Use of dissipative macroscopic currents to convert heat into electricity.

Abstract

The quantum device (10; 20; 30; 40; 50) comprises a transmitting structure (1; 1a, 1b; 21; 31; 41; 51) made of a topological material (10A; 20A; 30A; 40A; 50A) and at least two terminals (5; 8; 24; 34; 42, 43; 52; 53) connected to the transmitting structure, the transmitting structure being capable of generating a current of electrons, holes, or other quasiparticles between the at least two terminals, and at least one or more of the region (A, B) of the topological material (40A, 50A), one or both of the at least two terminals (5; 8), and the further element or material (23; 33) being capable of inelastically scattering at least a portion of the current.

Description

[Technical Field]

[0001] The present disclosure relates to devices and methods for converting thermal energy into voltage, current, magnetic field, or temperature inhomogeneities or electron spin distributions. In particular, the devices utilize a combination of incoherent electron transport, characterized by non-unitary, dissipative charge transport, and a conducting channel that carries dissipationless macroscopic current. [Background technology]

[0002] In the following description, the following documents are referred to: [Mannhart,2018a]J.Mannhart,JournalofSuperconductivityandNovelMagnetism31,1649(2018) [Mannhart,2018b]J.MannhartandD.Braak,JournalofSuperconductivityandNovelMagnetism31,1649(2018) [Bredol,2021]P. Bredol etal.,Phys.Rev.B.104.115413(2021) [Mannhart,2019]J.Mannhart etal.,PhysicaE109,198-200(2019) [Mannhart,2020]J.Mannhart,JournalofSuperconductivityandNovelMagnetism33,249(2020) [Mannhart,2021]J.Mannhart,H.Boschker,andP.Bredol,NanoExpress2,014998(2021) [Gerhardts,2008]RRGerhardts,Phys.Stat.Sol.(B)245,378(2008) [Gerhardts,2009]R.R. Gerhardts et al., ‘Quantum Hall Effect’ pp572 in Compendium of Quantum Physics, ed.: D. Greenberger et al. Springer (2009). [Panos,2014]K. Panos et al., New J. Phys. 16, 113071 (2014) [v. Klitzing,2005a]K. v. Klitzing, Phil. Trans. R. Soc. A363, 2203 (2005) [v. Klitzing,2005b]K. v. Klitzing et al., Phys. Journ. 4, 37 (2005) [v. Klitzing,2017]K. v. Klitzing, Annu. Rev. Condens. Matter Phys. 8, 13 (2017) [Weis,2011]J. Weis et al., Phil. Trans. R. Soc. A369, 3954 (2011) [Oh,2013]S. Oh, Science 340, 153 (2013). [Ando,2013]Y. Ando, J. Phys. Soc. Jpn. 82, 102001 (2013) [Buttiker,1988]M. Buttiker, Phys. Rev. B. 38, 9375 (1988) [Kane,2011]C. Kane et al., Physics World, Feb. 2011, pp.32

[0003] Literature [Mannhart,2018a,b;2019;2020;2021; Bredol,2021] Patent Application [PCT / EP2019 / 052634] [PCT / EP2019 / 058649] [PCT / EP2019 / 074347] [PCT / EP2020 / 068812] The above-mentioned publications and patent applications disclose nanoscale devices that convert heat into an electrochemical potential difference between two contacts of a nanodevice through the process of non-reciprocal propagation and inelastic scattering of electron quantum wave packets. Many applications require macroscale devices with lengths of millimeters or more. European Patent Application No. 21209862.8 (currently unpublished) discloses a means of fabricating such macroscale devices by appropriately assembling large quantities of nanodevices. Fabricating macroscale devices by these means can be complex and costly, as the nanodevices must first be patterned with resolution comparable to the Fermi wavelength and then connected with the correct polarity throughout the assembly. Summary of the Invention

[0004] A quantum device according to a first aspect of the present disclosure includes a transmitting structure made of a topological material; at least two terminals connected to the transmitting structure; the transmitting structure functions to generate a current of electrons, holes, or other quasiparticles between the at least two terminals; At least one or more of the region of topological material, one or both of the at least two terminals, and the further element or material function to inelastically scatter at least a portion of the current.

