eddy currents and related energy dissipation still is present as the material remains electrically conductive above this temperature. Therefore, induction heating for ferromagnetic materials also occurs above the Curie temperature.
In some cases, current cancelation occurs when heating by generation of eddy currents in the heating elements. Typically, the eddy currents concentrate on the surface of the electrically conductive element having a diameter d and only penetrates a limited depth δ. The phenomenon of current cancellation may become relevant at specific ratios of d over δ. When the radius r (r = 0.5·d) and the penetration dept are about equal, induced eddy currents circulating on opposite sides of the element may cancel each other. Due to this current cancellation less or no heat is generated. If for a given element diameter d the penetration depth δ increases or if for a given penetration depth δ the element diameter d decreases current cancellation may become relevant. Significant current cancellation takes place below a ratio of diameter over penetration depth of 3.2 (d/δ < 3.2).
The penetration depth δ is a function of the electrical resistivity of the element (ρ) the relative magnetic permeability of the element (µr) and the alternation frequency (F) of the heating coil F: δ (m) = 503 · [ρ - µ_r ^-1 - F^-1]^0.5. In the state of the art of induction heating, for a given material for a element (having a specific ρ and µ_r at a given reactor temperature) and having a specific d of the element, an appropriate F is selected for which no or limited current cancellation takes place, so for which the heating efficiency is high. The penetration depth δ relates to the frequency F wherein δ ~ F^--0.5. So, a higher frequency gives a smaller δ. To make efficient use of induction heating the skilled person normally selects a frequency so that d/δ > 3.2. As typically the frequency is adjustable, the skilled person may try different frequencies and select a frequency that provides for efficient heating and for the targeted temperature. In a preferred embodiment of the invention, no eddy current cancelation takes place in the heating elements. Preferably for the invention, the coil efficiency η_el is higher than 0.6, preferably higher than 0.8, and/or the power transmission factor F_TRANS is higher than 0.6, preferably higher than 0.8. The coil efficiency η_el represents the ratio of the power induced in the element P_W, to the total of P_W and electrical losses P_{EL, LOSS}. The power transmission factor concerns the power that is transferred from the coil to the electrically conductive elements in the reactor. For a F_TRANS of 1 all power is transferred to the elements, for a F_TRANS of 0 no power is transferred to the elements. Preferably the alternation frequency is in the range of 10 Hz-4 MHz, more preferably 1 kHz - 3 MHz, even more preferably 10 kHz - 3 MHz, even more preferably 100 kHz - 3 MHz, even more preferably 500 kHz - 3 MHz, most preferably 1 MHz - 3 MHz. Preferably for the invention a frequency and heating element diameter is selected so that d/δ is larger than 3.2, preferably larger than 4.0.

