Patent Application
20120260962
The invention relates to an electric generator based on a thermoelectric effect including at least:
The invention also relates to a method for implementing such an electric generator based on a thermoelectric effect.
The invention also relates to a method for manufacturing such a generator.
Currently, large efforts are made for developing transportable electric generators.
One of the possible ways to produce a transportable electric generator consists in using a thermoelectric converter, i.e. an operative device based on the principle of the conversion of a thermal energy into electric energy, by Seebeck effect.
In particular, as represented in
Each couple is in particular constituted of a thermoelement 2a, formed by a conducting or semiconducting material having a positive Seebeck coefficient, and constituted of a thermoelement 2b, formed by a conducting or semiconducting material having a negative Seebeck coefficient. The thermoelements 2a and 2b are connected to each other in twos by junctions 3a (hot-side junctions) or 3b (cold-side junctions), formed by an electrically conducting material, in order to obtain an electric connection in series of said thermoelements. Moreover, the thermoelements 2a and 2b composing the module are thermically connected in parallel in order to optimize the heat flux through the module from its first face 4a towards its second face 4b as well as its electric resistance, thanks to the application of a heat gradient (ΔT=Tc−Tf) from the face 4a towards the face 4b. Such a heat flux then involves a displacement of the charge carriers and thus the appearance of an electrical current I.
In particular, when the thermoelements 2a and 2b have respectively a positive Seebeck coefficient (noted S2a) and a negative Seebeck coefficient (S2b), their Seebeck coefficients add to each other and the potential difference V for the module has the following formula:
V=N×(S2b−S2a)×ΔT=N×Snp×ΔT
where:
The maximum power output by such a thermoelectric converter, for a resistive charge equal to the internal resistance of the converter rint, has the following formula (1):
where Tc and Tf are respectively the temperature of the hot area and the temperature of the cold area of the converter.
In addition, the ideal output of such a converter corresponds to the ratio of the useful electric power, output to a load resistor R equal to the internal resistance of the converter, to the heat flux through the material. It has in particular the formula (2) below:
with
and ZTm corresponding to a coefficient called “factor of merit” directly depending on the electric and thermal properties of the thermoelectric materials used to constitute the thermoelements 2a and 2b.
The equations (1) and (2) then show that the power and the output of the thermoelectric converters are directly related to the heat gradient (ΔT=Tc−Tf) applied between the two faces of the thermoelectric converter. Thus, in any system of the thermoelectric converter type, the existence of a heat gradient is thus determining for its performance and in particular to obtain a good energy effectiveness.
The heat gradient applied to the thermoelectric converter depends in particular on the heat source used to maintain the face 4a of the thermoelectric converter at a temperature called hot temperature Tc and on the heat dissipater used to maintain the face 4b of the thermoelectric converter at a temperature called cold temperature Tf.
For the heat source, it has been proposed to exploit the heat produced by an exothermic chemical reaction as the catalytic combustion of hydrogen or another fuel such as butane, propane or ethylene. The generated heat then makes it possible to maintain the face 4a of the thermoelectric converter at a sufficiently high temperature Tc.
On the other hand, to obtain a sufficiently high heat gradient in order to guarantee a high energy performance, the system allowing to evacuate the thermal power accumulated by the thermal converter, on the cold side thereof, must be powerful in order to obtain a temperature Tf much lower than Tc (ΔT>200° C. at the thermal balance). However, the powerful solutions in terms of cooling currently used are bulky and consume a part or the totality of the energy produced by the thermoelectric converter. That explains the low densities of power and the poor overall energy yield of the existing systems: energy lost in the ventilators, in a system with a circulation of water or cooling fluid or in the production of a cooling fluid and of fuel.
As an example, the U.S. Pat. No. 6,313,393 proposes an electric power generator using a microstructured architecture in order to improve the transfer of heat. Such a generator contains a combustion chamber in the form of a microchannel, used as a heat source. The chamber is in particular the center of an exothermic chemical reaction of a fuel. The heat released by the chemical reaction is then transferred to a device intended to convert thermal energy into electrical current, which can be then transmitted outside the device by means of output wires. Then, after flowing through the device, the heat flux is recovered by a heat dissipator. This dissipator can be formed for example by a thermal desorbor, a system provided with microchannels for the passage of a cooling fluid in order to form a heat exchanger, a microchemical reactor, center of an endothermic chemical reaction.