[0005] An array according to a second aspect of the present disclosure is an array of two or more quantum devices according to the first aspect, The quantum devices are connected in series and / or in parallel or anti-parallel to each other.

[0006] A use according to a third aspect of the present disclosure is the use of a quantum device according to the first aspect or an array of two or more quantum devices according to the second aspect in an electrical device, wherein two terminals of the quantum device or two external terminals of the array of quantum devices are connected to the electrical device.

[0007] A method according to a fourth aspect of the present disclosure is a method for converting thermal energy into one or more of voltage, current, magnetic field, or temperature inhomogeneity or electron spin distribution, the method comprising: fabricating a quantum device according to the first aspect or an array of two or more quantum devices according to the second aspect; connecting the quantum device to a load or other device configured to generate one or more of a voltage, a current, a magnetic field, or a temperature inhomogeneity or electron spin distribution.

[0008] Those skilled in the art will recognize additional features and advantages upon reading the detailed description herein and upon viewing the accompanying drawings. [Brief explanation of the drawings]

[0009] The accompanying drawings are included to provide a further understanding of the embodiments, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and, together with the description, serve to explain the principles of the examples. Elements in the drawings are not necessarily to scale. Like reference numerals in the drawings indicate corresponding like parts. Other embodiments and many of the intended advantages of the embodiments will be readily appreciated, as they become better understood by reference to the following detailed description. [Figure 1] Figure 1A, 1B, and 1C show the macroscopic dissipationless current flowing along the edge of a QHE (Quantum Hall Effect) sample. Figure 1B shows the macroscopic dissipationless current flowing along the edge of a quantum spin Hall effect sample (a 2D topological insulator). Figure 1C shows the macroscopic dissipationless current flowing along the edge of a quantum anomalous Hall effect sample. This figure is taken from [Oh, 2013]. [Figure 2] Schematic of a macroscopic dissipationless current flowing along the surface of a 3D topological insulator. This figure is adapted from [Ando, ​​2013]. [Figure 3]It consists of FIGS. 3a, 3b, and 3c, which are the electron orbits (left) in real space and the corresponding band diagrams of (a) a standard semiconductor, (b) a QHE sample, and (c) a 2D topological insulator with a magnetic field applied perpendicular to the plane, respectively. The left panel of (b) shows the Landau cylinders formed in QHE, and the skipping edge electron orbits proposed in [Buttiker, 1988] are shown at the lower end. The panel of (c) shows the edge currents and band structures in the upward and downward spin directions of electrons. This figure is cited from [Kane, 2011]. [Figure 4] It shows the energy and filling of Landau levels in a QHE sample. The energy of the Landau levels increases towards the two edges of the sample, forming incompressible stripes (1) and compressible stripes (2). This figure is cited from [Panos, 2014]. [Figure 5] It is a top view of a QHE sample with a magnetic field Bz applied. The incompressible stripes (1) form closed loops along the edges of the sample. This loop carries a non-dissipative current that circulates around the loop with a line density Jx. This current is a drift current induced by an electric field Ey. [Figure 6] It is a top view of a QHE sample with a magnetic field Bz applied. A part of the sample is modified to reduce the mean free path of electrons to Ie << rc (normal phase (3)). Therefore, the incompressible stripes (1) are formed only in the remaining part of the sample where the mean free path for QHE is Ie >> rc. Labels a, b, c indicate the positions where the energy diagrams shown in FIG. 7 were obtained. [Figure 7] It shows the Landau levels in the left, middle, and right parts (positions a, b, c respectively) of the sample shown in FIG. 6. The incompressible stripes are formed only at the edge of the left sample (the Landau levels are either completely filled or empty). The Landau levels at the right end are broadened and overlapped due to Ie << rc. Therefore, at the right end, the electron system does not have a complete gap. As a result, no incompressible stripes are generated at the edge of this sample, and it shows ohmic behavior for charge transport. [Figure 8] Top view of a QHE sample with a magnetic field Bz applied. In some parts of the sample's edge, an incompressible stripe (1) is formed. In other parts, the electric field induced by the gate electrode (4), here located on the sample side, modifies the electric field Ey, suppressing the formation of the incompressible stripe at each part of the sample's edge. In this example, the gate voltage has a negative value relative to the sample. Labels a, b, and c indicate the positions where the energy diagram shown in Figure 7 was obtained. [Figure 9] Figure 8 shows the Landau levels in the left, middle, and right parts of the sample (positions a, b, and c, respectively). Incompressible stripes form only at the edge of the left sample (the Landau levels are either completely filled or empty). At the edge of the right sample, the electrochemical potential remains at the Landau level. As a result, band bending at the edge of the right sample is suppressed, no incompressible