Heating elements

When in this document heating elements are referred to, these heating elements concern heating elements made of a ferromagnetic material. When in this document ferromagnetic material is referred to, this concerns the ferromagnetic material of the heating elements. Different shapes of heating elements are known in the art. The skilled person is routinely capable of selecting a shape that is suitable for a process. Aspects like the particle size of particulate solids present in the reactor such as catalysts, sorbents or solid reactants, the reactor diameter, the required maximum temperature gradients in the reactor and the targeted temperature affect the selection of the shape. Preferably shape, diameter and spatial distribution of the heating elements is selected such that the heating elements evenly heat the reactor. The concentration of heating elements may differ in the radial direction, to account for shielding effects induced by the elements placed near the wall of the reactor. In a preferred embodiment the heating elements are coated with a catalyst. In this way the heat generated may be directly transferred to the reaction surface. In a preferred embodiment the heating elements are placed on a carrier (for example as demonstrated by C. Scarfiello, et al., International Journal of Hydrogen Energy, Volume 46, Issue 1, 2021, Pages 134-145 in which Ni₆₀Co₄₀ particles are place on a γ-Al₂O₃ carrier). Preferably, the heating elements are coated with a catalyst and/or the heating elements are placed on a carrier.
In a preferred embodiment the heating elements are selected from the list consisting of particles, wires, rings, and any combination thereof. Preferably at least a part of the wires form closed loops of connected wires. A wire is also named a strand. Wires are defined by a length and a diameter. The length of a wire is at least a factor 10 larger than its diameter, preferably a factor 20, more preferably a factor 50. Relatively thick wires, typically with a diameter larger than 0.5 cm are sometimes named rods. Wires typically have a circular cross-section, although oval or polygonal cross-sections are also known. Particles are known in the art and may have different shapes. Typically for a particle the maximum diameter of the particle is less than a factor 10 larger than the minimum diameter of the particle, preferably less than a factor 5. In a preferred embodiment of the invention the particles are selected from the group of spheres, discs, solid cylinders, annular pellets, trilobal pellets, irregularly shaped pellets, and any combination thereof. The maximum diameter of a particle typically is a factor 20 smaller than the reactor inner diameter, preferably a factor 40.
The diameter which is relevant for current cancelation is dependent on the type of element. In the current document when referring to the diameter of an element the relevant diameter of that element is referred to. For a wire, the relevant diameter is the diameter of the wire. For a sphere, the relevant diameter is the outer diameter of the sphere. For a disc, the relevant diameter is the outer diameter of the disc. For a solid cylinder, the relevant diameter is the outer diameter of the solid cylinder. For an annular pellet, the relevant diameter is the outer diameter of the annular pellet. For a trilobal pellet, the relevant diameter is the outer diameter of the lobes of the trilobal pellet. For an irregularly shaped pellet, the relevant diameter is the largest outer diameter of the irregularly shaped pellet.
One form of heating elements is closed loops formed by connected wires. Examples of such structure is chicken wire or in figure 1a. The wires are electrically connected, so a closed loop forms a circuit. Preferred closed loops and benefits of closed loops are described in WO 2023/126484 . Since in such closed loop an additional current is present (a current that circulates over the whole loop) the closed loops provide for efficient heating. Moreover, thermally conductive wires may be attached to the closed loops or may form part of the 2-demensional or 3-dimensional network so that the generated heat is distributed even better in the reactor by conduction through the thermally conductive wires.
The closed loops may be part of a 2-dimensional network of connected wires or a 3-dimensional network of connected wires. In a preferred embodiment the closed loops form part of a 2-dimensional network of connected wires or a 3-dimensional network of connected wires. In a preferred embodiment of the invention the reactor comprises one or more 2-dimensional networks of connected wires and/or one or more of a 3-dimensional network of connected wires, preferably comprising closed loops formed by connected wires. The one or more 2-dimensional or one or more 3-dimensional networks preferably are distributed in such a way in the reactor that the reactor is evenly heated.
Evenly heated in the context of the current invention is that the temperature difference in the axial and/or radial direction is less than 100°C, preferably less than 50°C, more preferably less than 20°C. Preferably evenly heated for a tubular reactor is that the temperature difference in the radial direction is less than 100°C, preferably less than 50°C, more preferably less than 20°C. For a closed loop, the diameter of the loop is the relevant diameter.