The object of the invention is to propose an electric generator based on a thermoelectric effect whose energy performance is improved compared to the state of the art.
In particular, the object of the invention is to provide an electric generator based on a thermoelectric effect including a thermoelectric converter to which a high heat gradient can be applied without the overall energy performance of the electric generator not being degraded, while being sufficiently compact and easy to implement.
According to the invention, this objective is reached in that an electric generator based on a thermoelectric effect including at least:
According to the invention, this objective is also reached by a method of implementation of such an electric generator characterized in that said product of the chemical endothermic reaction forming said reagent of the exothermic chemical reaction is hydrogen.
This objective is also reached by a manufacturing method for such an electric generator based on a thermoelectric effect, characterized in that it is obtained by at least one step of powder injection molding, in particular of nanometric size.
Other advantages and characteristics will more clearly arise from the following description of specific embodiments of the invention given as nonrestrictive examples and represented in the annexed drawings, in which:
According to a specific embodiment illustrated in
Moreover, the thermoelectric converter 1 is interposed between a heat source and a heat dissipator in order to obtain a heat gradient ΔT=Tc−Tf between the two opposite faces 4a and 4b of the thermoelectric converter 1. The faces 4a and 4b form respectively the heat and cold areas of the thermoelectric converter 1, each of them being submitted to a heat flux, a hot one (φhot) for the hot area and a cold one (φcold) for the cold area.
In
The first circulation channel 5 comprises at one of its ends at least one input opening 7 intended to supply the first circulation channel 5 with at least one reagent of the exothermic chemical reaction. In
The exothermic chemical reaction is advantageously a catalytic combustion reaction for hydrogen (H2 or dihydrogene). According to this chemical reaction, hydrogen reacts with oxygen, for example provided by the ambient air, in order to produce water and heat. The combustion reaction for hydrogen is very energetic with a calorific value of 140 MJ/kg. Moreover, the flame temperature of this combustion reaction can reach more than 1500° C. In certain cases, hydrogen can be replaced by another fuel, such as butane. However, hydrogen remains the preferred fuel because it has a calorific value higher than that of another fuel. Butane has a calorific value of 50 MJ/kg.
In general, such a reaction is carried out with the help of a catalyst, which makes it possible to improve its output. The catalyst advantageously used is platinum, for example in a nanometric form such as platinized carbon, but it can also be selected among ruthenium, thorium, silver, copper, zinc, an alloy of iron and palladium, nickel and manganese. As an example, the first circulation channel 5 can be formed for example by a wall whose internal surface is covered with platinum particles, of micrometric or nanometric size. Moreover, the catalytic coating formed by these platinum particles can in an advantageous way have a strong porosity, in order to increase the surface reaction, and thus the reaction yield.
Thus, the exothermic chemical reaction occurs inside the first circulation channel 5, which then forms a catalytic combustion chamber, while making a rectional mixture circulate inside it from the input openings 7 and 8 to the output opening 9. Initially, the rectional mixture is mainly composed of air and of hydrogen, then it becomes gradually richer with water steam along the circulation channel 5 in favor of the initial reagents, until it mainly includes water at the output opening 9, indeed only water according to the length of the first channel 5. The water produced by the exothermic chemical reaction can advantageously be collected at the output opening 9. In particular, according to the type of endothermic reaction, it can be used as a reagent for the endothermic chemical reaction. Thus, the heat released along the first circulation channel 5 is decreasing from the input openings 7 and 8 to the output opening 9. This is more particularly illustrated in
In addition, the second circulation channel 6 comprise at one of its ends at least one first input opening 10 intended to supply the second circulation channel 6 with at least one reagent of the endothermic chemical reaction. It also comprises, at its other end, an output opening 11, intended to evacuate the rectional mixture of the endothermic chemical reaction and more particularly at least one product of the endothermic chemical reaction. In addition, this output opening 11 is connected to the input opening 8 of the first circulation channel 5.