stripes are formed at this edge, and charge transport exhibits ohmic behavior. [Figure 10] This is a top view of a portion of a QHE sample with a magnetic field Bz applied. As shown in the example of Figure 6, the portion of the sample is in the ohmic phase, with the incompressible stripe 1 and the compressible stripe 2 forming straight stripes, not loops. Two contacts 5 are connected to these two stripes via tunnel barriers 6. This arrangement is called a "striped element." These two contacts 5 may be connected to the normal phase 3 and to at least one other striped element. [Figure 11] Top view of a QHE sample subjected to a magnetic field Bz according to Fig. 6. The two-dimensional electron system of the sample has been locally eliminated by cutting a slit 7 into the sample. [Figure 12]This figure shows an example of an arrangement according to the second embodiment, specifically a top view of a QHE sample subjected to a magnetic field Bz selected to contain a compressible electron fluid in the main part of the sample. This sample shows the series connection of the incompressible stripe elements shown in FIG. 10. The polarity of the voltage induced in each stripe element is determined by the polarity of Ey induced by the left edge of the sample, which is the same for each element, and Bz. Therefore, the polarity of each element is the same, and in the structure shown, the induced voltages of the elements are added together, resulting in a voltage across the outermost contacts (8). [Figure 13] Figure 1 shows a top view of a QHE sample subjected to a magnetic field Bz selected to contain a compressible electron fluid in the bulk of the sample. In this example, the formation of incompressible stripes at the sample's edge is suppressed by a positively charged gate electrode 4b placed close to the sample's edge. In addition, five arbitrarily shaped gate electrodes 4a, which are negatively charged and may have different gate voltages, are placed on the sample surface and used to induce corresponding incompressible stripes in the sample's interior region away from the sample's edge. [Figure 14] This figure shows an example of an arrangement according to the second embodiment, specifically a top view of a QHE sample subjected to a magnetic field Bz selected to contain a compressible electron fluid in the majority of the sample. The sample shows a series connection of incompressible stripe elements 1a distributed across the sample's surface. In this case, the incompressible channel is induced by a gate electrode 4a, which in this example is negatively charged with respect to the sample. A contact 5 is placed on an insulating layer 9 to insulate it from the semiconductor. As shown on the right side of the figure, the negatively charged gate electrode 4 can induce unintended secondary incompressible stripes 1b depending on the sample layout. The formation of such unintended incompressible stripes 1b can be suppressed by adding a positively charged gate electrode 4b to the sample, which shields the electric field of the negatively charged gate electrode 4a, as shown at the bottom of the sample. [Figure 15]It is a cross-sectional view of a QHE sample that receives a magnetic field Bz selected such that an incompressible electron fluid is included in the main part of the sample. In this incompressible fluid, current is flowing in a thermal equilibrium state (large arrow). This current is coupled to a further conductor via a barrier. Due to this coupling, current also flows in the further conductor (small arrow). As shown in the figure, when the further conductor is restricted by two contacts, a voltage accumulates between these contacts. [Figure 16] It is a top view of a QHE sample that receives a magnetic field Bz selected such that current circulates in a ring-shaped incompressible stripe with diameter D and width w when the thermal equilibrium is not disturbed under the applied Hall electric field. This device includes a gate electrode G depicted as a small black dot with diameter d deposited on a gate insulator grown on the ring. When charge is applied to the gate electrode, a carrier depletion region with diameter s > w is formed, and the circulating current passes through the shunt resistance RL. [Figure 17] It is a diagram showing the value of the current circulating in the incompressible ring shown in FIG. 16 as a function of time t. The gate electrode is charged only between ton < t < toff. During this time, the circulating current passes through RL and decays with a time constant τ proportional to the inductance L of the ring. [Figure 18] It is a schematic top view showing an example of a quantum device according to the first aspect, and in the use of the device, it shows that chiral surface currents flowing along the edge of a topological insulator are locally coupled to a spatial region including these inelastic scattering centers. [Figure 19] In the quantum device shown in FIG. 18, it further shows the operation of the device in which operating electrons flow from region C to region D at a higher rate than from region D to region C. [Figure 20] It shows a further example of a quantum device according to the first aspect, and this quantum device is a further developed form of the quantum device according to FIG. 18, with additional terminals and inelastic scattering centers added. [Figure 21]1 shows a further example of a quantum device according to the first embodiment. The device utilizes a topological insulator with two contacts 1, 2 and embedded in a thermal bath. The topological insulator forms chiral edge states at its top and bottom edges, as shown, that carry spin-polarized electrons. DETAILED DESCRIPTION OF THE INVENTION

