RU2275713C2 - Thermoelectric converter and method for heat energy conversion - Google Patents

Abstract

FIELD: heat energy conversion.

SUBSTANCE: highly doped n* region can serve as emitter region wherefrom charge carriers can be injected to gap region. The latter may be p region made of dope-free conductor or it may be moderately doped n region. Hot ohmic contact is connected to n* region, Cold ohmic contact functions as collector and is connected to other end of gap region. Cold ohmic contact has recombination region formed between cold ohmic contact and gap region as well as compensation blocking layer that reduces thermoelectric reverse-current component. Emitter heated with respect to collector generates electromotive force inducing current through series-connected load.

EFFECT: enhanced efficiency of conversion.

63 cl, 50 dwg

Description

BACKGROUND OF THE INVENTION

1. The technical field

The present invention relates to the conversion of thermal energy into electrical energy and to the use of electrical energy for cooling, and, in particular, to a solid-state thermionic converter using a semiconductor diode.

2. The prior art

Thermoelectronic energy conversion is a method of direct conversion of thermal energy into electrical energy using thermionic emission. In this process, thermionic emission from the metal surface occurs during heating, with a part of the electrons being given enough energy to overcome the restraining forces on the metal surface to go beyond the metal. Unlike most conventional methods of generating electricity, the conversion of heat into electricity during thermionic conversion does not require the use of intermediate forms of energy or a working fluid other than electric charges.

A conventional thermoelectronic energy converter, in its most elementary form, consists of one electrode connected to a heat source, a second electrode connected to a heat sink and separated from the first electrode by an intermediate gap, conductors connecting the electrodes to an electrical load, and a housing. From the space inside the housing, either the air is completely pumped out or it is filled with the corresponding rarefied steam, such as cesium vapor.

The process in a conventional thermionic converter mainly proceeds as follows. Heat from a heat source at a sufficiently high temperature enters one electrode-emitter, from which electrons escape in the form of thermionic emission into the interelectrode space, in which vacuum is maintained, or which is filled with rarefied steam. Electrons move through this space toward another collector electrode, whose temperature is kept low with a nearby heat sink. Here, the electrons settle and return to the hot electrode through external electrical leads and an electrical load connected between the emitter and the collector.

An embodiment of a conventional thermionic converter 100 is shown schematically in Figure 1. Such devices typically comprise an emitter 110 or a cathode with a low electron work function, a collector 112 or a relatively cooler anode with a high electron work function, a housing 114, corresponding electrical conductors 116, and an external load 118. Heat flux 120 is supplied to emitter 110, as a result of which electrons 122 are emitted from this cathode, the flow of which, thus, closes the electric circuit and, thanks to which generates an electric current in the load 118. As indicated above, in conventional thermionic converters from the space 130 between the electrodes, the air is pumped out, or it is filled with rarefied steam.

The flow of electrons through an electrical load is maintained due to the temperature difference between the electrodes. In this case, electrical work is applied to the load.

Thermoelectronic energy conversion is based on the concept that a cathode with a low electron work function in contact with a heat source emits electrons. These electrons are absorbed by a cold anode with a high electron work function, and these electrons can flow back to the cathode through an external load, where they do useful work. The use of thermionic generators in practice is limited by the values of the work function of available metals or other materials that are used as cathodes. Another important limitation is the space charge effect. The presence of charged electrons in the space between the cathode and the anode creates an additional potential barrier that reduces the thermionic current. These limitations have a negative effect on the maximum current density and, thus, represent the main problem in the development of large-scale thermionic converters.

Typically, thermionic converters are classified as vacuum converters or gas filled converters. In vacuum converters, air is evacuated from the space between the electrodes. Such converters are of limited practical use.

In embodiments of gas-filled converters of the first class, a vaporous substance is used in the space between the electrodes that allows the generation of positive ions. As such a vaporous substance, alkali metal vapors such as cesium, potassium or rubidium are usually used. Due to the presence of these positive ions, free electrons can more easily pass from the emitter to the collector. The emitter temperature in conventional devices of this type is partially determined by the evaporation temperature of a substance that produces positive ions. Typically, the temperature of the emitter should be at least 3.5 times higher than the temperature of the source material of positive ions, if in such conventional devices it is necessary to achieve a level of effective ion production.

In embodiments of second-class gas-filled transducers, a third electrode is provided for generating ions. An inert gas such as neon, argon or xenon is used as the gas filling the space between the electrodes in these conventional devices. Although these transducers can operate at lower temperatures, approximately 1500 K, they have a more complex design.

Typical thermionic emitters usually operate at temperatures in the range from 1400 to 2200 K, and collectors operate at temperatures in the range from 500 to 1200 K. Under optimal operating conditions, the overall energy conversion efficiency is from 5 to 40%, the specific electric power is about 1 up to 100 W / cm ² and the current density is on the order of 5 to 100 A / cm ² . In general, the higher the emitter temperature, the higher the efficiency, specific power and current density when using designs that take into account radiation losses. The voltage with which energy is generated in one typical converter unit is from 0.3 to 1.2 V, that is, approximately at the same level as in a conventional electrolytic cell. Thermoelectronic systems with higher power are often made up of many blocks of thermoelectronic converters connected in series electrically. Each thermionic converter unit usually has a power of 10 to 500 watts.

The requirement to use high temperature in thermionic converters is preferable for some applications, but is a limitation for other cases. This is because the temperatures required for the emitter to operate are usually outside the practical capabilities of many conventional heat sources. In contrast, typical thermoelectric converters operate at a heat source temperature in the range from 500 to 1500 K. However, even under optimal conditions, the total efficiency of thermoelectric energy converters is in the range of only 3 to 10%, and the specific electric power is usually less than a few W / cm ² and the current density is on the order of 1 to 100 A / cm ² .

From the point of view of physics, thermoelectric devices are similar to thermoelectronic devices. In both cases, a temperature gradient is applied to the metal or semiconductor, and both cases are based on the concept that the motion of an electron is electricity. However, the motion of an electron also carries energy. Forced current transfers energy in both thermoelectronic and thermoelectric devices. The main difference between thermoelectric and thermoelectronic devices is the transfer mechanism: ballistic and diffusion transfer and ohmic transfer in thermoelectric devices are used in thermoelectronic devices. The ohmic flow has diffusion at the microscopic level, but not at the macroscopic level. A distinctive feature is the presence of excess media. In thermoelectric devices, current is usually generated by carriers present. In thermionic devices, current arises due to the introduction of excess carriers into the gap. A thermoelectronic device has a relatively high efficiency if the electrons ballistically fly through the gap. In thermionic devices, all kinetic energy is transferred from one electrode to another. The motion of electrons in a thermoelectric device is a quasi-equilibrium state and is ohmic, and can be described using the Seebeck coefficient (coefficient of thermoelectric effect), which is an equilibrium parameter.

In structures with narrow barriers, the electron spacing is usually not large enough to take into account their collisions when crossing the barrier. Under these conditions, the ballistic version of the theory of thermionic emission is a more accurate representation of current transfer. In this case, the current density can be represented by the following equation:

,

,

где m представляет эффективную массу электрона и

- редуцированная постоянная Планка.where A ₀ represents the Richardson constant (constant of thermionic emission), φ is the height of the barrier (work function of the electron), e is the charge of the electron, k _in is the Boltzmann constant and T is the temperature. The constant A ₀ Richardson is determined by the following equation

where m represents the effective mass of the electron and

is the reduced Planck constant.

The above electron current density equation is a quantitative expression for explaining some of the observations described above. For example, the equation for the emission current shows that the emission rate increases sharply with increasing temperature and decreases exponentially depending on the work function.

In the prior art, attempts have been made to find a solution to the above problems by using vacuum transducers or gas-filled transducers. As a result of attempts to reduce the effect of the space charge effect using vacuum transducers, the gap between the electrodes was reduced to a few micrometers. As a result of attempts to reduce the same effect in gas-filled converters, the introduction of positive ions into the electron cloud in front of the emitter began to be used. However, such conventional devices still have disadvantages associated, for example, with limiting the maximum current density and temperature conditions. Therefore, there remains a need for a more satisfactory solution for converting thermal energy into electrical energy at lower temperature conditions with high efficiency and high specific power values.

SUMMARY OF THE INVENTION

The present invention has been developed to meet the need for a device that can efficiently convert thermal energy into electrical energy at relatively low operating temperatures and specific power and efficiency values high enough for commercial use. The present invention also operates in reverse mode to provide efficient cooling.

The present invention addresses a number of problems that are inherent in the prior art, as described above. More specifically, the device and method in accordance with the present invention constitute important progress in the field of thermionic energy conversion, as evidenced by the characteristics of the embodiments of the present invention.

, коллектор и область промежутка между эмиттером и коллектором, находящуюся в электрическом и тепловом соединении с эмиттером и коллектором. Область промежутка содержит полупроводник со вторым донором с концентрацией N_d, которую выбирают таким образом, чтобы натуральный логарифм отношения

находился в пределах от 0 до приблизительно 7.In General, the objectives of the present invention are achieved by using a solid state transducer containing an emitter, including at least a region containing a first donor with a concentration

, the collector and the region of the gap between the emitter and the collector, in electrical and thermal connection with the emitter and the collector. The gap region contains a semiconductor with a second donor with a concentration of N _d , which is chosen so that the natural logarithm of the ratio

ranged from 0 to about 7.

Other embodiments of the present invention include a solid-state thermionic converter, which uses an embodiment on a semiconductor diode comprising an emitter including an n * -type region; the gap region between the emitter and the collector, the gap region near the indicated n * -type region and the cold ohmic contact connected to the specified gap region, wherein the cold ohmic contact has a collector recombination region formed between the cold ohmic contact and the specified gap region. In some embodiments of the present invention, the collector comprises a recombination element, and such a collector is electrically connected to a cold ohmic contact. The hot ohmic contact is electrically connected to the emitter. The gap region may be n-type, p-type, or may contain pure semiconductor. In order to use the electric current generated in the converter embodiments according to the present invention, the electric circuit is usually closed externally using an electric load connected to the hot ohmic contact and the cold ohmic contact. It should be understood that terms such as "electrical connection", "electrical connection" and "electrical contact" refer to the relationship between the elements, due to which an electric current can flow between such elements, while such elements are either in direct contact or electric current flows through at least one conductor connecting such elements.

In other embodiments of the present invention, a plurality of plates are used, each of which is an emitter and a collector with a gap region between them.

In the embodiments used for cooling, carrier transfer is accomplished using an external electric field. The first ohmic contact on the emitter, containing, in one embodiment, an n * -type region, is connected to a heat load, which is cooled by removing the heat flux using electrons emitted by the emitter. As described in connection with embodiments of heat-to-electricity converters according to the present invention, electrons in the embodiments used for cooling circulate from an emitter, preferably from a hot ohmic contact in the n * -type region, to the gap region. The gap region in one embodiment is located in close proximity to the emitter, and a second ohmic contact containing a recombination region of the collector is formed between the second ohmic contact and the gap region. The gap region in the embodiments according to the present invention may be either n-type or p-type, or may be made of pure semiconductor. The heat exchanger dissipates the heat of the hot electrons at the second ohmic contact, which is connected to the gap region.

BRIEF DESCRIPTION OF THE DRAWINGS

For a more complete understanding of the method of achieving the above advantages and objectives of the present invention, a more detailed description of the present invention will be given with reference to its specific embodiments, which are illustrated in the accompanying drawings.

Figure 1 schematically depicts an embodiment of a conventional thermionic converter;

figure 2 is a view in cross section of a thermal diode, in accordance with the present invention;

figure 3 is a graph of the normalized conductivity parameter χ as a function of temperature for the InSb material, provided that N * _D = 10 ²⁰ cm ^-3 ;

figure 4 - the concentration of electrons and holes for the thermionic structure n * pn * in InSb, where the concentration of the donor in the areas of the emitter and collector is 10 ²⁰ cm ^-3 , and the concentration of the acceptor in the region of the gap is 10 ¹⁷ cm ^-3 ;

figa - electron concentration for thermoelectronic structure n * nn * in InSb, where the donor concentration in the emitter and collector regions is 10 ²⁰ cm ^-3 , and the donor concentration in the gap region is 10 ¹⁴ cm ^-3 ;

5B is a graph of normalized conductivity χ as a function of temperature for several semiconductors;

Fig.6 is the concentration of electrons and holes for the thermionic structure n * nn * in InSb, where the donor concentration in the emitter and collector regions is 10 ²⁰ cm ^-3 and the donor concentration in the gap region is 8 × 10 ¹⁷ cm ^-3 ;

Fig.7 is the normalized height of the barrier Δu as a function of the concentration of impurities;

Fig. 8 is a graph of short circuit currents as a function of the height of the normalized barrier for the structures described above at T _max = 600 K and Δτ = 0.5;

Fig.9 - current and voltage characteristics for a structure of InSb material with a thickness of 625 μm, with an emitter electron concentration of 10 ²⁰ electrons / cm ³ , at an emitter temperature T _max = 600 K and a collector temperature T _min = 300 K;

figure 10 - load power per unit area for a structure of InSb material with a thickness of 625 μm, with an emitter electron concentration of 10 ²⁰ electrons / cm ³ , emitter temperature T _max = 600 K and collector temperature T _min = 300 K;

11 is a graph of thermal power dissipated per unit area as a function of voltage for an InSb structure of 625 μm thick, with an electron concentration on the emitter of 10 ²⁰ electrons / cm ³ , at an emitter temperature T _max = 600 K and a collector temperature T _min = 300 K;

12 is a graph of intensity as a function of stress for an InSb structure. The calculation results are given for the density of impurities in the interval 10 ¹⁷ (the lowest curve on the graph), 3 × 10 ¹⁷ , 5 × 10 ¹⁷ and 7 × 10 ¹⁷ (the highest curve on the graph) in units of cm ^-3 ;

Fig. 13 shows the results of numerical optimization of the efficiency as a function of alloying the gap material in a wide range of impurity concentrations at a fixed concentration of ionized emitter impurities of 10 ²⁰ cm ^-3 ;

Fig - results for optimal efficiency as a function of doping the emitter with a fixed level of impurities in the interval N _D = 7 × 10 ¹⁷ cm ^-3 ;