The endothermic chemical reaction is indeed selected so that at least one of its products forms one of the reagents of the exothermic chemical reaction. In particular, if the exothermic chemical reaction is a catalytic combustion reaction for hydrogen, the endothermic chemical reaction is selected among the reactions making it possible to obtain hydrogen so that, once formed, it is redirected towards the input opening 7 of the first circulation channel 5.
There are several endothermic chemical reactions making it possible to produce hydrogen: reforming of ethanol, methanol, ethylene glycol, methylcyclohexane, glycerol, hexane, methane, or ammonia cracking.
Tables 1 and 2 below show, as an example, the energy balance of the endothermic chemical reactions for a reforming process from two compounds: methanol and methylcyclohexane.
In a preferential way, the endothermic chemical reaction is a steam reforming reaction from methanol, consisting in making methanol react in the presence of heat and water steam to produce hydrogen and carbon dioxide. Indeed, among various hydrocarbons, the choice of methanol has many advantages: it is easy to produce, it is not very toxic in particular in comparison with ammonia and other hydrocarbons. It reforms at moderate temperatures (typically 150-250° C.) compared to other hydrocarbons (mainly from 300° C. to 600° C.) and it has a high endothermic energy density during the chemical reaction (evaporation+reforming). It absorbs a lot of energy (≈720 kJ/kg), which makes it possible to balance the heat fluxes between the heat and cold areas in the thermoelectric converter 1.
In addition, the yield of the steam reforming reaction for methanol can be close to 100%, for temperatures between 250° C. and 300° C., when the reaction is carried out in the presence of a catalyst chosen among copper, zinc, aluminum, zirconium and palladium. Thus, the second circulation channel 6 can be formed, like the first circulation channel 5, by a wall whose internal surface is covered with catalyst particles, of micrometric or nanometric size.
In the generator 1 according to
In the case of a reforming reaction for methanol or other hydrocarbons, the reagents of the endothermic chemical reaction that have not reacted at the output opening 11 can be directed, like hydrogen, towards the input opening 7 of the first circulation channel 5, in order to be consumed there by catalytic combustion and to contribute to the heat flux.
In
Thermically coupling a thermoelectric converter with a heat source forming the center of an exothermic chemical reaction in order to generate the hot flow and with a heat dissipator forming the center of an endothermic chemical reaction in order to generate not only the cold flow but also the reagent of the exothermic chemical reaction makes it possible to obtain an electric generator whose performance, in particular energy performance, can be improved.
This is partly carried out by choosing a specific endothermic chemical reaction, so that at least one of the products of said reaction is also one of the reagents of the exothermic chemical reaction, the latter being then at least partly redirected from the heat dissipator to the heat source.
With such a solution, it is thus possible to obtain a high heat gradient at the thermal balance, while being freed from the problems of spatial requirement for storing the reagent of the exothermic chemical reaction or due to the big size of the heat dissipators according to the anterior art and from the problems of energy supply to make the heat dissipator work.
In addition, the fabrication of such an electric generator as well as its starting are simplified. Moreover, it is obtained a considerable reduction of the mass of the reagents at work, which makes it possible to increase the mass energy density of the electric generator based on a thermoelectric effect. Moreover, the electric generator based on a thermoelectric effect is energetically autonomous, has small spatial requirement and is energitically optimized.
However, obtaining such a performance and in particular energy performance with such an electric generator remain delicate because many factors related to the operation of this generator must be taken into account to find the single and optimal operating point for the electric generator. In particular, it is necessary to obtain:
These three types of coupling are, in addition, dependant on each other.
The catalytic combustion reaction of hydrogen is, for example, five times more energetic than the reforming reaction from methanol. Thus, if the reforming reaction for methanol is complete, only one fifth of the produced hydrogen is used for the exothermic catalytic-combustion reaction.