[0010] Each quantum device described below is characterized in that it is adapted to extract thermal energy from an ambient heat bath and convert that thermal energy into the generation of an electric current. Furthermore, the quantum device is adapted to generate an electric current even when at least two terminals of the quantum device are held at the same temperature. In particular, quantum devices can have a nonreciprocal transmission structure, meaning that particle current is stronger in one direction between the contacts than in the other direction. Some quantum devices operate based on the quantum Hall effect and require the application of a magnetic field. Other quantum devices require the application of an additional electric field to induce the quantum Hall effect, while others require neither a magnetic nor an electric field.

[0011] The problem of converting heat into an electrochemical potential difference between two contacts using nonreciprocal electron transport in macroscale devices, which does not require high-resolution sample patterning or nanodevice alignment through one or more fabrication steps, is solved by appropriately combining an incoherent electron system characterized by nonunitary, dissipative charge transport with a conducting channel that carries a dissipative, i.e., nominally lossless, macroscopic current. A dissipative channel can be obtained, for example, by using a structure that exhibits the quantum Hall effect (QHE) or related topological phenomena. In describing this disclosure, a solid consisting of an n-doped semiconductor held in thermal equilibrium at temperature T is used as an example, although this disclosure is not limited to n-type semiconductors. 1. Generation of dissipationless macroscopic current

[0012] Various topological phenomena in two-dimensional (2D) and three-dimensional (3D) solids are known to be associated with dissipative currents that flow in closed loops along the edges and surfaces of two-dimensional (2D) and three-dimensional (3D) solids in thermal equilibrium, as shown in Figures 1 and 2, respectively. Examples of such topological phenomena include the integer quantum Hall effect, fractional quantum Hall effect, anomalous quantum Hall effect, quantum spin Hall effect generated by 2D topological insulators, and surface currents in 3D topological insulators. Transport along the edges and surfaces of a sample is often described by surface bands that intersect with the chemical potential in the band structure of the sample (see Figure 3). These surface bands are determined by the topological properties of the solid.

[0013] Before describing this disclosure, it is necessary to provide a definition of the term "topological material."

[0014] First, the term "topological materials" generally refers to two different types of materials, as described below: materials that already have intrinsic topological properties, i.e., materials that possess these properties without external influences such as the application of magnetic or electric fields, and materials that are not intrinsically topological but can acquire these properties through external influences such as the application of magnetic or electric fields.

[0015] Following the literature usage [B. Yanan and S.-C. Zhang, Rep. Prog. Phys. 75, 096501 (2012); C. Choi, Spectrum IEEE, p. 6, July 2021], the term "topological material" refers to a material whose electronic structure possesses at least one of the following properties: it can be counted as an integer, cannot be continuously changed without a qualitative change in the band structure, and has integer values ​​different from those of trivial materials such as silicon, air, or vacuum. These numbers are called topological numbers and are exemplified by the so-called "TKNN invariants" or "Tschern numbers" [M. Z. Hasan and C. L. Kane, Rev. Mod. Phys. 82, 3045 (2010); M. Franz and L. Molenkamp, ​​Topological Insulators, Elsevier, 2013]. Specifically, for the integer and anomalous quantum Hall effects, the topological number is given by the number of completely filled Landau levels. For topological insulators, a commonly used topological number, the so-called Z2 invariant, refers to whether the number of electronic bands across the band gap between two Kramers degeneracy points is even (topologically trivial materials) or odd (topological materials).