The reactor

In a preferred embodiment the reactor of the invention is a tubular reactor. A tubular reactor takes the form of a cylindrical column having a reactor diameter and a reactor length. The reactor length is in the axial direction of the column. The reactor has two ends or opposite sides at the opposing sides of the column. The distance between these opposite sides is the length of the reactor. In view of the improved heating according to the invention, the size of the reactor is not limited. The invention can deal with reactor diameters up to 4 meter or even bigger. In a preferred embodiment, the tubular reactor has a diameter up to 4 meter, more preferably the tubular reactor has a diameter in the range of 0.01 m - 4 m, even more preferably in the range of 0.1 m - 2m. The heating elements are placed in the interior of the reactor. In a preferred embodiment the volume of the total of the heating elements is less than 40 vol.% of the total empty reactor volume, preferably less than 20 vol.%. In case the volume of heating elements is too high, the elements placed near the wall of the reactor may shield the heating elements placed near the centre of the reactor, leading to less heating of the centre of the reactor.
The reactor has an interior wherein a chemical reaction, physical reaction and/or separation process may occur. In a preferred embodiment the reactor is for a chemical reaction and/or a sorption process. The skilled person understands that a sorption process involves adsorption and desorption. Typically, first adsorption takes place and subsequently desorption. As will be understood by the skilled person, the reactor typically has one or more inlets for receiving a feed and one or more outlets for discharging a product. The interior of reactor is heated by the induction heating, which may be in view of the endothermic nature of the chemical reaction and/or sorption process, but could also be for other reasons, such as catalyst activation, start-up of a non-endothermic reaction (exothermic or autothermal reaction), sorbent activation. Preferable processes to be carried out in the tubular reactor of the invention are the reverse water-gas shift reaction, steam reforming or dry reforming (of hydrocarbons and/or oxygenates), pyrolysis, calcination reactions, dehydrogenation of hydrocarbons, evaporation, temperature-swing adsorption, and pressure-swing adsorption, or combinations thereof.
A preferred reaction type that is to be performed in the reactor is sorption-enhanced reactions, wherein one product of the chemical reaction is removed from the equilibrium mixture by sorption which consequently pushes the equilibrium towards the product side.
Preferably the reactor comprises one or more solids. These one or more solids need to be heated for proper operation. The solid or solids are placed in the interior of the reactor. The solid may be a catalyst, a sorbent, a reactant, or a combination of two or all three thereof. The material of the solid is not susceptible to induction heating itself, but to heating by thermal conduction. In a preferred embodiment, the solid is a catalyst and/or a sorbent. Preferably, the solid is present in the form of particles, so as a particulate solid, such as in fixed bed reactors, moving bed reactors, rotating bed reactors, fluidized bed reactors or slurry reactors. Preferably the sorbent is for adsorbing H₂O, CO₂ and/or impurities, preferably for adsorbing H₂O or CO₂, more preferably the sorbent is selected from the list consisting of zeolites, activated carbon, metal organic frameworks (MOF), carbon molecular sieves (CMS).
Preferably the tubular reactor is a fixed bed reactor, a moving bed reactor, a rotating bed reactor, a fluidized bed reactor or a slurry reactor, more preferably the tubular reactor is a fixed bed reactor or a fluidized bed reactor, most preferably the reactor is a fixed bed reactor.
As such, the solid material is perfectly compatible with the heating elements that are also comprised in the reactor. The heating elements are placed in the interior of the reactor. The material of the solid is preferably not electromagnetically shielding, such as may be the case for electrically conducting materials. Thus, in a preferred embodiment, the solid is not electrically conducting. Preferable solid materials include heterogeneous catalysts and sorbents such as metal chlorides, metal oxides, zeolites (aluminosilicates), silica gel, amorphous carbon, polymers, or mixtures thereof, including materials that are partially composed thereof.