Fig - graph of thermal power and power in the load per unit area for the construction of a thermal diode;

Fig is a graph of the efficiency of the design with an electron emitter concentration of 10 ²⁰ electrons / cm ³ and an impurity density in the gap of 7 × 10 ¹⁷ cm ^-3 ;

17 is a graph of optimized efficiency as part of a thermodynamic limitation;

Fig. 18 is a flow of heat power under conditions of optimal energy conversion at various temperatures;

Fig. 19 is a cross-sectional view of a compensated thermal diode;

Fig - current dependence as a function of the amount of impurities in the gap for the construction of the thermal diode InSb;

на уровне 7×10¹⁷, 10¹⁸, 2×10¹⁸ и 3×10¹⁸ см^-3;Fig - graphs of the optimization of efficiency as a function of impurities of the gap with compensation of the p-type, using concentrations

at the level of 7 × 10 ¹⁷ , 10 ¹⁸ , 2 × 10 ¹⁸ and 3 × 10 ¹⁸ cm ^-3 ;

Fig is a view in cross section of a separate compensated thermal diode with increasing temperature in the direction indicated by the arrow marked with the letter T;

figa is a graph of efficiency under optimized conditions as a function of emitter temperature for various levels of doping in the interval with perfect compensation. The use of InSb material as a compensated thermal diode with a thickness of 625 μm, with an emitter electron density of 10 ²⁰ cm ^-3 and a collector temperature of 300 K is assumed;

figv - efficiency, normalized with respect to the thermodynamic limit, under optimized conditions as a function of emitter temperature for various cases, shown in figure 23A;

24 is an embodiment using four diodes mounted in a row;

25 is an embodiment in which a plurality of diodes installed in a row are used, forming a curved line contour of a wedge-shaped shape;

Fig. 26 is an embodiment in the form of a series of diodes in which the circuit approaches the ideal curve shown in Fig. 25;

figa-27B - the dependence of efficiency as a function of temperature for optimized embodiments of the compensated thermal diodes, in which the collector temperature is approximately 300 K, in accordance with the present invention;

28 and 29 are graphs of the dose dependence necessary to create a compensated layer in a wide range of ion energy in an n-type InSb diode doped to a concentration of n;

Fig. 30 shows the results for the dose doping of the ohmic contact required to obtain a fine layer of Te impurities with a concentration of 10 ²¹ cm ^-3 in the InSb construct;

Fig - depth of penetration of ions for Fig;

Fig - the results of the introduction of Ag impurities to obtain ohmic contact;

Fig - the results of the introduction of impurities Ag to obtain ohmic contact;

Fig - temperature characteristics of the gap InSb;

Fig - temperature dependence on the height of the barrier for a layer of Te impurity to a concentration of 3 × 10 ¹⁹ cm ^-3 deposited on InSb with Te impurity to a concentration of 1 × 10 ¹⁸ cm ^-3 with an In emitter;

figa - state of the metal-semiconductor contact surface;

Fig. 36B is a schematic embodiment of the present invention, which comprises a layer at a metal-semiconductor interface designed to reduce a barrier;

Fig - curves I-V of the current versus voltage for one diode and installed in a row of three InSb diodes;

Fig is a graph of the efficiency of one plate for InSb as a function of the dose of ion doping Ar for a sample with dimensions of 0.50 × 1.0 × 1.5 mm ³ ;

Fig. 39 is a graph of the depth of ion doping ⁴ Not as a function of ion energy for an InSb target;

Fig - simulation results of many vacancies per ion as a function of the energy of the ⁴ He ion for ion doping, shown in figure 39;

Fig. 41 is a graph of the output current density for an embodiment containing a sample of Hg material _0.86 Cd _0.14 Te, as a function of the temperature of the hot side of the sample with a copper (Cu) emitter layer, and a sample with an In-Ga emitter layer;

Fig. 42 is a graph of the output current density for an embodiment containing a sample of Hg _0.86 Cd _0.14 Te material as a function of the hot side for a sample with an aluminum (Al) substrate and for a sample with an In-substrate;

Fig. 43 is a graph of absolute efficiency as a function of temperature for an embodiment of a compensated thermal diode containing a sample of material Hg _0.86 Cd _0.14 Te;

Fig. 44 is a graph of the efficiency expressed as a percentage of the ideal Carnot efficiency cycle as a function of temperature for the same embodiment as shown in Figure 43;

Fig. 45 is a graph of absolute efficiency for an embodiment of a transducer with a layered structure as a function of the temperature of the hot plate;

FIG. 46 is a graph of efficiency expressed as a percentage of the ideal efficiency of the Carnot cycle as a function of the temperature of the hot plate for the same embodiment as shown in FIG. 45;

Fig. 47 is a graph for a normalized indicator of material quality Hg _1-x Cd _x Te with respect to this indicator for InSb as a function of x;

Fig. 48 is a thermal diode for cooling;

Fig. 49 is a compensated thermal diode for cooling;

Fig. 50 is a graph of the performance as a function of temperature for embodiments intended for cooling.

DETAILED DESCRIPTION OF THE PRESENT INVENTION

The present invention is embodied in the form of a solid-state thermionic energy converter 10, which, in General, is depicted in figure 2, and is directed to a method and apparatus for converting energy. One embodiment of the solid-state thermionic energy converter 10 in accordance with the present invention comprises a semiconductor diode including an n * -type region 14 used as an emitter, a gap region 16 located close to the n * -type region 14, a hot ohmic contact 12, connected to the indicated region 14 n * -type, and a cold ohmic contact 20, which is a collector and which is connected to the specified region 16 of the gap. In one embodiment, the cold ohmic contact 20 has a collector recombination region 18 formed between the cold ohmic contact 20 and said gap region 16.

The recombination region in some embodiments of the present invention contains a distinct layer. In other embodiments of the present invention, a recombination region is obtained by treating and / or causing damage to the surface of an ohmic contact or collector. The formation of a recombination region in the context of the present invention thus includes procedures for embedding a recombination layer and procedures for treating and / or damaging an ohmic contact or collector surface.

The term n * region, as used herein, denotes an n-type region that has a higher electron concentration than usual an n-type region. Illustrative embodiments of materials constituting n * -regions are given below. The general characteristics of the n * region and n region, depending on the relative donor density Nd * and Nd, are given below. Examples of n * -type regions are regions that include InSb with Te impurity at a concentration of from about 10 ¹⁶ cm ^-3 to about 10 ¹⁹ cm ^-3 . Concentrations of the order of 10 ²⁰ cm ^{−3 are} also considered as characteristic of material impurities in n-type regions in embodiments of the present invention. Examples of n * -type regions are represented by regions that include InSb with Te impurity at concentrations from about 10 ¹⁹ cm ^-3 to about 3 · 10 ¹⁹ cm ^-3 . Concentrations of approximately 3 × 10 ²⁰ cm ^{−3 are} also considered characteristic of material impurities in n * -type regions in embodiments of the present invention. In addition to Those in some other embodiments of the present invention, the impurities include at least one of the following elements: sulfur, selenium and tin (S, Se and Sn). In addition, the symbol n ** is used here to denote the n * region with a higher electron concentration than in the n * region. Examples of n ** -type regions are represented by regions that include impurity materials such as indium, tellurium, gallium, and iron (In, Te, Ga, and Fe).

In an embodiment of the converter according to the present invention, an generated electric current flows through an electric load R _L connected to the hot ohmic contact 12 and the cold ohmic contact 20. The emitter may be metallic. The gap region 16 may be an n-type region, a p-type region, a region with a moderate amount of impurity, or may consist of a pure semiconductor. Electrons are collected in the region 18 of the recombination of the collector. The heated emitter generates an electromotive force with respect to the collector, which leads to the formation of current through a series-connected load.

It should be noted that the principle of operation in accordance with the present invention is the same for hole conductivity as well as for electrons. In addition, metal references provided herein also include alloys.

Unlike conventional thermionic devices, embodiments of the transducers in accordance with the present invention are solid state devices. However, only devices based on the use of vacuum between the electrodes, or in which the space between the electrodes are filled with gas, are described in the prior art. The general characteristics of such conventional devices are briefly described above.

Instead of a space with a vacuum or a space filled with gas, embodiments of the present invention comprise a semiconductor material.

Semiconductors are used not because they have electrical conductivity, but because they have two unusual properties. First: the concentration of free carriers in semiconductor materials, and therefore the electrical conductivity increase exponentially with increasing temperature (approximately 5% per degree Celsius at ordinary temperatures). Second: the electrical conductivity of a semiconductor can be significantly increased and to a precisely controlled extent by adding small amounts of impurities in a process called doping of the semiconductor. Since there are two types of mobile charge carriers (electrons and holes) with opposite signs, unusual distribution of charge carriers can be created. This property is used in a semiconductor diode. Semiconductors, pure or doped, p-type or n-type, are two-sided; current flows in them in any direction with equal force. If, however, the p-type region is located in close proximity to the n-type region, a carrier gradient is created that is unidirectional; while the current flows freely in only one direction. The resulting device — a semiconductor diode — exhibits a very useful carrier transfer control property that can be used to convert energy.

The following description and graphic material relate to models and / or simulations of a phenomenon associated with working embodiments of the present invention. References to these models and / or simulations are not intended to limit the description of the present invention. It should be understood that the present invention is not limited and not limited to any single explanation of the underlying physical processes. Models and / or simulations are intended to provide appropriate variables that can be used to develop additional embodiments envisioned within the scope of the present invention, even if such embodiments were not described in detail in the context of the present description. When such tools are used in the present description, one of ordinary skill in the art may develop further embodiments that are within the scope of the present invention and claims. Accordingly, the following description and graphic material represent embodiments of the present invention and models that can be used to develop additional embodiments contemplated within the scope of the present invention.

The following are the results for a thermal diode with InSb material, since InSb is one of the semiconductor materials used for the embodiments of the present invention. It was shown that the behavior of the embodiment on InSb is consistent with the injection of carriers from the emitter into the region of the gap, with allowed transfer through the gap to the collector. These results are comparable with the dependence of the efficiency on the degree of doping of the gap material, since the degree of doping of the gap material determines the height of the barrier and the current strength. These results also show that the optimization of the thermal diode efficiency, in accordance with the present invention, using InSb can reach 5.5% at 600 K on the emitter and at an electron density on the emitter of 10 ²⁰ cm ^-3 .

The following is also a description with reference to compensation drawings used as a technology for improving efficiency in embodiments of the present invention. Compensation includes reverse current suppression. Methods of forming ohmic contacts in embodiments of the present invention are described below.

Examples of embodiments of the present invention that contain an InSb material with a compensation layer include InSb plates doped with an n-type impurity such as tellurium Te, and a tellurium emitter layer of Te implanted using magnetron sputtering technology. The compensation layer in these embodiments is formed by implantation of p-type impurities. Such p-type impurities contain at least one ion of a type such as argon or helium (Ar and He), which compensate for the n-type impurity.

Another material for constructing an n * / n type emitter in accordance with the present invention contains Hg _1-x Cd _x Te. For example, in embodiments according to the present invention, an Hg _0.86 Cd _0.14 Te plate is used to construct an n * / n emitter by reacting an Hg _0.86 Cd _0.14 Te with a substrate containing n-type impurities, such like aluminum and indium gallium (Al and In-Ga), this creates an n * electron injection region. One form of indium gallium (In-Ga) material used for this purpose is an In _0.75 Ca _0.25 alloy. Embodiments with such an emitter make it possible to obtain an output current density that rises as a function of the temperature of the hot side. It is shown below that in this case, in the embodiments, efficiency is achieved that exceeds 30% of the ideal efficiency of the Carnot cycle.

In other embodiments, embodiments of the present invention, part of a multi-plate or layered configuration is Hg _1-x Cd _x Te. For example, one embodiment of these layered structures, in accordance with the present invention, comprises an InSb plate doped with an n-type material such as tellurium (Te) and an InSb emitter layer coated with tellurium (Te) and coated with such material, such as In-Ga, more specifically, In _0.75 Ga _0.25 . The second plate in such a laminate contains Hg _1-x Cd _x Te, where x, in one embodiment, is 0.14.

Examples of embodiments in accordance with the present invention include the structural parameters of uncompensated thermal diodes, compensated thermal diodes and Schottky diodes. In addition, the converters in accordance with the present invention include converters of thermal energy to electricity, and embodiments for cooling. As described below, two embodiments include the same basic components, and they work either as thermionic converters for cooling, or as thermal diodes for converting thermal energy into electricity.

Moreover, in the context of the present invention, it was determined that the material Hg _1-x Cd _x Te, at x equal to from about 0.08 to about 0.15, exhibits a high value of thermoelectronic quality index, while remaining a semiconductor, and with the possibility using the design of the n * emitter layer / compensation layer, and exhibiting the characteristics described above with respect to other materials. In addition, in the context of the present invention, it has also been determined that Hg _0.92 Cd _0.18 Te exhibits excellent thermoelectric properties.

1. Solid state thermionic converter

The high doping region 14 n * in embodiments of the present invention can be used as an emitter region from which charge carriers can be expelled by heat to the gap region 16. The n * region contains a semiconductor doped with a high concentration of donor impurities (impurities that create an excess of electrons). For example, the InSb material can be doped with tellurium or sulfur (Te or S). Moreover, in the context of the present invention, it was determined that the energy conversion is a function of the normalized conductivity χ of the semiconductor, which, in turn, is a function of the material parameters and the peak doping of the emitter semiconductor material.

Appropriate material parameters were investigated in the context of the present invention to determine beneficial operating conditions for a number of semiconductors. This rating is presented below to illustrate how the appropriate semiconductor parameters were selected from a number of materials, and how this selection and rating can be extended within the scope of the present invention to the additional materials indicated below.