[0016] Qualitative changes in band structure are most often caused by the disappearance of the band gap at an electrochemical potential. Thus, when two materials with different topological numbers are joined at an interface, the interface becomes conductive due to the disappearance of the band gap, even if the interior of the material is insulating. In particular, if one of the materials is air, the interface becomes the surface of the material, and surface conduction occurs. The existence of such interface or surface states is described as "topologically protected" or "protected by symmetry properties."

[0017] Furthermore, interface states in topological materials can carry dissipationless currents, as in the case of incompressible stripes in quantum Hall devices. Also, in topological insulators, interface and surface states can carry currents without energy loss in the absence of inelastic and spin-flip scattering. Such currents are those that flow along the edges of devices, as shown in Figure 1.

[0018] Furthermore, in topological insulators with spin-orbit coupling energy, the direction of electron motion and the direction of spin are linked by the spin-orbit coupling. Such a current is called a chiral surface current. The dispersion relation of this current is linear in the electrochemical potential. Therefore, the absolute value of the electron velocity is constant. This means that the absolute value of the current density is proportional to the electron density.

[0019] Topological materials exist in two or three spatial dimensions. In 2D topological materials (shown in Figure 1), the interface or surface states are 1D states that extend along the edges of the sample, while in 3D topological materials (see Figure 2), the interface or surface states are 2D states that extend along the surface of the sample.

[0020] To understand the gist of this disclosure, it is necessary to consider the microscopic mechanisms driving current flow in more detail than just band structure diagrams. These currents are best understood and experimentally confirmed in the case of the QHE. Therefore, while this discussion focuses on the QHE, it is not limited to the QHE and applies to other phenomena in which a macroscopic dissipationless current is induced in thermal equilibrium.

[0021] The most recent understanding of QHE is provided by what we call the "incompressible stripe model." Pioneering work on this model was conducted by Professor v. Klitzing, the discoverer of QHE, and his group, and is described in [v. Klitzing, 2005a; v. Klitzing, 2005b; Gerhardts, 2008; Gerhardts, 2009; Weis, 2011; Panos, 2014; v. Klitzing, 2017]. As explained in [v. Klitzing, 2005b; Weis, 2011; v. Klitzing, 2017], this understanding improves upon the "skipping orbit model" previously proposed by Buttiker (Buttiker, 1988). This skipping orbit model is mentioned here only because it has been frequently presented in older literature.

[0022] The incompressible stripe model has advanced our understanding of the generation of dissipative macroscopic currents in the QHE, as summarized below. z and the mean free path of the electron is B z The cyclotron radius I corresponding to e >>r c In a semiconductor sample much larger than , the electronic states form a Landau cylinder (shown in Figure 3a, b) whose energies are given by the Landau levels. At thermal equilibrium, the potential φ eWhile ρ is constant throughout the bulk of the sample, it rises at the sample's edge, forming a depletion layer with a width on the order of the semiconductor's Debye length. As a result, the sample edge is filled with alternating stripes of incompressible and compressible electron fluid. Hereafter, these stripes are referred to as incompressible and compressible stripes (Figure 4). In the incompressible stripes, the chemical potential lies between fully filled and fully empty Landau levels. Here, ν denotes the number of fully filled Landau levels. The number of incompressible stripes is determined by the value of υ. Of these incompressible stripes, typically only the one furthest from the sample's edge can carry a dissipation-free current. For this reason, without loss of generality, we will focus on this stripe below.

[0023] The band bending at the sample edges is due to the electric field E y (Fig. 5). This electric field penetrates the stripe, causing the electronic state to drift in the direction of the stripe, resulting in a current flowing parallel to the x-direction along the stripe. This is a somewhat counterintuitive phenomenon for the movement of electrons in the presence of an electric field and a magnetic field perpendicular to it. In the semiclassical picture, an electron at rest at the start moves in a direction parallel to the x-direction, E y (left end in the -y direction). After that, it moves in the -y direction at a finite speed, so B z The Lorentz force acting in the x direction is due to the force acting in the x direction. Thus, the electron traces an arc in the x direction, while the Lorentz force acts in the -y direction, and this process is repeated. As a result, the electron traces a cycloidal orbit that is guided, on average, in the x direction.