Method for induction heating

In a preferred embodiment of the invention for the method form induction heating of a reactor the reactor is part of a reactor system comprising the reactor and an induction heating coil surrounding the reactor and a power source suitable for supplying an alternating current to the induction heating coil. For this embodiment, the one or more heating elements are heated by supplying an alternating current to the induction heating coil, and the temperature may be lowered by reducing the power provided to the induction heating coil, preferably by reducing the current through the induction heating coil. The power P expressed in Watt (W) is the product of the current I expressed in ampere (A) and the voltage (V); P = I.V.
The heating elements are heated by induction heating to a first temperature above the Curie temperature. The inventors surprisingly found that by increasing the temperature of the heating elements above the Curie temperature, the induction heating is more efficient. Only a relatively small increase in current resulted in an unexpected large increase of the temperature of the heating elements. In a preferred embodiment the first temperature is 10 - 500 °C, more preferably 15 - 400 °C, even more preferably 20 - 300 °C above the Curie temperature of the ferromagnetic material and/or at least a factor 1.10 above the Curie temperature of the ferromagnetic material, preferably at least a factor 1.15, more preferably at least a factor 1.20, wherein the Curie temperature is expressed in degrees Celsius. Preferably the first temperature is at least a factor 1.10 above the Curie temperature of the ferromagnetic material, preferably at least a factor 1.15, more preferably at least a factor 1.20, wherein the Curie temperature is expressed in degrees Celsius. When is this document a factor above the temperature is mentioned, the factor is relative to that temperature, for example when a factor above the Curie temperature is mentioned, the factor is relative to the Curie temperature. For a temperature that is factor 1.10 above the Curie temperature, this temperature is 1.10 times the Curie temperature expressed in degrees Celsius. For example, for a Curie temperature of 770°C a temperature that is factor 1.10 above the Curie temperature amounts to a temperature of 1.10 · 770°C = 847°C.
The inventor found that surpassing the Curie temperature by increasing the induction heating power slightly results in a sudden increase in temperature. In practice a temperature that is near the Curie temperature of the heating elements is therefore difficult or even impossible to reach. In case the required temperature in the reactor is well above the Curie temperature of the heating elements this does not constrain the practical use of this more efficient heating. However, to work close to the Curie temperature of the heating elements, so to have a second temperature relatively close to but above the Curie temperature of the heating elements, it may be necessary to first surpass the Curie temperature and subsequently reduce the induction heating power to come to a second temperature of the heating elements that is below the first temperature but above the Curie temperature. Surprisingly, the power required to maintain this second temperature which is above the Curie temperature is considerably lower than the power needed to heat the heating element to a temperature below the Curie temperature. Therefore, with a surprisingly relatively low heating power a high temperature can be maintained. Preferably for the method of the invention the heating elements have a temperature in the range of 200 - 1200°C, more preferably in the range of 300 - 1000°C, most preferably in the range of 350 - 900°C.
In a preferred embodiment the second temperature of the heating elements is 25 - 300 °C, more preferably 15 - 200 °C, even more preferably 10 - 100 °C above the Curie temperature of the ferromagnetic material and/or the second temperature is less than a factor 1.20 above the Curie temperature of the ferromagnetic material, more preferably less than 1.15, even more preferably less than 1.10 wherein the Curie temperature is expressed in degrees Celsius. Preferably the second temperature is less than a factor 1.20 above the Curie temperature of the ferromagnetic material, more preferably less than 1.15, even more preferably less than 1.10 wherein the Curie temperature is expressed in degrees Celsius. In a preferred embodiment the second temperature of the heating elements equals the first temperature or is lower than the first temperature. Preferably the second temperature of the heating elements is at least a factor 0.01 lower than the first temperature wherein the first temperature is expressed in degrees Celsius, more preferably at least a factor 0.02, even more preferably at least a factor 0.05, even more preferably at least a factor 0.07 wherein the first temperature is expressed in degrees Celsius. Preferably the second temperature is at least 5°C lower than the first temperature, more preferably at least 10°C, even more preferably at least 20 °C. For all embodiments of the invention the second temperature of the heating elements is above the Curie temperature of the heating elements.
Typically, for its operation a reactor is kept at a constant temperature, so the heating elements are kept at a constant temperature. In a preferred embodiment of the invention the temperature of the heating elements is kept constant for at least 10 seconds, more preferably for at least 100 seconds, even more preferably for at least 1000 seconds. For the method of induction heating of a reactor of the invention, the temperature of the heating elements is kept constant at the first or second temperature, preferably the temperature is kept constant at the first temperature or at the second temperature for at least 10 seconds, more preferably for at least 100 seconds, even more preferably for at least 1000 seconds. Typically to keep the temperature of the heating elements constant the induction heating power is kept constant. In a preferred embodiment the induction heating power is kept constant for at least 10 seconds, more preferably for at least 100 seconds, even more preferably for at least 1000 seconds. The induction heating power is the power provided to the induction heating coil. A constant temperature is a temperature that remains within a bandwidth of 30°C during a period, preferably within a bandwidth of 10°C. A constant power is a power that remains within a bandwidth of 10% of the average power during a period, preferably within a bandwidth of 5%.
The ferromagnetic material of which the heating elements are made of may be selected based on the process that is carried out in the reactor. In this way an optimal efficiency may be obtained. Several aspects are of importance. One of these aspects is the temperature at which the process is carried out. The reactor is operated at the required temperature of the process that is carried out in the reactor. Typically, this operation temperature is the same as the second temperature of the heating elements. In a preferred embodiment the ferromagnetic material has a Curie temperature in the range of 25 - 300 °C, preferably 15 - 200 °C, even more preferably 10 - 100 °C below the required temperature of the process that is carried out in the reactor. Preferably the ferromagnetic material has a Curie temperature that is a factor in the range of 0.83 - 1.0 of the required temperature of the process that is carried out in the reactor wherein the Curie temperature is expressed in degrees Celsius, preferably a factor in the range of 0.87 - 0.99, more preferably a factor in the range of 0.91 - 0.97. By selecting these temperature range optimal use can be made of the more efficient heating the inventors identified. Preferably, the process that is carried out in the reactor is a chemical reaction, sorption and/or desorption, wherein the chemical reaction preferably is an endothermic reaction.
A preferred process to be carried out in the reactor is an endothermic reaction. The method for induction heating of a reactor is especially suitable for such reaction as an endothermic reaction requires a constant supply of heat, otherwise the temperature in the reactor will drop leading to a poor efficiency. In addition, endothermic processes benefit from relative high temperatures which can be efficiently provided by induction heating. A preferred embodiment the invention therefore encompasses a method for carrying out an endothermic reaction in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to the invention wherein a feed provided to the reactor is converted to a reaction product that exits the reactor, preferably wherein the reactor comprises a catalyst in the form of a particulate solid and/or wherein the heating elements are coated with a catalyst. Endothermic reactions are known in the art. In a preferred embodiment the endothermic reaction is a sorption enhanced endothermic reaction. In sorption enhancement a sorbent is present in the reactor that adsorbs a reaction product. By adsorbing a reaction product by a sorbent during the reaction the reaction equilibrium beneficially shifts to the reaction products.
A preferred endothermic reactions is the reverse water gas shift reaction (RWGS reaction) wherein the feed comprises H₂ and CO₂ and the reaction product comprises CO and H₂O. In this reaction CO is produced that can be utilized as a chemical building block. Preferably for this embodiment the molar ratio H₂ to CO₂ is in the range of 0.5 - 10; preferably the catalyst is one or more metals and/or metal oxides wherein the metal is selected from the lists consisting of Cu, Pt, Ni, Fe, Co and Zn; preferably the reaction temperature is in the range of 400 - 900 °C. The RWGS reaction may be improved by sorption enhancement. Preferably the reverse water gas shift reaction is a sorption enhanced reverse water gas shift reaction. The induction heating method of the invention has as additional benefit that it can be utilized both during the reaction as during the regeneration of the sorbent.
Another preferred process to be carried out in the reactor is desorbing an adsorbate from a sorbent. When a sorbent is saturated with the adsorbate, the sorbent often is regenerated by increasing the temperature. In a preferred embodiment the invention concerns a method for desorbing an adsorbate from a sorbent in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to the invention, wherein the reactor comprises a sorbent preferably in the form of a particulate solid and wherein the adsorbate that is desorbed from the sorbent exits the reactor, preferably wherein a feed is provided in the form of a carrier gas.