Table 1 provides a list of material parameters, including electron mobility and thermal conductivity, and associated estimates of χ. Estimates of the values for the normalized electrical conductivity χ were obtained using the concentration level N * _D = 10 ²⁰ cm ^-3 , and a wide range (almost four orders of magnitude) of possible values of the normalized electrical conductivity χ can be seen. The data for constructing this table are based on CRC Handbook of Chemistry and Physics, 67th edition. Materials with a small χ value are more preferred. According to this criterion, the preferred material among the semiconductors shown in Table 1 is mercury selenide, with a corresponding value of χ = 0.014. With such normalized electrical conductivity and a reduced potential of 5, the efficiency of the optimized barrier reaches approximately 13.3% and 23.8% for the ratio of the temperature differences of the emitter-collector to the emitter temperature (Δτ) of 0.3 and 0.5, respectively. These efficiency values are close to half the maximum values of thermodynamics.

TABLE 1 Semiconductor

χ Hgse 20000 0.010 0.014 Hgte 25,000 0,020 0,022 Insb 78000 0.160 0,057 CdSnAs ₂ 22000 0,070 0,089 InAs 33000 0.290 0.25 Pbte 1600 0,023 0.40 Pbse 1000 0.017 0.47 Zno 180 0.006 0.93 Pbs 600 0.023 1,07 Gaas 8800 0.370 1.18 Gasb 4000 0.270 1.89 Cdo one hundred 0.007 1.96 Ge 3800 0.640 4.72 Inp 4600 0,800 4.87 Si 1900 1,240 18.3 Gap 300 1,100 103 Bi _0.9 Sb _0.1 50,000 0,050 0,028 Se + Te _1-x 10,000 0,035 0,097 Bi (|| triangular axis) 35,000 0,053 0,042

Both mobility and thermal conductivity are generally functions of temperature. In InSb, both mobility and conductivity decrease with increasing temperature. The resulting temperature dependence of χ is shown in Figure 3. Figure 3 shows the normalized conductivity parameter χ (also called the "quality indicator") as a function of temperature for the InSb material, provided that N * _D = 10 ²⁰ cm ^-3 . The results are shown for regions of the n-type gap with doping densities N _D = 10 ¹⁵ , 10 ¹⁶ , 10 ¹⁷ , 10 ¹⁸ units cm ^-3 . As can be seen, the normalized conductivity decreases with increasing temperature. In addition, the normalized conductivity decreases in the presence of background charge carriers, since the electron mobility in InSb decreases with increasing doping density. More generally, Figure 5B shows the normalized conductivity χ for several semiconductors.

The semiconductors shown in Table 1 are materials that can be used in embodiments of the present invention. InSb is one of the materials among these semiconductors. As will be shown below, the Hg _1-x Cd _x Te material is another semiconductor and exhibits the property of normalized conductivity, which is approximately half the value for such an HgSe parameter.

In the context of the present invention, it has been determined that concentrations in the emitter and gap regions can be associated with an emitter-gap potential barrier. More specifically, it was determined that the potential barrier between the emitter and the doping concentration of the p-type gap is

и

представляют концентрации ионизированного донора и акцептора, n_i - плотность беспримесного носителя, и Т_max представляет максимальную температуру эмиттера.where Δu represents the emitter-gap potential barrier,

and

represent the concentration of the ionized donor and acceptor, n _i is the density of the pure carrier, and T _max represents the maximum temperature of the emitter.

The barrier in accordance with an embodiment of the present invention is up to about 7, preferably in the range from about 1.5 to about 7, and more preferably from about 3 to about 7. Unfortunately, this value is unattainable for InSb in the absence of independent gap displacement for the region of the p-type gap at temperatures close to room temperature.

In the case of the transition, the n * emitter is an n-type gap, it was determined that the barrier height is

The normalized value of the barrier height 5-7 corresponds to the doping level e ^{5th 7} , which in digital terms is 150-1100. If the n * region is doped to a level of 10 ²⁰ cm ^–3 , then the doping of the gap region should be in the range from 9 × 10 ¹⁶ cm ^–3 to 7 × 10 ¹⁷ cm ^–3 .

2. Results for InSb Thermal Diode

but. Charge carrier injection

As shown in figure 2, the emitter is a hot region 14 n * depicted on the left side. The gap region 16 is a wide region in the center, which can be either an n-type region or a p-type region (although it has been determined that a higher efficiency is achieved using the n-type gap). The collector is depicted here as the collector recombination region 18 and the cold metal contact 20. The initial premise of the embodiment presented here, in accordance with the present invention, is that the charge carriers are expelled by heat from the hot emitter region 14 to the gap region 16 where they transferred to the collector area 18 and contact 20.

This section describes issues related to the injection of charge carriers from the emitter into the gap region, with three different variations of the gap region type (p-type, pure semiconductor and n-type). The injection of electrons into the region of the p-type gap is a much simpler task for analysis, but there is a significant barrier that arises in the depletion region. Optimum efficiency is achieved when the barrier is of the order of 4 k _B T. The barrier between the n * emitter and the p-type gap approaches 8-9 k _B T. Therefore, a lower barrier is required to inject more charge carriers. Lower barriers are formed when using regions of the n-type gap with a moderate concentration, however, it should be understood that most of the charge carriers will be injected charge carriers.

In the case of thermal electron emission in the region of the p-type gap, in the context of the present invention, an npn transistor with bipolar junctions was used to show that such a description of the injection of charge carriers is in principle correct. Figure 4 shows the numerical solutions of charge emission from a hot n * -type emitter to the region of the p-type gap. It can be seen that the electrons enter the region of the gap and are screened by the majority of charge carriers in the form of holes, and the transfer of a smaller number of charge carriers mainly occurs due to diffusion. In this case, the holes act as a screen for the field in the gap region.

A significant decrease in the height of the barrier occurs when the gap region is made on the basis of pure semiconductor. Since the density of the injected charge carriers can be significant, an n-type gap region with minor doping with a donor number of 10 ¹⁴ cm ^{-3 is used} , which simulates the gap region based on an unalloyed semiconductor. Such a simulated gap region based on a pure semiconductor allows a lower potential barrier to be obtained than using a purely pure gap region. The results of the injection of charge carriers in this case are presented in Figure 4. Figure 4 shows the electron and hole concentrations for the n * pn * thermionic structure in InSb. The donor concentration in the emitter and collector regions is 10 ²⁰ cm ^-3 , and the acceptor concentration in the gap region is 10 ¹⁷ cm ^-3 . The emitter was heated to a temperature of 600 K, and the collector had a temperature of 300 K. Three cases are shown in the drawing: an open circuit i = 0 (equilibrium), in which carrier injection does not occur; short-circuit ν = 0 with maximum current; and the intermediate case when ν is equal to half the open circuit voltage. It can be seen that the electron and hole densities become almost equal, which implies maintaining a neutral charge level in the gap region, as far as this can be understood on a graph of this type. In this case, it was verified numerically that the injection of charge into the region of the pure semiconductor is possible, and that the electrons injected into the region of the gap can be transferred to the collector region.

The results of the injection of the charge carrier in this case are additionally shown in FIG. 5. FIG. 5 shows the electron concentrations for the n * nn * thermionic structure in InSb. The donor concentration in the emitter and collector regions is 10 ²⁰ cm ^-3 for this case, and the donor concentration in the gap region is 10 ¹⁴ cm ^-3 . The emitter is heated to a temperature of 600 K and the collector temperature is 300 K. The drawings show three cases: open circuit i = 0 (equilibrium state) without injection of charge carriers; short circuit ν = 0 at maximum current; and the intermediate case when ν is equal to half the voltage in the open circuit state. It can be seen that the densities of electrons and holes become almost equal, which implies that the neutral charge is maintained in the gap in the range that can be seen on this type of graph. In this case, it was confirmed numerically that the injection of charge into the region of the pure semiconductor is possible and that the electrons injected into the region of the gap can be transferred to the collector region without any difficulties.

A case will be described below when charge carriers are injected from the n * -type emitter region into the n-type gap region with moderate doping. The numerical results are shown in Figure 6. Figure 6 shows the electron and hole concentrations for the n * nn * thermionic structure in InSb. The donor concentration in the emitter and collector regions is 10 ²⁰ cm ⁻³ , and the donor concentration in the gap region is 8 × 10 ¹⁷ cm ⁻³ . The emitter was heated to 600 K, and the collector temperature was 300 K. Three cases are shown in the drawings: open circuit i = 0 (equilibrium state) without injection of the charge carrier; short circuit ν = 0 at maximum current; and the intermediate case when ν is equal to half the open circuit voltage. It was observed that electron injection occurs, and that the transfer is still, to a greater or lesser extent, a diffusion transfer.

The above results show that the injection of electrons from the n * emitter region into the gap region occurs, as expected, in the case of the p-type gap and in the gap region from the pure semiconductor and the n-type gap. It can be expected that injection into the region of the p-type gap occurs in accordance with the diode pattern. Heating the emitter region with respect to the collector leads to the thermal generation of EMF. The collector in this configuration plays the same role as the metallized contact in the diode. Therefore, the current can be calculated directly from the characteristics of the diode. However, when electrons are injected into the gap region from an unalloyed semiconductor and an n-type semiconductor, which will be described below, there is no reason to expect a priori that the diode pattern will be satisfied. Therefore, it is interesting to investigate whether the behavior of the diode type extends to new modes, and what kind of changes can be expected.

To conduct such a study, it is necessary to obtain the characteristics of the dependence of current on voltage, taking into account the fact that there is probably no way to regulate the voltage in the thermal energy converter. The height of the barrier can be changed by choosing the characteristics of the doping region of the gap. The figure 7 shows the normalized height of the barrier for the example described above, as a function of the degree of doping of the gap. Figure 7 shows the normalized barrier height Δu as a function of doping concentration. It is assumed that the emitter is doped in such a way that it contains 10 ²⁰ electrons / cm ³ . It is assumed that donors and acceptors in the gap are completely ionized. Using this result as a relationship between the degree of doping and the height of the barrier, the current of the short-circuited circuit was simulated as a function of the degree of doping of the gap, and the results were shown as a graph of the dependence of the current of the function on the height of the barrier. This is depicted in Figure 8. Figure 8 shows the short-circuit current as a function of the normalized barrier height for the structures described above at T _max = 600 K and the ratio of the temperature difference of the emitter-collector to the emitter temperature of 0.5. In this case, it was observed that the diode characteristics manifested themselves in the p-type mode, far from the pure semiconductor. You can also see that the behavior characteristic of the diode pattern extends to the n-type mode. This behavior corresponds to the assumption of the injection of charge carriers from the emitter into the region of the gap with the possibility of transfer through the gap to the collector. This makes it possible to numerically verify that the models developed for this system, based on the behavior of the diode characteristic, may be appropriate.

b. The ratio of current-voltage, power and efficiency

After establishing the main thermionic effect, we turn to the description of the main issue related to the optimization of efficiency.

The figure 9 presents curves showing the calculated values of the current and voltage characteristics for the example InSb material with a thickness of 625 μm described above, with an electron concentration of 10 ²⁰ electrons / cm ³ in the emitter and at an emitter temperature T _max = 600 K and a collector temperature T _min = 300 K. The results are shown for donor densities in the interval 10 ¹⁷ (the lowest curve on the graph), 3 × 10 ¹⁷ , 5 × 10 ¹⁷ and 8 × 10 ¹⁷ (the highest curve on the graph) in units of cm ^-3 . It was shown that the interdependences of current and voltage can be perfectly approximated by linear dependencies. In addition, it is noted, as defined above, that the electrons ejected from the emitter lead to an increase in negative current, as well as negative voltage when the circuit is broken. The figure 9 shows the values of both current and voltage.

The figure 10 shows the calculated power in the load per unit area for the example InSb material with a thickness of 625 μm described above, with an electron concentration of 10 ²⁰ electrons / cm ³ in the emitter, at a temperature T _max = 600 K and a collector temperature T _min = 300 K. The results are shown for donor densities in the range 10 ¹⁷ (the lowest curve on the graph), 3 × 10 ¹⁷ , 5 × 10 ¹⁷ and 7 × 10 ¹⁷ (the highest curve on the graph) in units of cm ^-3 .

The thermal power per unit area dissipated by the device for the conditions used in the previous examples is shown in Figure 11. Figure 11 shows the calculated thermal power dissipated through a unit area for the example 625 μm InSb material described above with an electron concentration of emitter 10 ²⁰ electrons / cm ³ , at an emitter temperature T _max = 600 K and a collector temperature T _min = 300 K. The results are shown for donor densities in the gap of 10 ¹⁷ (the lowest curve in the drawing), 3 × 10 ¹⁷ , 5 × 10 ¹⁷ and ^{17 October} 7 × (the vertices curve gives the graph) in units of cm ^-3. It can be seen that the power consists of a constant component, in accordance with Fick's law of diffusion, at zero current (with an open circuit voltage), and a term that is linearly dependent on the current (and, therefore, linearly with respect to voltage).

Figure 12 shows the efficiency as a function of voltage for the InSb example described above. Shown here are the results of calculations for the donor densities of the gap of 10 ¹⁷ (the lowest curve in the graph), 3 × 10 ¹⁷ , 5 × 10 ¹⁷ and 7 × 10 ¹⁷ (the highest curve in the graph) in units of cm ^-3 . The dots indicate the efficiency values at the optimum points.

The results presented above show that the efficiency depends on the degree of doping of the gap, since the degree of doping of the gap determines the height of the barrier and the corresponding magnitude of the current. Doping with a donor of the gap at a level of approximately 7 × 10 ¹⁷ cm ^-3 is apparently a value close to optimal. For an additional study of such optimization, FIG. 13 shows the results of numerical optimization of efficiency as a function of the degree of gap doping in a wide range of doping densities at a fixed concentration of ionized emitter impurity of 10 ²⁰ cm ^-3 . If the degree of doping of the gap is close to the pure semiconductor, the efficiency will be low. This, in the end, is due to the fact that, correspondingly, a high barrier is formed at the emitter-gap transition. Accordingly, embodiments with a doping degree of the gap close to the pure semiconductor at which a high emitter-gap transition barrier is formed have a low efficiency. If the degree of gap doping becomes too high, the positive effect of the low emitter-gap transition barrier is compensated by the negative influence of the reverse ohmic current. This issue is further discussed below.

from. Scale Study

The design efficiency of a thermal diode is a function of the degree of doping of the emitter. It is assumed that the thickness of the material is 625 μm, the degree of doping of the gap is maintained at 7 × 10 ¹⁷ cm ^-3 , and the temperature of the emitter and collector are assumed to be 600 K and 300 K, respectively. The results are shown in figure 14 for optimal efficiency as a function of the degree of emitter doping for a fixed gap doping degree N _D = 7 × 10 ¹⁷ cm ^-3 , while it can be seen that the efficiency monotonically increases with the emitter electron concentration, but to a lesser extent than according to linear law. This is due to two effects: the emitter-gap transition barrier increases at higher values of the degree of doping of the emitter, since the degree of doping of the gap is maintained at a fixed level; and electron mobility decreases with a higher concentration of charge carriers. Both of these effects together reduce the positive effect of a greater degree of doping of the emitter.