Consequently, to obtain a balance, it can be useful to moderate the chemical catalytic-combustion reaction by controlling the air flow through the additional input opening 8. An oxygen deficit or excess indeed makes it possible to lower in a significant way the yield of the catalytic combustion reaction.
According to another alternative, it is also possible to associate the thermoelectric converter with a fuel cell in order to consume the hydrogen produced in excess by the endothermic chemical reaction, after a possible filtering. As an illustration,
In particular, in
Conversely, in another embodiment represented in
According to another development of the invention, the electric generator can include other elements than the fuel cell. For example, it can include a series of microturbines arranged before the input of the heat dissipator, in order to collect the mechanical energy produced by the circulation of the reagents of the endothermic chemical reaction. The addition of this series of microturbines can make it possible to increase the energetic efficiency of the electric generator.
In addition, concerning the endothermic chemical reaction, it is preferable to use a low reforming temperature in order to guarantee the highest possible heat gradient between the hot and cold areas in the thermoelectric converter. This can be obtained, for example, by controlling the ratio “steam to carbon”, also written SIC. In the case of methanol, it could be possible to carry out the reaction of steam reforming at 50° C. with a ratio SIC from 4 to 5. However, at 50° C., the steam reforming reaction also produces carbon monoxide. However, this product can be harmful, in particular when the generator includes a fuel cell. Thus, the steam reforming reaction for methanol is advantageously carried out at reforming temperatures between 200° C. and 400° C. and advantageously between 250° C. and 300° C.
It also appears that the main parameter making it possible to obtain a good operating point is the thermal conductivity of the thermoelectric converter. However, to be able to control this parameter in order to ensure the correct operation of the device and thus to find the best compromise to generate a sufficient electric power, it is necessary to be ensured that:
1) the temperature Tc in the combustion chamber is maximum (750-800° C. for a catalytic combustion reaction for hydrogen coupled with a steam reforming reaction for methanol), in order to obtain a very high heat gradient and thus a maximum electric power output,
2) the temperature Tf in the heat dissipator is minimal in order to ensure the correct operation thereof and to maximize the reaction yield. For example, Tf is about 200° C. to 400° C. in the case of the steam reforming from methanol coupled with a catalytic combustion reaction of hydrogen,
3) the heat source and the heat dissipator have dimensions (size of the channels, catalyst, . . . ) allowing for each element to obtain the highest possible reaction yield and
4) the thermoelectric converter is thermically dimensioned.
Contrary to the heat source and to the heat dissipator, the thermoelectric converter cannot be dimensioned only by taking account of its own performance. It is not possible indeed to attempt to obtain a maximum electric power output, with a maximum filling ratio and a low height for the thermoelements in order to decrease the internal resistance of the converter. Indeed, the thermoelectric converter depends on the heat source and the heat dissipator, because it is in physical contact with these two elements. It is thus necessary to dimension its thermal conductivity so that, for a Tc value fixed at a preset temperature (for example between 700° C. and 800° C.) and for a quantity of absorbed energy predetermined by the endothermic reaction, the value Tf is maintained at a predetermined temperature (for example between 200° C. and 400° C. in the case of a steam reforming from methanol coupled with a catalytic combustion reaction for hydrogen).
In short, the thermoelectric converter must be designed, from a dimensional point of view, in order to obtain a compromise between the optimal thermal conductivity, which makes it possible to ensure the correct operation of the device, and the electric power output by the device. More particularly, it is thermically dimensioned to comply with the following formulas:
φconduction=φcold−φSeebeckCold−φjoule
φconduction=φhot−φSeebeckHot+φjoule
in which:
φconduction is the heat flux through the thermoelectric converter from the first area to the second area,
φcold and φhot are the heat fluxes respectively absorbed by the endothermic chemical reaction and the exothermic chemical reaction,
φSeebeckHot and φSeebeckCold are the Seebeck heat fluxes respectively in the area (4a) in contact with the heat source and in the area (4b) in contact with the heat dissipator and, more particularly, that is to say φSeebeckHot=N.S.I. Tc and φSeebeckCold=N.S.I. Tf with N corresponding to the number of thermoelements forming the thermoelectric generator, S corresponding to the Seebeck coefficient of the material, I being the current through the thermoelements and Tf and Tc are respectively the temperatures in the areas 4a and 4b, and
φjoule is the heat flux produced by Joule effect in the thermoelectric converter.