General Definitions

In this document and in its claims, the verb "to comprise" and its conjugations is used in its non-limiting sense to mean that items following the word are included, but items not specifically mentioned are not excluded. In addition, reference to an element by the indefinite article "a" or "an" does not exclude the possibility that more than one of the element is present, unless the context clearly requires that there be one and only one of the elements. The indefinite article "a" or "an" thus usually means "at least one". The word "about" or "approximately" when used in association with a numerical value (e.g. about 10) preferably means that the value may be the given value more or less 1% of the value.
The present invention has been described above with reference to a number of exemplary embodiments. Modifications and alternative implementations of some parts or elements are possible, and are included in the scope of protection as defined in the appended claims. All citations of literature and patent documents are hereby incorporated by reference.
The following symbols are used in this document, that represent the following quantities (the units are between brackets):

ρ: electrical resistivity (Ω·m)
σ: electrical conductivity (S/m)
µr: relative magnetic permeability (-)
µ0: magnetic permeability of free space (4π × 10^-7 H/m)
µ: magnetic permeability (H/m)
B: magnetic flux density (Tesla, kg·s^-2·A^-1)
H: magnetic field intensity/strength (A/m)
F: frequency (Hz)
δ: penetration depth (m)
d: diameter (m) - the diameter of a wire, particle or closed loop of connected wires.
P: Power (W)
V: Voltage (V)
I: Current (A)