It is possible to implant tellurium Te (which is the donor with the lowest level of ionization energy) into the emitter at concentrations of the order of 10 ²⁰ cm ^-31 . Modeling using TRIM-91 codes shows that such a high density of the dopant leads to the formation of an amorphous emitter layer. Such a layer will have a gap with a different width, a different effective mass and mobility than in the models above. In addition, it can be expected that the degree of recombination will be very high. Some consequences of this can be predicted. The injection of electrons into the gap will be limited by the available emitter densities of the order of one recombination length into the emitter when measured from the side of the gap. This will occur at spatial scales of the order of the mean free path of an electron.

As a result, a large region of an amorphous emitter with a relatively short length of about 1 micron is very likely to represent an effective doping density for the gap, which will be much lower than the peak density obtained at the edge of the semiconductor. On the other hand, a sharp n * profile can lead to the formation of a free electron flow from an amorphous region with a high degree of doping to the crystalline intermediate region. The latter dependence may more closely correspond to the assumptions underlying the model described below.

Another important point is that the ionization at the donor levels used in the emitter is likely to be incomplete. If the emitter region were crystalline at high doping levels, the corresponding density of the state conduction band could not be especially large, as a result, the donor ionization balance would probably create favorable conditions for the substantial use of donors. Data for the donor’s ionization energy are available (tellurium Te in InSb probably has a donor ionization energy of the order of 50 meV); therefore, part of the ionization fraction can be estimated. The use of a metal contact with a low work function of the emitter electrons can lead to related problems, since thermionic injection from a metal can be quite high.

Depending on the model used to model the effectiveness of the embodiments in accordance with the present invention, it can be concluded that the optimum efficiency should be independent of the gap length or this independence should be practically maintained for the gap thickness in the range from 200 μm to 2 mm. It was determined that the thermal power should be proportional to the reciprocal of the gap thickness in any model described in the context of the present invention, as shown in figure 15 for a thermal diode.

In the above description, an InSb substrate with a total thickness of 625 μm was used. In the absence of a recombination effect, thick layers will be preferred for applications with a large T _max value, since the associated heat flux may be correspondingly smaller. The electron recombination length in the mass of n-type InSb crystalline material for the described gap densities is at least 10 times greater than the thickness of the investigated substrate. In addition, radiation recombination predominates in general recombination, which is, strictly speaking, the radiation enclosed in large crystals, such as those considered here. As a result of this, structures of the order of 1-2 mm should be of interest for use in energy conversion.

The effect of recombination on the efficiency of the device under conditions when recombination becomes an important factor has not been investigated. Although the effect of recombination can, in general, be considered as a universally negative effect for thermionic energy conversion, it is likely that its consideration should probably not be left in the background, in the absence of simulation results. The reason for this is that recombination increases the injection current above the level of the diffusion limit considered here. Although heat loss is inversely proportional to the length of the gap, the current will be inversely proportional to the length of the recombination. It can be assumed that, within these limits, a net result will be a net increase in efficiency. This can be assumed based on the fact that the required hole current to maintain recombination will lead to a corresponding potential drop. Since the mobility of holes in the InSb material is small, the potential drop associated with this is likely to be significant.

The figure 16 shows the efficiency of the design with an electron emitter concentration of 10 ²⁰ electrons / cm ³ and a donor density in the gap of 7 × 10 ¹⁷ cm ^-3 . The results are shown as a function of the emitter temperature, provided that the collector temperature is 300 K. In two classes, very close results were obtained. The optimized efficiency as part of the thermodynamic limit is shown in Figure 17. It can be seen that such a design works to a greater or lesser extent in the same way as when compared with the thermodynamic limit for all temperatures presented.

At a lower temperature difference, the flow of thermal energy will be less for this design. The figure 18 shows the flow of thermal energy under conditions of optimal energy conversion for structures with a thickness of 626 microns and 1250 microns at different temperatures. The flow of thermal energy is in the range of several hundred W / cm ² under optimal operating conditions of interest.

d. Short description

In the previous description, a thermal “diode” model based on an embodiment using InSb material was considered. This device uses an emitter region with a high degree of doping, a gap region, which can be either p-type or n-type, and an ohmic metal collector with a sufficiently large work function, set in such a way that it has an insignificant thermionic injection current compared to with an ohmic component, due to the equilibrium of the charge carriers at the contact point of the collector.

The above results show that in order to obtain the highest thermionic injection current, the emitter-gap barrier is preferably of the order of 4-5 k _B T _max , which implies that the gap should preferably be an n-type semiconductor. Therefore, the present invention contemplates embodiments in which the emitter-gap barrier is preferably in the range of from about 4 k _B T _max to about 5 k _B T _max , as in other embodiments in which this barrier is outside this range, but which can be developed based on the description given here.

As shown above, the region n * allows electrons to be injected into the region of the n-type gap, and transport in the gap region is carried out, to a greater or lesser extent, due to diffusion. In addition, it was also shown above that a thermal diode can operate as an energy converter based on thermionic emission from an emitter into a gap, followed by transfer to a collector. It has also been shown above how to optimize embodiments according to the present invention as a function of the concentration of the gap donor. The optimal thermal diode efficiency can be 5.5% at an emitter temperature of 600 K, provided that an electron density of 10 ²⁰ cm ^-3 can be formed in the emitter. These parameters are characteristic of the embodiments in accordance with the present invention, although embodiments with other characteristics that differ from this particular optimization option and which are designed as described herein are also considered to be within the scope of the present invention.

3. Results for a compensated thermal InSb diode

The efficiency of thermionic energy conversion in such embodiments as schematically shown in FIG. 2 is ultimately limited by the presence of an ohmic reverse current arising from the thermoelectric response of the semiconductor. If this reverse current could be suppressed, a substantial increase in efficiency could be obtained. This section shows that when the reverse current is reduced, it is possible to increase the efficiency by approximately half.

One of the schemes designed to reduce the reverse current includes compensation of the n-type substrate by doping of the p-type to obtain a layer close to the pure semiconductor before the collector contact in the embodiments according to the present invention, while the supply of available electrons is significantly reduced that could initiate an ohmic reverse current. With this approach, a certain compromise is obviously used, since too large p-type compensation can limit the thermionic current in the collector.

The work performed, in the context of the present invention, shows that there is a small window in the parameter space in which the compensated layer can be matched with the gap doping in order to simultaneously provide approximately free thermionic current and practically zero ohmic reverse current. The resulting designs have a very high calculated energy conversion efficiency and can be competitive with the best thermoelectric devices. Since there are other semiconductors with better mobility to thermal conductivity ratios (such as HgSe and HgTe), a compensated diode circuit may have the potential to achieve conversion efficiencies that exceed the best performance for thermoelectric devices. As will be shown below (see, for example, the description of FIG. 38), compensation layers developed in the context of the present invention can achieve significant performance improvements for materials such as InSb. This improvement is reflected in an increase in efficiency of approximately 2 times. For Hg _1-x Cd _x Te, such an improvement translates into an efficiency that approaches the absolute limit of 50% of the ideal efficiency of the Carnot cycle.

Since these and other embodiments of compensated diodes with such semiconductors exhibiting high mobility to thermal conductivity ratios can be developed by those skilled in the art using the descriptions and examples given in the present description, they are considered to be within the scope of the present invention.

The basic structure of the thermal diode as an embodiment with InSb material was considered as an energy converter. It was determined that the efficiency of the device is limited at a level slightly exceeding 10% of the thermodynamic limit. Compensation is as follows. As indicated above, it is possible to increase the efficiency by suppressing the ohmic reverse current, which has the opposite sign to the thermionic current. One of the options for this is to use p-type doping to obtain a compensated layer on the inner side of the collector contact, which prevents the injection of electrons from the collector side of the device (see figure 19). Figure 19 schematically shows an embodiment of a compensated thermal diode in accordance with the present invention. The emitter is a hot region of 14 n * located on the left. The gap region 16 is represented by a wide n-type region located in the center. The collector is depicted here as a cold metal contact 20. The hot ohmic contact 12 is located in close proximity to the hot region 14 n *. On the inner side of the metal contact, a compensated region 19 is formed, made by adding p-type impurities, which suppresses the reverse electron current. Adding p-type doping allows one to obtain a p-type semiconductor layer, which, if it is not precisely matched with the doping of the substrate, can resist the thermionic electron flux to the collector. It was determined that in the parameter space there is a small region around the ideal value of compensation, in which a number of design parameters can be selected, which allows for the simultaneous supply of thermionic current to the collector and compensation of the ohmic reverse current. It was determined that the design efficiency of such a device is significantly increased compared with the main structure of the diode. This section describes such a device and its related features.

but. Reverse current

Before moving on to the present description, it is necessary to establish the presence of reverse current in the main thermal diode. One way to achieve this is to determine the characteristics of the current, including its sign, as a function of the concentration of the donor gap. The result is shown in Figure 20. Figure 20 shows the current as a function of the degree of doping of the gap for an example of the InSb thermal diode design described in the last section. For the thermionic mode, which corresponds to a negative current (electrons move from left to right), the current is calculated under conditions of optimal efficiency. For the thermoelectric mode, which corresponds to a positive current (electrons move from the collector to the emitter), the current is obtained under conditions when the voltage is half the heat-induced EMF, which is obtained when an approximately optimal thermoelectronic mode is provided. In this case, a change in the sign of the heat-induced current is observed. With low doping of the gap, the electrical conductivity of the gap is insufficient to maintain a high level of the ohmic component, which could compete with thermionic injection. With an increase in the degree of doping of the gap at a certain point, the ohmic current exceeds the thermionic current in magnitude, and then thermionic energy conversion will no longer be possible. Such an ohmic reverse current is formed due to free electrons, which are formed on the collector side of the device and are transferred to the emitter primarily by drift.

b. Efficiency optimization using a blocking layer

, составляющих 7×10¹⁷, 10¹⁸, 2×10¹⁸ и 3×10¹⁸ см^-3. Линией из точек показан полученный КПД в отсутствие блокирующего слоя. Максимальный КПД получается, когда концентрация акцептора слоя компенсации регулируется таким образом, чтобы она соответствовала концентрации донора подложки. При этом получают существенное повышение оптимального КПД по сравнению со случаем отсутствия компенсации.Next, we consider the alleged operation of the compensated thermal diode, which includes an internal blocking layer. In this model, a compensation layer is modeled using p-type Gaussian doping with a characteristic length of 5 μm. The results for different acceptor densities are shown in Figure 21. Figure 21 shows the optimization of efficiency as a function of doping the gap with p-type compensation using concentrations

constituting 7 × 10 ¹⁷ , 10 ¹⁸ , 2 × 10 ¹⁸ and 3 × 10 ¹⁸ cm ^-3 . A line of dots shows the obtained efficiency in the absence of a blocking layer. The maximum efficiency is obtained when the concentration of the acceptor of the compensation layer is adjusted so that it corresponds to the concentration of the donor substrate. At the same time, a significant increase in optimal efficiency is obtained compared with the case of no compensation.

The shape of the efficiency curves shown in Figure 21 can be qualitatively represented using a simple analysis. The efficiency is maximized under the conditions of an internal blocking layer, which at the same time allows the transmission of thermionic current from the emitter with minimal formation of reverse current. At a lower concentration of the gap donor, the compensation layer forms a p-type region, which in this application works to a greater or lesser extent as a diode with reverse bias, which prevents the flow of thermionic current. At a higher donor concentration in the gap, the compensation will be insufficient to eliminate excess electrons. In this case, a reverse current begins, the strength of which approximately linearly depends on the concentration of electrons in the blocking layer. Therefore, a linear decrease in efficiency from the optimal value can be observed.

From a practical point of view, the possibility of transmitting electric current through the terminals limits the current density to 10 ² -10 ³ A / cm ² . If these values are exceeded, the voltage drop in the conductors becomes unacceptable. In addition, a temperature drop of 200-300 ° C passes through the diode. For the InSb thermal conductivity coefficient, this value can be converted into a gap thickness, which should be approximately 1 cm.This thickness creates a problem associated with the recombination length, which becomes comparable to the gap thickness, and in addition, technological problems associated with polishing thick substrates arise. For example, most substrate processing equipment is designed for thicknesses of less than 1 mm.

Typically, wide gaps are obtained by installing the diodes in a row. Since the current through the series-installed diodes in a row will be the same, the series-installed diodes must be matched by current. If one of the diodes generates a larger current, this will lead to an increase in the voltage drop across the other diodes and to a deterioration in the overall characteristic due to additional potential barriers.

The following description is intended to provide current matching, which may be embodied in the context of the present invention.

(1) Current matching using gap doping concentration

In the example below, it is assumed that the material of the InSb diode is used, all diodes having the same structure, and the temperature of the heat source is 530 K and the heat sink temperature is 460 K. The configuration with one diode is shown in Figure 22. The arrow in Figure 22 indicates that the temperature T of the hot ohmic contact 12 is higher than the temperature of the collector 20. Based on the graph shown in Figure 23A, and when the efficiency level is set to 6%, four diodes can be installed in a row, as shown in Figure 24, where the first diode (D ₁ ) has k the concentration of the doping gap is 5 × 10 ¹⁷ cm ^-3 , D ₂ ~ 7 × 10 ¹⁷ cm ^-3 , D ₃ ~ 10 ¹⁸ cm ^-3 and D ₄ ~ 2 × 10 ¹⁸ cm ^-3 . In this embodiment, all four diodes generate the same current, and the overall efficiency is 6%. Although some embodiments of composite diodes in accordance with the present invention comprise diodes whose respective elements are made of the same materials for each diode, it should be understood that embodiments based on composite diodes, in the context of the present invention, are not limited to such series. Some embodiments of the composite diodes, in accordance with the present invention, contain diodes, the corresponding elements of which are made of various materials. For example, in some embodiments based on composite diodes, emitters in different diodes contain different materials, and / or gaps in different composite diodes contain different materials, and / or collectors in different composite diodes contain different materials.