As an example, for a flow of 250 g/h of methanol and water (that is φendo=90 W/cm2), for a filling ratio (surface of the converter/cumulated surface of the thermoelements) of 50%, and a value of Tc regulated at 700° C. (via the exothermic reaction yield), it is necessary that the thickness of the thermoelements be between 0.8 mm and 1.5 mm, so that Tf lies between 200° C. and 400° C.
The nature of the material(s) forming the thermoelements 2a and 2b of the thermoelectric converter 1 are advantageously selected according to the range of temperature considered, both for the hot area and the cold area. Thus, the thermoelements 2a and 2b can be made of SiGe if the endothermic chemical reaction is a catalytic steam reforming reaction of methanol and the exothermic chemical reaction is a catalytic combustion reaction of hydrogen. Indeed, SiGe has the best thermoelectric performance in the range of temperature considered for this couple of chemical reactions, i.e. a maximum temperature in the hot area (Tc=900° C.) and a minimal temperature in the cold area (Tf of about 150° C.), with an average temperature of about 525° C. between the heat and cold areas.
As example, in the case of a SiGe-based thermoelectric converter and by using the reforming of methanol to absorb the combustion energy on the cold side and to produce hydrogen, the theoretical maximum electric energy density Eelecmax is equal to 38 Wh/kg (methanol+water), with Tc=800° C. and Tf=300° C. Moreover, by using a nanostructured-SiGe-based thermoelectric converter, it is possible to obtain an energy density of about 60 Wh/kg.
Other thermoelectric materials can also be used, as the alloy Bi2Te3. In particular, this alloy can be used in combination with the alloy SiGe in order to form segments in the module, the segment formed by couples of thermoelements 2a and 2b made of Bi2Te3 is advantageously arranged in the area where the temperature Tc is lower.
Advantageously, if the main reagent of the endothermic chemical reaction is in a liquid form, as it is the case for ammonia, methanol or ethanol mixed with water, the endothermic vaporization energy for said reagents, before the endothermic chemical steam reforming reaction itself, can be collected in order to obtain a better cooling on the cold side of the thermoelectric converter 1. Moreover, the thermoelectric converter 1 can advantageously include, at the input of the second circulation channel 7 and the output of the first circulation channel 6, an area formed by a purely thermically conducting material, without any thermoelement, in order to optimize the energy balances.
As an example and as illustrated in
Moreover, the thermoelements 2a and 2b can be formed by nanostructured thermoelectric materials, favoring the diffusion of the phonons at the interfaces between the nanoaggregates, which makes it possible to reduce considerably their thermal conductivity and at the same time to improve their thermoelectric performance.
The first and second circulation channels are not necessarily rectilinear like those represented in
The second circulation channel 5 forming the heat dissipator can advantageously be replaced by any other means making it possible to carry out the endothermic chemical reaction, insofar as it makes it possible to redirect at least one of the products of the endothermic chemical reaction towards the heat source, so that the product of the endothermic chemical reaction can be used as a reagent for the exothermic chemical reaction. As an example, the second circulation channel 6 can be replaced by a porous thin film, for example made of alumina or nickel, including a plurality of pores, advantageously of micrometric size and covered by a catalytic coating for the endothermic chemical reaction. In this case, the deposition of the catalyst can be carried out by a technique of impression, such as the technique of ink jet impression, or by techniques of chemical vapor deposition. It is the same thing for the first circulation channel 5.
The introduction of the reagents of the endothermic chemical reaction can also be carried out by any suitable means. For example, it is possible to use a pump for injecting the reagents when those are liquid. Another solution can also consist in using the heat released by the exothermic chemical reaction for pumping reagents when those are liquid. As an illustration, one can use the socalled capillary pumping technique consisting in evaporating the fluid to be pumped inside a capillary tube. At the interface fluid-gas, it is then formed a meniscus generating a pressure intended to aspire the fluid. In the case of reactive gases, if the gas is under pressure (for example in the case of butane), this pressure can be used to supply the heat dissipator with the reagent, for example by using an injector also carrying other fluids.