Examples

Method

The experiments were carried out in an experimental setup comprising a power source (Ambrell model 8310), a 9-turn induction coil (diameter 3cm, hight 8 cm), a quartz reactor placed inside the induction coil having an internal diameter of about 16 mm; a set of mass flow controllers that independently feed different gasses to the reactor. Different heating elements may be placed in the reactor on top of a porous quartz plate. A thermal camera is used to monitor and record the temperature of the heating element from the top view of the reactor (FLIR A500 24 ° Standard Science kit). For the non-reactive experiments N₂ is fed at the bottom of the reactor with a flow rate of 50 mL/min during the measurement. The electromagnetic field strength can be adjusted by adjusting the current through the coil delivered by the power source.
The heating elements were made from metal foil using by laser cutting. The Fe heating elements were made by 0.20 mm foil (99.5% Fe), the Ni heating element was made by 0.20 mm foil (99.6% Ni), The Cu heating element (control heating element) was made by Cu foil (0.254mm of thickness, 99.9% Cu).
For the reaction experiments about 1 g of a catalyst bed was provided on top of the porous quartz place with 2 iron rings placed in the bed. A Cs-Cu-Fe-Al₂O₃ catalyst was used according to Pastor-Pérez et al., Journal of CO2 Utilization, 21:423-428, 2017, having a particle size distribution in the range of 250 - 500 µm. The current was stepwise increased. The corresponding temperature is measured by a pyrometer. The reverse water gas shift reaction was carried out in the reactor. H₂ and CO₂ were provided to the reactor. The flow was held constant with a weight hourly space velocity (WHSV) of 12.500 mLg^-1s^-1 with an H₂/CO₂ ratio of 4:1 balanced in 50v% nitrogen. The outlet gas composition was monitored using an infrared gas analyzer (model gms815p). CO₂ conversion was calculated based on the incoming and outgoing CO₂ flow (any volume changes was compensated for by relating the incoming and outgoing CO₂ flow to the metered incoming and (measured) outgoing N₂ flow).

Results

Iron mesh

An iron mesh was tested. In figure 1a the mesh is depicted. The current was step-wise increased (59A, 68A, 77A) and subsequently stepwise decreased (68A, 59A, 50A). Figure 1b demonstrates that the temperature followed the stepwise increase of the current, with a sudden large increase around the Curie temperature (from 630 °C to 930 °C). The jump demonstrates that induction heating above the Curie temperature of the heating element is more efficient than below the Curie temperature, since a small increase in current give a large increase in temperature.
Surprisingly, when the current subsequently was reduced after having reached the Curie temperature, the temperature of the heating element remained relatively high. After having reached the Curie temperature, the same current results in a much higher temperature compared to the temperature before having reached the Curie temperature (at 68A, 870°C versus 630°C).

Iron disk

The experiment was repeated with an iron disk. The step-size in current was smaller compared to the iron mesh experiment. As can bee seen in fig 2a, the iron disk shows a similar behaviour as the iron mesh. The temperature gradually increases with increase current, with a large step-change near the Curie temperature when the current was increased from 52 to 53 Ampere. Contrary to the experiment with the iron mesh, the current was further increased after having surpassed the Curie temperature. Only after 5 subsequent steps of increasing the current (after having reached the Curie temperature) the current was stepwise reduced with a same step size. During the first current reduction steps the temperature of the disk for a same current was about the same, regardless of if the current was selected in the ascending cycle or the descending cycle. Interestingly, when the current was set in the descending cycle at or below the current that in the ascending cycle resulted in a temperature below the Curie temperature, the temperature remained surprisingly high, above the Curie temperature. So, with a same current a much higher temperature could be reached in the descending cycle. This effect remained in the descending cycle until the temperature of the heating element neared the Curie temperature. A sudden drop in temperature was observed when lowering the current near this temperature.
This effect is further visualized in figure 1b. The figure demonstrates a working areas in which a surprisingly high temperature can be maintained in relation to the current provided. So, by first increasing the temperature above the Curie temperature of the heating element to a first temperature and subsequently lowering the temperature (by lowering the induction heating power) to a second temperature above the Curie temperature (but evidently lower than the first temperature) a surprisingly efficient heating can be realized. In figure 2b this the more efficient heating is at a temperature of about 900 °C or lower. 900 °C is a factor 1.17 higher than the Curie temperature of iron (770 °C). The temperature gain (the difference in temperature between the ascending cycle and the descending cycle at a same current) is about 200 °C.