(2) Coordination of current by selection of geometric dimensions

In the example below, the line of highest efficiency is used according to figure 23A, which corresponds to a donor concentration of 2 × 10 ¹⁸ cm ^-3 . A thin diode in a series row at a temperature of 500 K allows one to obtain a current that is approximately two times greater than the current obtained in a diode that is not installed in a row. The coordination of the current is achieved if the diode has a temperature of 400 K and the area is two times larger than the area of the diode, not installed in a row. The results of this embodiment are shown in FIG. 25, where a change in area leads to the formation of a wedge-shaped row of diodes. The dial boundary is not really a straight line, but rather a curve that takes into account the non-linear behavior of the efficiency curve shown in Figure 23A. The kit shown in FIG. 26 represents another embodiment where the border of the kit approximately corresponds to the ideal curve shown in FIG. 25. The arrows in FIGS. 25-26 indicate an increase in temperature T from low temperature (T _COLD ) to high temperature T of the _{hot side} (T _HOT ).

The design of the compensated thermal diode has been optimized to work with maximum efficiency with a hot emitter of 600 K. It is interesting to determine the efficiency of the device at other emitter temperatures. The numerical results for the efficiency are shown in figure 23A for various degrees of doping of the substrate, subject to ideal compensation. Figure 2 ZA shows the efficiency under optimized conditions as a function of emitter temperature for various degrees of gap doping with ideal compensation. It is assumed that the structure is used compensated thermal diode InSb with a density of 625 microns thick electron emitter October ²⁰ cm ^-3 and a temperature of 300 K. The collector gap concentrations of donor and acceptor concentration agreed up (to improve efficiency in accordance with schedule) 7 × ^{17 October,} 10 ¹⁸ , 2 × 10 ¹⁸ and 3 × 10 ¹⁸ cm ^-3 . A line of dots represents the resulting efficiency in the absence of a blocking layer. The results for the efficiency, normalized to the thermodynamic limit, are shown in figure 23B. Figure 23B shows the efficiency normalized to the thermodynamic limit under optimized conditions as a function of emitter temperature for various cases shown in figure 23A. In this case, it can be observed that the compensation layer operates efficiently at a high emitter temperature. In addition, optimization at high temperature, apparently, allows to obtain a relative optimum at other temperatures so that it is not necessary to use separate structures optimized for different temperature conditions. More advanced designs that work best around the temperature calculated for them, and do not provide the same efficient operation at other temperatures, will be described below.

from. Examples

Figures 27A-27B show efficiency as a function of temperature for optimized embodiments of compensated thermal diodes in accordance with the present invention. The curves in figures 27A-27B are indicated in accordance with the material of the gap, and the numbers in square brackets represent the concentration of charge carriers. Figure 27B shows the relative efficiency with respect to the Carnot cycle.

The compensated layer in the n-type semiconductor can be made, including, but not limited to, by introducing acceptors. For InSb doped with Te tellurium (donor impurity), the ionization energy of the donor is 50 meV. The same ionization energy is characteristic of acceptors created by vacancies. A compensated layer exists if the number of vacancies corresponds to the initial concentration (n) of the donor.

The number of vacancies is determined by the dose of ions per unit area D (ions / cm ² ) and the number of vacancies created by one V ion if vacancies are induced by implantation of an inert gas ion. V is a function of the energy E of the ion, V = V (E). The number of vacancies is N _v (E) = D · V (E). In this case, the concentration of the vacancy depends on the depth of the ion content in the material R, which, in turn, is a function of the ion energy R = R (E). Thus, for a given type of ion, the concentration of vacancies as a function of ion energy represents

For the compensated layer, C (E) = n, or

R (E) and V (E) were modeled using computer code TRIM-91 for InSb material and Ne, Ar, and Xe ions. The simulation results are shown in figures 28 and 29. The dose required to create a compensated layer in a wide range of ion energies in an n-type InSb diode doped to a concentration of n can be determined using figures 28 and 29 and the equation for D above. In figures 34 and 35, respectively, lines 32 and 38 represent the dependences for argon, lines 34 and 36 for neon, and lines 30 and 40 for xenon.

Ion implantation creates a vacancy concentration profile that is more pronounced in the last 20-30% of the ion content depth. These 20-30% of the ion content depth can be reduced to a value smaller than the tunneling distance in InSb, which is usually 100-150 Å, to exclude the possibility of the formation of additional barriers.

The choice of ions used requires a balance of pros and cons, such as the number of thin layers to be created, relative to the degree of damage to the structure of the solid. For example, xenon (Xe) allows you to get thinner layers, but leads to a greater degree of damage, and the result is a semi-amorphous crystal.

d. Short description

Modification of the design of the main thermal diode, presented in the last section, was described in the form of a search for the possibility of reducing the reverse ohmic current, which limits the maximum efficiency of the device. It was demonstrated that ohmic reverse current is present in the calculations, and that it prevails over thermionic current with a high degree of gap doping. As it was shown, the addition of a compensated layer on the inner side of the collector leads to a decrease in the reverse current, which allows to achieve a higher optimal efficiency. The simulation results show that ideal compensation allows you to achieve the highest levels of efficiency at a given donor concentration in the gap. This conclusion corresponds to an a priori analysis based on simple physical arguments. Therefore, preferred embodiments of compensated thermal diodes in accordance with the present invention are characterized by high efficiency, which means that a greater degree of compensation leads to a higher efficiency. The peak efficiency calculated for the compensated thermal diode is comparable with the characteristics of the best thermoelectric devices.

4. Ohmic contact

An ohmic contact is defined as a metal-semiconductor contact, which has a negligible contact resistance with respect to the volume or distributed resistivity of the semiconductor. (See publication of Sze, S. M., Physics of Semiconductor Devices, New York, John Wiley & Sons, 1981, pp. 304-311, the entire contents of which are incorporated herein by reference). This section describes ohmic contacts and methods for forming such contacts in accordance with the present invention.

Metal-semiconductor surfaces introduce local potential barriers, commonly known as the Schottky barrier. In a simplified form, the height φ _{b of the} Schottky barrier, measured with respect to the Fermi level, can be written as φ _b = φ _m −χ _s , where φ _m represents the electron work function of the metal and χ _s represents the electron affinity of the semiconductor. Examples of Schottky barrier values are 0.70 eV for GaAs and 0.18 eV for InSb.

For a metal-semiconductor solid-state thermionic converter, the operating voltage range will be lower than the height of the Schottky barrier. This ratio makes it impossible to obtain the desired effect, or at least to reduce operating currents.

Typical operating voltage in accordance with the present invention is 10-100 mV and power is 1-10 watts. This allows you to get a working current I ₀ = 100-1000 A, while the energy loss at the Schottky barrier is W _loss = I _o φ _b . In order for the value of W _{loss to} be less than 1% of the total power, the value of φ _b must be less than 1 meV. The barrier is often expressed as contact resistance. Therefore, with steady currents, the contact resistance should be less than 10 ^-5 -10 ^-6 Ohms.

Published by Chang et al. Specific Contact Resistance of Metal-Semiconductor Barriers. Solid-State Electronics, Volume 14 (1971), pp. 541-550, and Shannon, JM, Control of Schottky Barrier Height Using Highly Doped Surface Layers. Solid-State Electronics. Volume 19 (1976), pp. 537-543, which are incorporated herein by reference, provides a method for forming an ohmic contact. The electric field at the metal-semiconductor section creates a region depleted in charge carriers in the semiconductor. With an increase in the concentration of ionized impurities, the width of the depletion region becomes narrower. This, in turn, leads to an increase in the transmission coefficient for tunneling. Consequently, even a high barrier contact can become ohmic if the barrier, in turn, is so thin that tunneling dominates in the process of carrier transfer. The doping level corresponding to 10 ^-6 Ohm / cm ² is 10 ²⁰ -10 ²¹ (Te in InSb) at a temperature of 300 ° C. The effective electron mass for tunneling increases with temperature, and at a temperature of 500 ° C the required concentration is more than 10 ²¹ than 10 ²⁰ . A layer with a high doping concentration should be sufficiently thin so that it does not introduce its own barrier at the interface with the semiconductor. In the author’s publication above, Shannon estimates this thickness as less than 150 Å. This approach is applicable to both n-type alloying and p-type alloying, taking into account the fact that the current sign will be opposite when passing from the n-type region to the p-type region.

The required implantation dose for thin doping of 10 ²¹ cm ^-3 was calculated using the computer code TRIM-91 (G. Ziegler, G. Biersack. IBM (1991)). The ion penetration depth and the required dose were calculated separately for In and Sb. The calculation results were averaged to approximate InSb. The difference between both In and Sb in this energy range did not exceed 20%. Tellurium Te was used as an n-type impurity, since tellurium Te has the lowest ionization energy (50 meV). Figure 30 shows the calculation results for this dose, while figure 31 shows the depth of penetration of ions.

For p-type alloying, known materials for InSb include Ge germanium (ionization energy 9 meV) and Ag silver (E _i = 30 meV). Ag silver is obviously the preferred ion because it is heavier than Ge germanium and has a lower penetration depth for the same ion energies, which allows a smaller depletion region to be formed. The results of calculations for silver alloying with Ag are presented in figures 32 and 33. In the process of ion implantation, vacancies are created, which subsequently must be processed by annealing.

Another way of forming an ohmic contact is by diffusion annealing. For example, an ohmic contact for an InSb-containing diode can be formed by annealing indium layers on an InSb substrate. The following procedure was performed in a quartz ampoule, purified acid. The ampoule was annealed for more than one hour under high vacuum at a temperature of 800 ° C. IndS coated InSb samples were loaded into quartz ampoules, from which gas was pumped out and which were filled with helium under a pressure of 10-100 Torr. High thermal conductivity helium provides rapid cooling. After annealing at different temperatures, the dependence of current on voltage on the samples was measured to confirm that ohmic contacts exist. Positive results were obtained in the temperature range of 250-400 ° C with annealing time of 10-60 minutes. At temperatures exceeding 500 ° C, indium was completely dissolved, which made it impossible to use samples, even if such samples showed ohmic characteristics.

5. Examples

but. Construction Parameters

As shown in figure 2, intermediate heat-conducting layers can be placed in other embodiments of the present invention between the ohmic contacts (12, 20) and the heat sink to provide thermal contact. For example, a deposited layer of indium In or the like can be used on the hot side, and a deposited layer of a eutectic In-Ga or similar material can be used on the cold side. These materials are sufficiently fluid to provide adequate thermal contact at low pressure (0.1-1.0 MPa).

Accordingly, the materials that can be used for these layers according to the present invention are heat-conducting materials with good flowability, although other materials may also be used in other embodiments. Another way to ensure thermal contact is to apply a paste, glue, fusible solder or their equivalents. An electrical and heat-conducting layer is then added, which serves as a barrier to diffusion between the heat-conducting layer and the semiconductor. In the present embodiment, the thermally conductive and electrically conductive layer is used as an emitter without an additional semiconductor emitter layer. The characteristics and functions of this layer include the following indicators: (1) conducts heat; (2) conduct electricity; (3) emits electrons; (4) creates a Schottky barrier at the metal-semiconductor interface; (5) creates a barrier to diffusion; (6) prevents the chemical reaction of a semiconductor with a subsequent layer; (7) harmonizes the thermal expansion of the semiconductor to prevent delamination; (8) is thermally stable in the range of operating conditions of the thermal diode; and (9) has high oxidation stability if it is not intended to be placed in a vacuum device or enclosed in an inert environment.

For example, the InSb material has a thermal expansion coefficient of 5.2-5.4 × 10 ^-6 K in the temperature range of 300-500 K. Other possible materials include, without limitation, molybdenum Mo, chromium Cr, tungsten W, tantalum Ta, rhenium Re, osmium Os, iridium Ir, lanthanides, as well as alloys of nickel, platinum Pt and soft metals such as indium In, gold Au, copper Cu or similar metals. From this list, tantalum Ta and lanthanides are susceptible to oxidation, and indium In has a low melting point.

High alloyed semiconductors and semimetals can also be used. For example, a thin layer of silicon Si has a sufficiently high thermal conductivity and electrical conductivity. In this case, however, certain precautions must be fulfilled, and in particular, it should be noted that a large forbidden zone in comparison with InSb leads to the formation of an internal barrier that prevents the flow of current due to the transfer of charge carriers.

The thickness of the heat-conducting and conductive layer is determined as follows. The thermal conductivity should preferably be higher than that of the semiconductor gap. With a gap thickness of 100-1000 microns, the layer thickness should preferably be less than about a few microns, since a thick layer increases heat loss. On the thin side, the layer thickness is determined taking into account several conditions. For example, the metal layer should preferably be thicker than the mean free path of the electron to maintain its bulk properties. Since the layer is located in close proximity to another metal (intermediate layer), it can affect the position of the Fermi level and change the electron emission in the semiconductor. It is known that this effect is significant in a metal layer with a thickness of less than 1000 Å. This figure is at least a few average electron mean free paths and can be considered as a lower practical limit to prevent unnecessary complications. Similar thickness conditions apply to the n * region of the semiconductor emitter.

For the emitter-gap region, a situation is preferred when the region has matched crystallography, that is, when the emitter region has been epitaxially extended over the gap region. For InSb material this can be obtained by maintaining the deposition temperature above 150 ° C (PVD - vacuum-plasma deposition). For other gap materials, such as Hg _1-x Cd _x Te, epitaxial extension is more difficult. In case of inconsistency at the emitter-to-interface interface, dissipation and degradation of the converter performance occurs.