The introduction of the reagents of the exothermic chemical reaction can also be carried out by any known means, if at least one part of the product obtained by the endothermic chemical reaction can be redirected towards the heat source in order to be used as a reagent in said exothermic reaction. Thus, the use of a conventional minipump for injecting gas reagents can be contemplated. It is also possible to use a pump based a a venturi effect, in the hydrogen distribution channels or a compressor of the type “Knudsen”.
The electric generator based on a thermoelectric effect can be carried out for example by techniques of deposition of thin films used in the field of microelectricity and which have the advantage of offering a great density of thermoelements and thus of electric energy. It can also be obtained by more traditional techniques, such as for example at least one step of powder injection molding or PIM, more particularly from powders of nanometric size (technical also called microPIM), or by thermoelectric material sintering or brazing.
In particular, the socalled “PIM” or “microPiM” technique indeed makes it possible to carry out, advantageously in only one step, parts formed by various types of materials, such as metals, ceramic materials, . . . , with complex patterns, while being inexpensive in order to carry out a mass production. Moreover, the “PIM” or “microPIM” method makes it possible to mix various materials, in order to carry out a co-injection. In particular, it is possible to produce a matrix of slightly electrically conducting material, such as porous silicon oxide intended to form the mechanical support of the thermoelements, with thermoelectric materials in the form of powders in the same operation. The fluid circulation channels and the catalyst deposition can also be carried out by this same technique, Thus, it is possible to integrate in the same plate:
The complete system can include several plates and the assembly (right or spiral geometry) can be done in a final sintering step.
According to a specific embodiment, an electric generator such as represented in
λ=1.5 W/mK,
ρ=2.5 mΩ·cm and
and S=400 μV/K
wherein λ is the thermal conductivity of the converter, ρ is the electric conductivity and S is the total Seebeck coefficient of the converter.
The generator contains a heat source 5, center of a catalytic combustion of hydrogen and generating temperatures Tc between 750° C. and 900° C. on the hot face of the thermoelectric converter. More particularly, it is composed of a thin film made of porous alumina with coated platinized carbon.
It also contains a heat dissipator 6, center of a methanol steam reforming reaction, with a temperature Tf between 200° C. and 300° C. and being able to absorb 722 Wh/kg of reagents on the cold face of the thermoelectric converter 1. More particularly, the heat dissipator is composed of a thin film of porous alumina coated with nanometric particles of ZnO and CuO.
The heat gradient likely to be obtained for this generator is about 500° C. to 600° C. at the terminal of the converter. Moreover, the thermoelectric converter 1 includes an evaporator 16 arranged at the reagent input of the heat dissipator 6. A venturi pump 17 is also arranged between the output of the heat dissipator 6 and the input of the heat source 5, in order to supply the heat source 5 with air and a tank 18 with reagents is arranged before the heat dissipator 6.
It should be noted that, for such an electric generator, the cold flux is limited by the evaporation limit for water. Indeed, beyond the water calefaction regime (about 100 W/cm2), the liquid reagent tank for reforming methanol is isolated from the heat source 6 by a gaseous water film. Thus, the maximum quantity of energy that can be absorbed by the liquid reagent tank 18 (water+méthanol) is about 150 W/cm2. The maximum power density of such a generator is thus about 7.5 W/cm2, which is higher than that of fuel cells.
The parameters of the electric generator according to
As illustrated in
φconduction=φcold−φSeebeckCold−φjoule
φconduction=φhot−φSeebeckHot+φjoule
Moreover, in
E
elec
max=40 Wh/kg (methanol+water).
| Number | Date | Country | Kind |
|---|---|---|---|
| 09 05537 | Nov 2009 | FR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/FR10/00769 | 11/16/2010 | WO | 00 | 6/21/2012 |