Nickel disk

The experiment with the iron disk was repeated with a nickel disk. The same effects were observed for the nickel disk around the Curie temperature of nickel (354 °C) as can be seen in figures 3a and 3b. Below a temperature of about 410 °C there is more efficient heating in the descending cycle. This more efficient heating remains present up to when the heating element reaches the Curie temperature of nickel. 410 °C is a factor 1.16 higher than the Curie temperature of nickel (354 °C). The temperature gain (the difference in temperature between the ascending cycle and the descending cycle at a same current) is about 100 °C.

Cupper disk

The experiment with the iron and nickel disk was repeated with a cupper disk. Cupper is not a ferromagnetic material. The disk was heated in steps to a temperature of 910°C and subsequently the temperature was stepwise reduced. For all currents, the temperature in the ascending cycle was about the same as in the descending cycle. No range was present in which the current in the descending cycle resulted in a higher temperature than the current in the ascending cycle.

Iron disk - duration test

The iron disk of the iron disk example was brought to a temperature of 951 °C with a current of 55A. Subsequently the current was reduced to 50A. In the ascending cycle this current resulted in a temperature of 630 °C. In the descending cycle 50A resulted in a temperature of about 862 °C (232 °C higher than the temperature at the same current in the ascending cycle). The current was kept at 50A for 10 hours. The temperature remained the same for this whole period. Only when the current was further dropped to 40A after 10 hours, the temperature dropped to below the Curie temperature to 521 °C.
This result demonstrates that the surprising effect that was found by the inventors is stable over time and therefore robust.

Reverse water gas shift reaction (RWGS)

The quartz reactor as described in the method section was fed with CO₂ and H₂. The reactor was heated by induction heating and the current through the coil surrounding the reactor was stepwise increased and subsequently decreased. The temperature in the reactor was measured and the CO₂ conversion could be calculated based on the metered incoming CO₂ feed and the measured CO₂ concentration in the outgoing flow. As expected, it was found that conversion increased with temperature (data not shown). Above 450°C this increase was almost linear.
In figure 5 the induction heating power (W) provided is depicted as a function of time. The power was increased in 6 steps and decreased in 7 steps. At about minute 60 the reactor surpassed the Curie temperature. Interestingly, when decreasing the power after having reached the Curie temperature, the conversion remained relatively high. For example, at 93 W the conversion in the ascending cycle was 50% whilst in the descending cycle the conversion is well above 60%. At lower power, the difference in conversion for a given power between the ascending cycle and the descending cycle is even larger. The higher conversion in the descending cycle can be explained by that the temperature is higher due to the activation effect demonstrated in the previous examples (activation of the induction heating element above the Curie temperature), which results in a higher temperature after activation for a same power input and therefore a higher conversion.
The experiment was repeated by placing the reactor in a convection oven. The power consumption of the oven for a same temperature was several times higher. Moreover, the time to reach a setpoint temperature is much larger for the convection oven (seconds for induction heating whilst it takes tens of minutes in a convection oven).

Conclusion

The example demonstrates that induction heating of a reactor by making use of heating elements made of a ferromagnetic material is surprisingly efficient when the temperature is brought above the Curie temperature of the heating elements. Especially when the temperature is first brought above the Curie temperature (which may be seen as an activation step) and subsequently the temperature is lowered to a second temperature still above the Curie temperature, the power input to maintain the second temperature surprisingly is much lower than expected. For a same power input without activation the temperature is at least 100°C lower than when the heating element is activated. This temperature difference seems to depend on the Curie temperature of the heating element, for iron with a Curie temperature of 770°C the temperature difference exceeds 200 °C whilst for nickel with a Curie temperature of 354 °C the difference exceeds 100 °C. For both materials a significant gain in heating efficiency may be realized by the method of the current invention.
The effect can be made to use for chemical reactions and sorption processes carried out in the reactor. The RWGS example demonstrates that activation of the heating element leads to that for a lower power input a significant higher conversion can be obtained.