Specialists in the field of electrical and thermal contacts are aware of many ways to ensure adequate electrical and thermal contact, and the scope of the present invention is not limited to the above examples, but also provides other embodiments developed in accordance with other criteria.

b. Uncompensated Thermal Diodes

The following examples are intended to illustrate individual embodiments of the present invention and are not limiting. The present invention can be embodied in other specific forms, without departing from its essence or essential characteristics. The concentration of the dopant in the following examples (1) to (9) is given in units of cm ⁻³ .

(1) metal ₁ / n / metal ₂ : eutectic InGa (in mass) / Cr or Ni (1000-4000 Å) / InSb (360 microns; doped 1.1 × 10 ¹⁸ Te, orientation 100) / Pt (1500 Å ) / In (in bulk). The thickness of metal ₁ metal ₁ cannot be less than the average mean free path of electrons for a given metal at a specific temperature, for example, for Ag gold, the average mean free path is approximately 400 Å.

(2) metal ₁ / n * / n / metal ₂ : eutectic InGa (in bulk) / Cr (1500 Å) / InSb (400 Å; doped 3.0 × 10 ¹⁹ Te) / InSb (360 microns; doped 1, 1 × 10 ¹⁸ Te) / Pt (1500 Å) / In (in mass).

(3) metal ₁ / n ** / n * / n / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / InSb (400 Å; doped 3.0 × 10 ¹⁹ Te) / InSb (360 microns; doped from 1.1 × 10 ¹⁸ Te) / In (100 Å) / Pt (1500 Å) / In (in mass).

(4) metal ₁ / n ** / n * / n / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 3.0 × 10 ²⁰ Te) / InSb (360 microns; doped 1.1 × 10 ¹⁸ Te) / In (100 Å) / Pt (1500 Å) / In (in mass).

(5) metal ₁ / n ** / n / n ** / metal ₂ : eutectic InGa (in bulk) / Cr (1500 Å) / In (100 Å) / InSb (360 microns; doped 1.1 × 10 ¹⁸ Te) / In (100 Å) / Pt (1500 Å) / In (in mass).

(6) metal ₁ / n ** / n / n ** / metal ₂ : eutectic InGa (mass) / Cr (1500 Å) / In (100 Å) / InSb (500 microns; doped 1.1 × 10 ¹⁸ Those) / In (100 Å) / Pt (1500 Å) / In (in mass).

(7) metal ₁ / n ** / n * / n / n ** / metal ₂ ; eutectic InGa (by mass) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 3 × 10 ²⁰ Te) / InSb (500 microns; doped 1.1 × 10 ¹⁸ Te) / In (100 Å) / Ni (1500 Å) / (in mass).

(8) metal ₁ / n ** / n / n ** / metal ₂ : eutectic InGa (in bulk) / Cr (1500 Å) / In (100 Å) / InSb (500 microns; doped 1.9 × 10 ¹⁷ Those) / In (100 Å) / Pt (1500 Å) / In (in mass).

(9) metal ₁ / n ** / n / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / In (100 Å) / InSb (500 microns; doped 1.9 × 10 ¹⁷ Those) / In (100 Å) / Ni (1500 Å) / In (in mass).

from. Compensated Thermal Diodes

The following examples are intended to illustrate selected embodiments of the present invention and are not limiting. The present invention can be embodied in other specific forms without departing from its essence or essential characteristics. Layers designated as having a low doping level (p) may also be n-type layers. The concentration of impurities in the following examples (1) to (5) is given in units of cm ^-3 .

(1) metal ₁ / n ** / n * / n / p / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 3 , 0 × 10 ¹⁹ Te) / InSb (500 microns; doped 1 × 10 ¹⁸ Te; 2 ° from (100)) / p-InSb (400 Å; doped 3.1 × 10 ¹⁴ Te) / In (100 Å) / Pt (1500 Å) / In (mass).

(2) metal ₁ / n ** / n * / n / p / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 3 , 0 × 10 ¹⁹ Te) / InSb (500 microns; doped 1 × 10 ²⁰ Te; 2 ° from (100)) / p-InSb (400 Å; doped 3.1 × 10 ¹⁴ Te) / In (100 Å / pt (1500 Å / In (mass).

(3) metal ₁ / n ** / n / p / n ** / metal ₂ : eutectic InGa (mass) / Cr (1500 Å) / In (100 Å) / InSb (500 microns; doped 1 × 10 ¹⁸ Te; 2 ° from (100)) / p-InSb (400 Å; doped 3.1 × 10 ¹⁴ Te) / In (100 Å) / Pt (1500 Å) / In (in bulk).

(4) metal ₁ / ** / n * / n / p / n ** / metal ₂ : eutectic InGa (in mass) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 3, 0 × 10 ¹⁹ Te) / InSb (500 microns; doped 1 × 10 ²⁰ Te; 2 ° from (100)) / P-InSb (2000 Å; doped 3.1 × 10 ¹⁴ Te) / In (100 Å / Pt (1500 Å) / In (by mass).

(5) metal ₁ / n ** / n * / n * / p / n ** / metal ₂ : eutectic InGa (in bulk) / Cr (1500 Å) / In (100 Å) / InSb (400 Å; doped 1.0 × 10 ²⁰ Te) / InSb (500 microns; doped 1 × 10 ¹⁸ Te) / p-InSb (400 Å; where Ar or Ne) / In (100 Å) / Pt ions were implanted in the p-type region (1500 Å).

6. Distributed Schottky Diode

The formation of the Schottky barrier at the metal-semiconductor interface is well described in the publication by Rhoderick, E.N. and Williams, RH, Metal-Semiconductor Contacts. Oxford, Clarendon Press (1988), which is incorporated herein by reference. There are two main models of the Schottky barrier. For the Schottky-Mott model, the barrier φ _{in is} considered as the difference between the work function φ _{m of} the metal yield and the degree of χ _s electron electron affinity of the semiconductor; φ _in = φ _m -χ _s . In fact, φ _in is practically independent of the work function φ _{m of the} metal. The explanation given by the author J. Bardeen is that static charges on the surface influence the barrier. The compensation of this charge is affected by the static charge Q _ss at the surface. And the electrically neutral condition is represented by the following expression Q _m + Q _d + Q _ss = 0, where Q _m represents the negative charge on the metal surface and Q _d represents the positive charge of uncompensated donors. The properties of the compensation barrier φ ₀ (neutral level) depend on the relative position φ ₀ and the Fermi level E _F. If φ _{0 is} measured with respect to the top of the valence band, the Schottky barrier is expressed as follows: φ _in ≈E _g -φ ₀ .

The energy E _{g of the} gap is a function of temperature and, to some extent, the concentration of the impurity. The accumulation of contamination on the surface affects φ _in , as described in section 4 in relation to ohmic contacts. Partial accumulation of contamination on the surface also affects the height of the barrier.

Figure 34 shows the temperature characteristic of the InSb gap (see Landolt-Börnstein, Numerical Data and Functional Relationships in Science and Technology, Group III: Crystal and Solid State Physics, (1983) Volume 22b, the entire contents of which are incorporated herein by reference) .

The values of the Schottky barrier can be determined by changing the slope of the curve obtained from external measurements of the dependence of current on voltage. At room temperature, the barrier height is 175–180 meV, regardless of the impurity concentration (Te) in InSb, up to 10 ²⁰ cm ^–3 (in contact). Figure 35 shows the barrier height as a function of temperature for a layer at the 2000 Å thick interface doped with tellurium (Te) to a concentration of 3 × 10 ¹⁹ cm ^-3 deposited on InSb doped with tellurium (Te) concentration up to 1 × 10 ¹⁸ cm ^-3 (500 μm) for an emitter from indium (In). Since the height of the barrier decreases with increasing temperature at a faster rate than E _g , this means that the neutral level φ ₀ will be higher than E _F , and the surface charge density increases with increasing temperature. In figures 34 and 35 presents the value of φ ₀ , the assessment of which should be carried out at 15-20 meV and a temperature of approximately 300 ° C. This type of barrier is shown in Figure 36A. The insulating film (oxide) shown in Figure 36A is made so thin that the charge carriers pass through the tunnel effect through it so that practically this barrier is completely invisible, even if it is present. Moreover, in the context of the present invention, it was determined that a diode of this type operates at elevated operating temperatures, in accordance with the present invention.

but. Experimental results

The samples were made on the basis of InSb substrates doped with Te tellurium up to 1 × 10 ¹⁸ cm ^-3 . The thickness of the substrate was approximately 500 microns, and the substrate was polished on both sides. After standard cleaning, a 2000 Å thick emitter layer of InSb doped with Te tellurium to a concentration of 3 × 10 ¹⁹ cm ^-3 was deposited on a substrate by magnetron sputtering. The samples were in the form of squares with sizes ranging from 1 × 1 to 3 × 3 mm ² , which were obtained using eutectic InGa (T _m = 35 ° C) on both sides. The process of applying the image was the application of some pressure to destroy the surface oxide layer.

36B schematically shows an embodiment of the present invention comprising a hot ohmic contact 12, an emitter 14, a gap region 16, a compensated region 19, and a collector 20. A region 15 has been formed in some embodiments on an emitter side that faces the hot ohmic contact 12, to reduce the barrier at the metal-semiconductor interface. Such a barrier reducing layer at the metal-semiconductor interface is formed in some embodiments by magnetron sputtering. Region 17 is formed in some embodiments on the side of the gap region, which faces the cold metal contact 20 of the collector, and is designed to reduce the barrier at the metal-semiconductor interface. This region is formed using technologies similar to the technology used to form region 15. Other embodiments of the present invention contain only region 15, and in still other embodiments of the present invention, only region 17 is used. As noted above, the presence of at least one of regions 15 and 17 in embodiments of the present invention, including compensated and uncompensated embodiments, increases operating temperatures.

The test device included a 400 W heating element placed in a massive silver-plated copper block and a water-cooled cold plate (silver-plated copper) mounted on a micrometric linear platform. The electrical leads were massive flexible copper strips (<10 ^-4 ohms). Temperature control was carried out using an Omega RTD device with a Keithly 2001 display. A custom-made bank of resistors represented a load of 10 ^-4 Ohms or higher. The voltage was measured with an accuracy of 0.01% and the current with an accuracy of 1%. Samples were mounted on a hot plate and pressed against a cold plate on a linear installation platform. Argon gas was introduced between the plates to prevent oxidation of the materials at elevated temperatures. The hot side was insulated from the mounting plate and ambient air.

The figure 37 shows an example of a curve of the dependence of the output current on voltage for one sample, indicated by number 42, and a set composed of three samples, the dependence curve of which is indicated by number 44, at an emitter temperature of 200 ° C. At the point of maximum received power, the output difference between them was less than 20% with a decrease in the heat flux by at least three times. This means a significant increase in efficiency in the configuration with a set of samples. In addition, each interface introduces thermal resistance due to imperfect contact and the effect of phonon mismatch. The minimum numbers for phonon mismatch were approximately 4% (See Swartz, E.T., Thermal Boundary Resistance, Vol. 61, No. 3 (July 1989), which is incorporated herein by reference). Each sample introduces two additional boundaries.

Changing the orientation of the sample in the set so that the emitter layer is on the cold side reduces the yield in the set of 5 samples by about 5 times at a temperature of 200 ° C, and by about 2 times at a temperature of 300 ° C, compared with a correctly constructed set. At 300 ° C and recalculation based on the thermal conductivity of InSb and the electrical output, in some cases the efficiency was better than 25% of the ideal Carnot cycle with a specific output power of 3-8 W / cm ² .

7. Additional examples

This section describes the characteristics of the test device, sample preparation technology, and more specific results regarding embodiments using InSb and / or Hg-Cd-Te based material.

but. Testing device and sample preparation

Testing devices were developed based on standard mechanical parts for use in laser technology, including the Coherent® stainless steel breadboard. A micrometric linear platform and a laser optical stand made it possible to linearly move vertically at a distance of 100 mm.

The hot side was mounted on a Macor ceramic ring on a linear platform and consisted of a massive copper block with a 400 W Ogden Scientific heating element. The copper block was thermally insulated using porous ZrO ₂ ceramics and fiberglass. Interchangeable copper rods made of oxygen-free copper, coated with silver 2 microns thick, were used to transfer heat to the sample. Each rod contained at least two holes configured to install temperature sensors. Based on temperature measurements at two points along the rod and knowing the thermal conductivity and the cross section of the rod, the heat flux to the sample was determined.

A water-cooled silver-plated cold plate was installed over the optical stand with a three-axis Newport ball and socket platform, which made it possible to combine cold and hot plates in parallel.

The findings for electric current were made of silver-plated intertwined copper wires with a resistance of approximately 10 ^-4 ohms. The load resistors in the range of about 10 ^-5 ohms to about 10 ^-1 ohms were made of copper and stainless steel and connected to the current leads with massive bolts.

Power was transferred to the heater from a constant voltage source of the Xantrex 300-3.5 type. The load voltage and sample resistance were measured using a NIST34420A NIST nanovolt / microohmmeter in a configuration with 4 conductors. Keithley 2001 multimeters were used as a reader for Omega thermocouples and RTD temperature sensors. Electric current was measured using an Amprobe® A-1000 transducer. Based on the known values of the load resistances and conclusions, an independent current determination was carried out. For all measured parameters, with the exception of currents below 1 A, the accuracy was better than 1%.

To prevent oxidation of the sample and contacts at elevated temperatures, argon gas was supplied between the hot and cold plates using a flexible Capton type curtain.

Material for sample preparation was prepared in the form of plates from InSb (WaferTech, UK) with a diameter of approximately 2 inches (5.08 cm) and a thickness of 500 μm. The plates were polished on both sides to a rms roughness value of approximately 20 Å. The standard concentration of tellurium impurities (Te) was approximately 10 ¹⁸ cm ^-3 . The emitter layer was deposited by magnetron sputtering. InSb doped with 3 × 10 ¹⁹ cm ^-3 tellurium (Te) was also used. The thickness of the emitter layer was selected in the range from about 400 Å to about 15,000 Å. The emitter thickness in the embodiments in accordance with the present invention was at least about 400 Å. In addition, the principles in the context of the present invention do not impose any restrictions on the thickness of the emitter and, therefore, embodiments of the present invention do not have a top limit with respect to such thickness.

To create a compensation layer of approximately 10 ¹⁸ cm ^{-3, a} p-type impurity was placed in a semiconductor to compensate for an existing n-type tellurium (Te) impurity with a concentration of approximately 10 ¹⁸ cm ^-3 . Vacancies in InSb are formed by p-type charge carriers with an ionization energy of approximately 60 meV (see, for example, Landolt-Bornstein's publication), which is approximately equal to the tellurium ionization energy Te. The implantation doses for forming the compensation layer were recounted using the TRIM-91 software.

The samples were then cleaned and digested to fit in an implantation unit. Samples were implanted with Ar ions at 40 keV (Core Systems, Inc., Santa Clara, California) at various doses. Each sample was tested in conversion mode at a temperature of approximately 200 ° C. The test results are shown in figure 38 together with the calculated doses required for the compensated diode. Doses varied because the concentration of impurities in the substrates was known with an accuracy of approximately 10%. The zero implant dose in FIG. 38 corresponds to the sample without compensation, and implant doses greater than zero in FIG. 38 refer to compensated samples. As shown in figure 38, a comparison of the obtained maximum efficiency for compensated samples with the efficiency of an uncompensated sample made it possible to determine that the compensation layer can improve performance by approximately 80%. For comparison, Fig.38 also shows the calculated efficiency, which was predicted for a given implantation dose.

The penetration depth for Ar ⁺ at 40 keV in InSb is approximately 400 Å, which is sufficient to create a compensation layer. The 400 Å layer is susceptible to losses due to the rapid diffusion of vacancies at elevated temperatures. To prevent such diffusion loss, in other embodiments, implantation with He ion is performed. The thickness of the He ion layer in these embodiments is of the order of several microns, which increases the effective life of the implanted layer. For example, an estimate of the half-life of InSb vacancy diffusion at a thickness of 1 micron is approximately 1 year at 200 ° C. Due to the fact that the compensation layer is located on the cold side in the embodiments of the present invention, diffusion problems are usually prevented when the compensation layer has a thickness of several microns. The calculated ion penetration depths and vacancy formation for ⁴ He ions in InSb are shown in figures 39-40.

b. Embodiments with Hg _1-x Cd _x Te

Hg _1-x Cd _x Te semiconductors (referred to herein as “MCT”) have very good thermionic values of the quality index when 0.08≤x≤0.15, where the upper and lower bounds are given approximately. A preferred x value is about 0.14. In embodiments of the present invention, a 500 micron thick Hg _0.86 Cd _0.14 Te plate (manufactured by Lockheed Martin IR Imaging Systems) was used. MCT reacts with various substrates to form heavily doped donor layers (interacts with metals such as indium In, iron Fe, gallium Ga, and aluminum aluminum) or acceptor layers (interacts with metals such as Ag silver, Au gold, and Bi bismuth). moreover, the reaction rate depends on the material and temperature. See the publication by P. Caper, Properties of Narrow Gap Cadmium-based Compounds, INSPEC 1994, which is incorporated herein by reference.

The reactivity of MCTs makes it easier to build n * / n emitter layers than when using InSb, since InSb has a lower degree of reactivity and requires more sophisticated technology to create an n * region. In addition, the use of InSb is limited by impurity concentrations of about 2-3 x 10 ¹⁹ cm ^-3 .

The performance properties of the embodiments of the present invention show that substrates that form donor impurities are preferred because they allow the generation of higher current densities. As shown in FIG. 41, a thermoelectric response without an injection layer of a charge carrier constitutes a current density that exhibits a slight change or no change with respect to temperature. For example, copper forms an acceptor impurity and should not form an n * region. In contrast, substrates containing aluminum Al, indium In and gallium Ga form n-type impurities in the MCT and they create electron injection regions n *. Figure 41 shows the dependence of the electric current density as a function of temperature for Hg samples of _0.86 Cd _0.14 Te, one of them was used with an emitter layer of copper Cu and the other with an emitter layer of In-Ga with an In _0.75 Ga substrate composition _0.25 . Contact resistance was monitored in both cases to ensure that oxide layers did not significantly affect the observed results. In particular, it was determined that In-Ga allows a slightly better contact than copper (approximately 92 mOhm for In-Ga compared to approximately 103 mOhm for copper). Figure 41 shows the electric current density as a function of temperature for a sample with copper, having an almost horizontal dependence. The MCT sample was left to cool and an In-Ga layer about 20-50 microns thick was placed on top of the copper substrate. As shown in figure 41, the electric current density in this case showed a change in temperature, which was similar to the dependence for the sample with copper only at temperatures up to approximately 70 ° C. The same figure shows that above this point the electric current density clearly increased with increasing temperature. This is due to the fact that the acceptor-type impurity was suppressed by n-type impurities, which led to the fact that the sample showed a mode of injection of charge carriers with a many times higher output current. The output voltage in both cases was approximately the same value from approximately 290 to approximately 350 μV / K, which corresponds to the Seebeck thermoelectric effect coefficient for MCT.

With different donor materials, different electric current densities were obtained. Figure 42 shows the dependence of the electric current density as a function of temperature for two samples of Hg _0.86 Cd _0.14 Te, one of them with a substrate of aluminum Al and the other sample with a substrate of In-Ga alloy. A preferred composition of these substrates has been embodied as In _0.75 Ga _0.25 . The In-Ga substrate allows the formation of a better emitter than aluminum Al, since the electric current density as a function of temperature for the In-Ga sample was correspondingly higher in the entire temperature range. Although this is not shown in the form of graphs of the dependence of the electric current density on temperature, indium In allows one to form a better emitter than gallium Ga, in particular, when used with a pure In substrate. Substrates of aluminum Al, indium In and gallium Ga are examples that form n-type impurities in MCTs that create n * electron injection regions.

The figure 43 shows the absolute efficiency shown in the embodiments of the MCT converter in accordance with the present invention, in which the n * emitter layer was formed by the reaction of the MCT with a eutectic In _0.75 Ga _0.25 . Absolute efficiency is defined as the ratio of the output electric power to the heat flux through the sample. The same data as shown in Figure 43 was recalculated as a percentage of the ideal Carnot cycle efficiency, and the conversion results are shown in Figure 44. The efficiency η _{from the} ideal Carnot cycle is defined as η _c = (T of the _{hot side} -T of the _{cold side} ) / T _{hot side} . It is assumed that a level above 30% of the ideal efficiency of the Carnot cycle is beyond the capabilities of any conventional thermoelectric device, and that conventional thermoelectric devices hardly reach the level of 20% of the ideal efficiency of the Carnot cycle. In contrast, FIG. 44 shows that the embodiments of the present invention continuously operate with an efficiency exceeding 20% of the ideal Carnot cycle efficiency in a temperature range from about 100 ° C. to about 175 ° C. at temperature T _{cold side} = 20.5 ° C, and with an efficiency above 30% of the ideal efficiency of the Carnot cycle at temperatures from about 150 ° C to about 160 ° C.

The decrease in the operating characteristics of the converter beyond the temperature range of 150-160 ° C can be explained by dissolution of the emitter layer and / or recombination of the injected charge carriers. To eliminate such a reduction in converter performance, embodiments of the present invention comprise a diffusion barrier. A ytterbium layer up to about 10 Å thick is an example of such a diffusion barrier. A thickness of up to about 10 Å is preferred since such a metal layer does not significantly affect the electron transfer properties. Other methods for constructing a diffusion barrier are described, for example, in an article by A. Raisanen et al. in Properties of II-VI Semiconductors, MRS Society Symposium Procceedings, Volume 161, pp. 297-302, 1990, which is incorporated herein by reference.

from. Embodiments with the InSb / Hg _1-x Cd _x Te Multilayer Package

As shown above, embodiments according to the present invention, which comprise a set of InSb plates in a configuration with an emitter on the hot side, exhibit significantly better efficiency. The efficiency for embodiments of this type was determined as follows.

InSb and MCT show the best performance at different temperatures: from about 300 ° C to about 350 ° C for InSb and about 150 ° C for MCT. Given this temperature difference, embodiments of the transducers in accordance with the present invention have been optimized for both materials.

The low thermal conductivity of the MCT makes it difficult to conduct direct measurements of the heat flux, especially when measurements should be carried out on small samples. In addition, the dimensions of some samples in the used embodiments in accordance with the present invention were, for the most part, a few square millimeters and due to such reduced sizes they were not allowed to conduct contact temperature measurements using available temperature sensors. In addition, the small size of these samples did not allow the use of standard infrared image cameras due to the limited spatial resolution of such infrared image cameras. Technology based on the use of infrared cameras with specially built optics has eliminated this problem.

Another methodology that was adopted in these experiments implied the following assumptions. The same heat flux passes through plates with the same thickness, so there is essentially no heat loss. The total temperature drop across the entire set is written as ΔT = ΔT ₁ + ΔT _2, where ΔT ₁ represents the temperature drop on the first plate, and ΔT ₂ represents the temperature drop on the second plate. The values of thermal conductivity, depending on the temperature of InSb and MCT, were presented as λ ₁ (T) and λ ₂ (T), respectively. Using these variables, we can describe the following system of equations:

-λ ₁ (T) ΔT ₁ = -λ ₂ (T) ΔT ₂

ΔT = ΔT ₁ + ΔT ₂

The ΔT value can be measured as the temperature difference between the cold and hot plates, and the first equation can be iterated using the values of ΔT ₁ , λ ₁ (T) and λ ₂ (T). The heat flux and temperature drop on each plate are evaluated in accordance with this iteration procedure. As noted above, the efficiency of the converter is calculated based on the ratio of the output electric power to the heat flux through the device. The thickness of the InSb plate was adjusted to change the range of the operating temperature of the converter from a value of less than 150 ° C to more than 300 ° C with essentially the same value of the component of the Carnot cycle efficiency exceeding 30%. Direct measurements using infrared imaging equipment showed a slightly lower heat flux through the converter, probably due to imperfect contacts, which led to a 3-4% increase in efficiency. The maximum observed efficiency in the embodiments of the present invention exceeded 40% of the ideal Carnot cycle. In contrast, the conventional energy converters currently available usually provide an efficiency of approximately 16% of the efficiency of the Carnot cycle in a narrow temperature range.

Figure 45 shows the efficiency of an embodiment with a multilayer converter in accordance with the present invention. In this embodiment, an InSb plate with a thickness of approximately 1 mm was used, and the impurity concentration (Te) was approximately 10 ¹⁸ cm ⁻³ . The emitter layer had a thickness of approximately 2000 Å and was a sputtered InSb layer with a concentration of approximately 3 × 10 ¹⁹ cm ^-3 Te. The plate was coated with an In-Ga containing layer. A preferred composition of such an In-Ga material in this embodiment was In _0.75 Ga _0.25 . The thickness of this layer ranged from about 30 microns to about 50 microns. The second plate was made of Hg _1-x Cd _x Te, with a value of x preferably satisfying the inequality 0.08 ≤ x ≤ 0.15, where the upper and lower boundaries are given approximately, a more preferred form of this composition has an approximate stoichiometry given by Hg _0.86 Cd _0.14 Te with a thickness of approximately 0.51 mm. The average cross section of the kit was approximately 1.70 × 1.52 mm ² . The fraction of efficiency from the ideal Carnot cycle as a function of the temperature of the hot plate for this embodiment is shown in Figure 46. The _{cold side} T value in Figures 45-46 was 20 ° C. As shown in FIG. 46, the percentage of the efficiency of the ideal Carnot cycle for this embodiment with the maximum performance was approximately the same as that shown in FIG. 44, but this embodiment displays it at a significantly higher temperature.

d. Quality Score Hg _1-x Cd _x Te

As shown in Table 1, the quality score for HgTe is approximately 2.5 times better than for InSb. The addition of cadmium Cd to HgTe improves the mobility of charge carriers and reduces thermal conductivity. The figure 47 shows the normalized dependence for the value of the quality indicator for Hg _1-x Cd _x Te with respect to InSb, as a function of x. For x = 0.08, the quality score for Hg _0.92 Cd _0.08 Te is approximately 0.0065, which is better than the quality score for HgTe by about 2 times. In other embodiments of the present invention, it is assumed that the quality score is at a level as low as approximately 0.001.

In the context of the present invention, it was also determined that the preferred embodiments of Hg _1-x Cd _x Te have a value of x in the range from about 0.08 to about 0.15 and that it is in this area that this material exhibits the highest values for its thermionic quality index , and while the material still exhibits the properties of a semiconductor. This maintains the interdependence between the n * emitter layer and the gap, and the compensation layer, and the gap described above.

8. Embodiments for cooling

The main components of the thermionic Converter 50 to provide cooling (see figures 48 and 49) are essentially the same as for the thermal diode 10, designed to convert heat into electricity, which is described above (see figures 1 and 19). Accordingly, the term "solid-state thermionic electronic thermal energy converter" generally refers to embodiments of thermal energy to electricity converters in accordance with the present invention, and to cooling embodiments in accordance with the present invention.

The figure 48 presents an uncompensated thermal diode and figure 49 shows a compensated thermal diode. The significant difference between the embodiments with the conversion of heat to electricity and the embodiments with cooling is that the transfer of charge carriers is carried out using an external electric field E _Ext , and the n * -type region 14 is connected to the heat load, which is cooled by the heat flux to the first ohmic contact 52 in the region of 14 n * -type. The n * -type region 14 is thermally insulated with an insulating material 54. Unlike the heated n * -type region 14, in the case of the embodiment with the conversion of heat to electricity, the heat load is cooled by a heat flux, Q _Loud to the 14 n * - region type in the thermal diode 50, as shown in FIG. 48. The region of the gap 16 is located in close proximity to the n * -type region 14, and the second ohmic contact 53 has a collector recombination region 56 formed between the second ohmic contact 53 and the gap region 16. The gap region 16 may be an n-type region, a p-type region or a pure semiconductor region. For the case of a compensated thermal diode, as shown in FIG. 49, the compensated region 19 is located on the inner side of the metal contact and is created by adding p-type impurities, which suppresses the reverse electron current. The rear surface of the second ohmic contact 53 acts as a heat exchanger, and the heat flux (Q _Exchange ) dissipates heat from the hot electrons.