Claims

A method for induction heating of a reactor wherein the reactor comprises one or more heating elements made of a ferromagnetic material, wherein
- in a first step the one or more heating elements are heated by induction heating to a first temperature above the Curie temperature of the ferromagnetic material, and subsequently the temperature of the one or more heating elements is kept constant at this first temperature; or wherein

- in a first step the one or more heating elements are heated by induction heating to a first temperature above the Curie temperature of the ferromagnetic material, and in a second step the temperature of the one of more heating elements is lowered to a second temperature by lowering the induction heating power, and subsequently the temperature of the one or more heating elements is kept constant at this second temperature, wherein the second temperature is below the first temperature but above the Curie temperature of the ferromagnetic material.
The method according to claim 1 wherein the temperature of the heating elements is kept constant at the first temperature or at the second temperature for at least 10 seconds, more preferably for at least 100 seconds, even more preferably for at least 1000 seconds.
The method according to any one of the preceding claims, wherein the first temperature is 10 - 500 °C, preferably 15 - 400 °C, more preferably 20 - 300 °C above the Curie temperature of the ferromagnetic material and/or at least a factor 1.10 above the Curie temperature of the ferromagnetic material of the heating elements, preferably at least a factor 1.15, more preferably at least a factor 1.20 wherein the Curie temperature is expressed in degrees Celsius.
The method according to any one of the preceding claims wherein the second temperature is 25 - 300 °C, preferably 15 - 200 °C, even more preferably 10 - 100 °C above the Curie temperature of the ferromagnetic material of the heating elements and/or wherein the second temperature is less than a factor 1.20 above the Curie temperature of the ferromagnetic material of the heating elements, preferably less than 1.15, more preferably less than 1.10 wherein the Curie temperature is expressed in degrees Celsius.
The method according to any one of the preceding claims, wherein the second temperature is at least a factor 0.01 lower than the first temperature, preferably at least a factor 0.02, more preferably at least a factor 0.05, wherein the first temperature is expressed in degrees Celsius.
The method according to any one of the preceding claims wherein a chemical reaction, sorption and/or desorption takes place in the reactor.
The method according to any one of the preceding claims wherein the reactor further comprises a solid selected from the list consisting of a catalyst, a sorbent, a solid reactant, and combinations thereof, preferably the reactor further comprises a solid selected from the list consisting of a catalyst, a sorbent, and combinations thereof.
The method according to any one of the preceding claims wherein the ferromagnetic material of the heating elements is selected from the list consisting of iron, iron alloys, nickel, nickel alloys, cobalt, cobalt alloys, manganese, manganese alloys, bismuth, bismuth alloys, antimony, antimony alloys, neodymium, neodymium alloys and mixtures thereof.
The method according to any one of the preceding claims wherein the volume of the total of the heating elements is less than 40 vol.% of the total empty reactor volume, preferably less than 20 vol.%.
The method according to any one of the preceding claims, wherein the heating elements are particles and/or wires.
The method according to any of the preceding claims wherein the heating elements comprise wires, and the wires are connected to form a 2-dimensional and/or 3-dimensional heating structure, preferably wherein the heating elements form a 2-dimensional mesh.
A method for carrying out an endothermic reaction in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to any one of claims 1-11 wherein a feed provided to the reactor is converted to a reaction product that exits the reactor, preferably wherein the reactor comprises a catalyst in the form of a particulate solid and/or wherein the heating elements are coated with a catalyst.
The method of claim 12, wherein the endothermic reaction is the reverse water gas shift reaction and the feed comprises H₂ and CO₂ and the product comprises CO and H₂O.
A method for desorbing an adsorbate from a sorbent in a reactor, wherein the reactor is heated according to the method for induction heating of a reactor according to any one of claims 1-11, wherein the reactor comprises a sorbent preferably in the form of a particulate solid and wherein the adsorbate that is desorbed from the sorbent exits the reactor, preferably wherein a feed is provided in the form of a carrier gas.