The figure 50 shows the efficiency (efficiency, relative to the reverse Carnot cycle) as a function of temperature for compensated diodes in the embodiments with cooling, in accordance with the present invention. The efficiency curves shown in FIG. 50 are labeled for different spacing values in each embodiment. In addition to significantly increased efficiency compared to conventional devices, FIG. 50 also shows embodiments, in accordance with the present invention, that operate at temperatures well below 200 K, in contrast to conventional devices that generally cannot work at temperatures below about 200 K.

For specialists in the field of heat exchangers it is clear that there are many means for implementing heat transfer, including, but not limited to, air and liquid cooling or their equivalents.

9. Industrial applicability

Since energy conversion is the foundation of modern civilization, an efficient energy converter can have many applications, such as existing power plants designed for power supply, solar power plants, power sources for residential buildings, electricity sources for domestic consumption / solar sources of electricity, cars, marine vessels, solar / marine power plants, portable electronic devices, energy converting heat pumps environment, cooling systems (cooling, air conditioning, etc.), aerospace technology, etc.

A huge amount of waste heat with a potential of 300 ° C and below is released at power plants. Conversion of waste heat with a Carnot efficiency of 20-40% when using embodiments of the present invention involves an additional increase in the overall efficiency of the power plant by 10-20% with equivalent fuel economy.

The proliferation of low cost energy converters in accordance with the present invention can lead to lower capital costs in the construction of power plants with a solar energy concentrator with higher efficiency than the currently used steam / electricity cycles. Lower operating temperatures also reduce operating costs.

Sources of electricity for domestic consumption, based on the direct conversion of heat into electrical energy, are ideal for remote areas where it is difficult or inconvenient to run power lines. The heat source may be either in the form of fossil fuels or in the form of solar energy concentrators. Concentrators of solar energy can also be made on the basis of water pools heated from the sun, using the temperature difference of the day / night. Several hundred cubic meters of water with one hundred square meters of surface, together with the embodiments of the present invention, can provide electricity to the house in areas with a temperature difference of about 10 ° C.

The thermal diode, in accordance with the present invention, in combination with a conventional engine driving an electric generator and an electric motor, can significantly increase the distance traveled per unit fuel consumption.

Direct energy conversion has a huge number of applications in electric cars. One embodiment involves the use of thermionic devices, in accordance with the present invention, with operating temperatures from about 150 to 200 ° C, as devices for increasing overall efficiency. Another embodiment may be used in an electric vehicle with a conventional engine coupled to an electric generator with a plurality of converters according to the present invention used as an intermediate radiator.

Variants of application in automotive and translational driving techniques can also be used at sea. In addition, solar power concentrators can be used on sailing ships. The combination of lightweight and low-cost plastic Fresnel lenses with thermal diode converters in accordance with the present invention can be mounted on modern rigid wing-type sails, providing the use of wind and solar energy to drive a ship with a solar energy component of approximately 100-200 W / m ² sails.

Since the embodiments of the converter in accordance with the present invention allow the use of very small temperature gradients in a self-sustaining mode, the temperature gradient between the heat sinks can be created using an asymmetric heat exchanger on the surface (for example, one of the heat sinks can be insulated). In addition, the system will work until something breaks down, cooling the environment and generating electricity. In general, the method and apparatus described in the present invention represent a significant improvement over the prior art in the field of energy conversion.

The present invention can be embodied in other specific forms without departing from its essence or essential characteristics. The described embodiments should be considered in all respects only as an illustration, and not as a limitation.

Claims (63)

1. Solid-state thermionic converter containing an emitter, including at least a region containing a first donor with a concentration

the collector and the region of the gap between the emitter and the collector, which is in electrical and thermal connection with the emitter and the collector, and containing a semiconductor, including a second donor with a concentration of N _d , chosen in such a way that the natural logarithm of the ratio

is a value in the range of 0 to 7. 2. The Converter according to claim 1, characterized in that it further comprises a compensated region located between the gap region and the collector and having such a configuration that suppresses electric current from the collector to the gap region. 3. The Converter according to claim 1, characterized in that the value of the natural logarithm of the ratio

is in the range of 3 to 7. 4. The Converter according to claim 1, characterized in that the emitter temperature is higher than the collector temperature when electric current flows from the emitter to the collector. 5. The Converter according to claim 1, characterized in that the emitter contains metal. 6. The Converter according to claim 1, characterized in that the gap region contains an n-type semiconductor. 7. The Converter according to claim 1, characterized in that it further comprises a recombination region located in an electrical connection between the gap region and the collector, or comprising a collector portion connected to the gap region. 8. The Converter according to claim 1, characterized in that the emitter contains InSb. 9. The Converter according to claim 1, characterized in that the emitter contains InSb doped with tellurium Te. 10. The Converter according to claim 1, characterized in that the emitter contains InSb doped with tellurium Te with a concentration in the range of essentially from 10 ¹⁶ cm ^-3 to 3 · 10 ¹⁹ cm ^-3 . 11. The Converter according to claim 1, characterized in that the emitter contains InSb doped with tellurium Te with a concentration in the range of essentially from 10 ¹⁸ cm ^-3 to 3 · 10 ¹⁹ cm ^-3 . 12. The Converter according to claim 1, characterized in that the gap region contains InSb doped with tellurium Te with a concentration of essentially 10 ¹⁸ cm ^-3 . 13. The Converter according to claim 1, characterized in that the thickness of the emitter is at least 400 Å. 14. The Converter according to claim 1, characterized in that the gap region contains a semiconductor for which the dimensionless value of the normalized conductivity χ is in the range of essentially from 1 to 0.001. 15. The Converter according to claim 1, characterized in that the gap region contains HgSe. 16. The Converter according to claim 1, characterized in that the gap region contains HgTe. 17. The Converter according to claim 1, characterized in that the gap region contains Bi _1-yr Sb _y , while y is in the range of essentially from 0.05 to 0.2. 18. The Converter according to claim 1, characterized in that the gap region contains Se _z Te _1-z , where z satisfies the inequality 0≤z≤1. 19. The Converter according to claim 1, characterized in that the region of the gap contains Hg _ix Cd _x Te, while x is in the range of essentially from 0.08 to 0.2. 20. The Converter according to claim 1, characterized in that the region of the gap contains Hg _1-x Cd _x Te, while x is essentially 0.08. 21. The Converter according to claim 1, characterized in that the gap region contains a doped semiconductor with an impurity concentration in the range of essentially from 10 ¹⁵ cm ^-3 to 10 ²⁰ cm ^-3 . 22. The Converter according to claim 1, characterized in that the energy barrier for injecting electrons from the emitter into the gap region is in the range of essentially from 4 k _B T to 5 k _B T, where k _B is the Boltzmann constant, and T represents the absolute temperature at which electron injection occurs. 23. The Converter according to claim 1, characterized in that the emitter is insulated. 24. The Converter according to claim 1, characterized in that it further comprises a first ohmic contact in electrical and thermal connection with the emitter; a layer for reducing the barrier at the metal-semiconductor interface between the first ohmic contact and the emitter, and the second ohmic contact in electrical communication with the collector. 25. The Converter according to paragraph 24, wherein the collector is formed on the second ohmic contact. 26. The Converter according to paragraph 24, characterized in that it further comprises a heat conducting layer deposited at least on one of the first and second ohmic contacts. 27. The Converter according to claim 1, characterized in that it further comprises a cold ohmic contact in electrical and thermal connection with the gap region and containing a collector located adjacent to the gap region, wherein the collector includes a collector recombination region and a compensated region located between the gap region and the collector configured in such a way that it suppresses electric current from the collector to the gap region. 28. The Converter according to item 27, wherein the region of the recombination of the collector is formed on a cold ohmic contact. 29. The Converter according to claim 1, characterized in that it further comprises a compensated region, wherein the gap region is located between the emitter and the compensated region, and the collector is in electrical and thermal contact with the compensated region having p-type doping, so that the electric current from the collector to the gap region is substantially suppressed while allowing thermionic current to flow from the gap region to the collector. 30. The Converter according to clause 29, wherein the emitter temperature is higher than the collector temperature when an electric current flows between the emitter and the collector. 31. A solid-state thermionic electronic converter comprising a plurality of individual converters installed in series, in which each of the individual converters is made in accordance with paragraph 29. 32. The Converter according to clause 29, wherein the compensated region is formed by implanting ions into the gap region. 33. The Converter according to clause 29, wherein the compensated region contains vacancies created by implanting ions. 34. The Converter according to clause 29, wherein the emitter is thermally insulated. 35. Solid-state thermionic thermal energy Converter containing an emitter, including at least the reaction product of Hg _1-x Cd _x Te with a substrate containing indium In; the collector and the region of the gap between the emitter and the collector, in electrical and thermal connection with the emitter and the collector, containing a semiconductor selected from the group consisting of n-type semiconductors, p-type semiconductors and pure semiconductors. 36. The Converter according to claim 35, characterized in that it further comprises a compensated region located between the gap region and the collector and whose configuration suppresses electric current from the collector to the gap region. 37. The Converter according to clause 35, wherein the substrate contains In-Ga. 38. The Converter according to clause 35, wherein the value of x is in the range of essentially from 0.08 to 0.25. 39. The Converter according to clause 35, wherein the value of x is in the range of essentially from 0.08 to 0.09. 40. The Converter according to clause 35, wherein the substrate contains In _1-w Ga _w , in which the value of w is in the range from about 0.1 to about 0.3. 41. The Converter according to claim 35, characterized in that a diffusion barrier is provided on the emitter. 42. The converter according to claim 35, wherein the emitter has a diffusion barrier containing ytterbium. 43. The Converter according to claim 35, wherein the emitter is thermally insulated. 44. A solid-state thermionic thermal energy converter containing a plurality of plates P _i , at 1≤i≤m, where m represents the total number of plates, each of which contains an emitter E _i having at least a region containing a first donor having a concentration

; the collector C _i and the region G _{i of the} gap between the emitter E _i and the collector C _i , in electrical and thermal connection with the emitter Ei and collector C _i , and containing a semiconductor including a second donor having a concentration of N _d chosen in such a way that natural the logarithm of the ratio Nd * / Nd is in the range of numerical values from 0 to 7; wherein each plate P _j having an emitter E _{j + 1} , a gap region G _{j + 1} and a collector C _{j + 1} has such a configuration that it is connected in series with a group of emitter E _j , a gap region G _j and a collector C _j , for 1≤j≤ (m-1), and the indices i and j are integers, and the collector C _j is in electrical communication with the emitter E _{j + 1} for each value of j satisfying the inequality 1≤j≤ (m-1 ) 45. The Converter according to item 44, wherein the value of the natural logarithm of the ratio

is in the range of essentially from 3 to 7. 46. The Converter according to item 44, characterized in that it further comprises a compensated region R _i located between the region G _{i of the} gap and the collector C _i and has such a configuration that it suppresses electric current from the collector to the gap region, with each plate P _j , including the emitter E _{j + 1} , the region G _{j + 1 of the} gap, the compensated region R _{j + 1} and the collector C _{j + 1 are} made in such a way that it is connected in series with the group of emitter E _j , the region G _{j of the} gap, the compensated region R _j and C _j to manifold values 1≤j≤ (m-1). 47. The Converter according to item 46, wherein the emitters E _i and E _j contain essentially the same materials, the collectors C _i and C _j contain essentially the same materials, and compensated areas R _i and R _j contain essentially the same materials for i ≠ j, and 1≤i≤m, 1≤j≤m. 48. The Converter according to item 44, wherein the temperature of each of the emitters E _{i is} higher than the temperature of each of the collectors C _i when an electric current flows between the emitter E _i and the collector C _i . 49. The Converter according to item 44, wherein the first plate P ₁ contains InSb materials. 50. The Converter according to item 44, wherein the first plate P ₁ contains InSb material doped with tellurium Te. 51. The Converter according to item 44, wherein the first plate P ₁ contains InSb material doped with tellurium Te at a concentration of essentially 10 ¹⁸ cm ^-3 . 52. The Converter according to item 44, wherein at least the emitter E _{1 of the} first plate contains InSb material doped with tellurium Te. 53. The Converter according to item 44, wherein at least the emitter E _{1 of the} first plate contains InSb material doped with tellurium Te at a concentration of essentially 3 · 10 ¹⁹ cm ^-3 . 54. The Converter according to item 44, wherein the first plate P ₁ is coated with a material containing In-Ga. 55. The Converter according to item 44, wherein at least the first plate P ₁ contains a coating of a material containing In _1-u Ga _u , in which the value u is in the range of essentially from 0 to 0.3 . 56. The Converter according to item 44, wherein at least the first plate P ₁ contains a coating of a material containing In _1-u Ga _u , in which the value of u is essentially 0.25. 57. The Converter according to item 44, wherein at least one of the plates contains Hg _1-x Cd _x Te with a value of x, which is in the range of essentially from 0.08 to 0.2. 58. The Converter according to item 44, wherein at least one of the plates contains Hg _1-x Cd _x Te when the value of x is in the range of essentially from 0.08 to 0.14. 59. The Converter according to item 44, wherein the first emitter E ₁ is thermally insulated. 60. A method of cooling using a solid-state thermionic converter, comprising creating an external electric potential difference on a thermionic converter containing a thermally insulated emitter, including at least a region having a first concentration of Nd * donor; the collector, the region of the gap between the emitter and the collector, in electrical and thermal connection with the emitter and the collector, and containing a semiconductor including a second donor having a concentration of N _d selected in such a way that the value of the natural logarithm of the ratio Nd * / Nd is in the range from 0 to 7; and connecting the heat load to the emitter so that the heat load is cooled by the heat flux as the applied difference in electric potentials creates an electric current between the emitter and the collector. 61. The method according to p. 60, characterized in that they maintain the temperature of the emitter in the range of essentially from 20 to 400 ° C. 62. The method according to p. 60, characterized in that the value of the natural logarithm of the ratio Nd * / Nd is in the range of essentially from 3 to 7. 63. The method according to p. 60, characterized in that the thermionic converter additionally creates a compensated region located between the gap region and the collector and having such a configuration that it suppresses electric current from the collector to the gap region.

Images