[0024] The value of this current in the x direction is the linear current density

number

[0025] For an incompressible stripe to be created and for dissipation-free current to flow, the band bending and the associated electric field E y It is essential to realize this. Note that such band bending and electric field can be induced even at locations far from the sample edge, for example, by using gate electrodes, as shown in Figures 13 and 14. The freedom afforded in designing a gate electrode or multiple gate electrodes, selecting locations far from the sample edge within the sample, and selecting the respective gate voltages is useful in many applications. This allows, for example, the interior of the sample to also be used to produce the desired results. Similarly, it is of course possible to suppress the formation of stripes by placing gate electrodes with appropriate voltages on the sample.

[0026] It should be noted that in polycrystalline samples, not only the outer edges or surfaces of the entire sample but also the internal grain boundaries can be used as sample edges, since these boundaries act as edges or surfaces of the respective grains, as well as twin boundaries in the corresponding single crystal. 2. Cutting the loop of the incompressible stripe

[0027] As a next step in explaining this disclosure, we consider the existence of QHE as a condition I e >>r c For example, by introducing defects or impurities into a semiconductor by irradiation or diffusion, the electron mean free path I ebecomes small enough, the semiconductor z For clarity of discussion, we will limit ourselves to the idealized case, I e Consider the case where other sample parameters such as electron density and dielectric properties remain unchanged except for the decrease in I. e The electron phase with reduced electron density is called the "normal phase."

[0028] By using lithographic patterning techniques, the sample can be fabricated with the large I required for the QHE effect. e and a small I e It is clear that the structure can be modified to include both the normal phase with a QHE and the normal phase with a QHE. As an example, consider the geometric configuration shown in Figure 6. A sample is constructed by patterning the incompressible stripes so that they do not form closed loops. These remaining stripes are oriented in the x direction. The remainder of the sample consists of a normal semiconductor with no Landau levels and QHE. Therefore, no incompressible stripes can be formed in the remainder of the sample, and no dissipative current flows there, as evidenced by the energy level diagram for the system (Figure 7). It will be appreciated that the patterning required to obtain the desired structure requires only moderate resolution, as the width of the incompressible stripes is typically on the order of 0.1 μm, and more commonly greater than 0.05 μm.

[0029] An alternative approach to obtain an incompressible stripe that does not form a closed loop is as follows: e Instead of reducing the band bending, one or more gate electrodes are used to eliminate the depletion layer, for example, by locally suppressing the band bending on one side of the sample (the right side in Figure 8). In this case, the sample is fully in the QHE state, but the loop due to the incompressible stripe is broken next to the gate electrode. This is evident from the energy level diagram of the system (Figure 9).

[0030] In both cases, the resulting sample consists of an incompressible stripe with both ends in electrical contact with ohmic conductors provided by the compressive electronic or normal phase of the QHE state. 3. Contact to incompressible stripe

[0031] Next, we consider the effect of establishing two contacts between the stripe and the ohmic semiconductor. While the incompressible stripe may establish contact in a variety of ways, for simplicity, we will assume that the contact is established by two arbitrary tunnel barriers, as shown in Figure 10. The configuration shown in Figure 10 will be referred to hereafter as a "stripe element." 4. Behavior of stripe elements

[0032] In the first step, the electric field E induced by the band bending at the sample edge y However, as mentioned above, the drift current

number

[0033] This state changes when a contact is placed on the sample. In this case, drift electrons reaching the bottom contact, as shown in Figure 10, can tunnel through the tunnel barrier and occupy available free states in the electronic density of states of the n-type semiconductor providing the bottom contact. This semiconductor is characterized by a Fermi-Dirac distribution for the state occupancy at finite T.

[0034] As electrons leave the stripe at the bottom contact, electrons within the stripe drift downward in the -x direction and are replenished at the top contact by thermally excited electrons tunneling into the stripe from the top contact. The process by which electrons move into and out of the stripe is an incoherent, inelastic process. Typically, this involves thermal excitation or a continuous electron density of states. Therefore, the device operation involves a combination of coherent electron transport, described by the Schrödinger equation, and incoherent transport, better described by the Born rule or Fermi's golden rule. In the example described here, the transition between these two transport processes occurs at the boundary between the incompressible state and the contact. However, in some cases, this transition can occur at a different location. For example, if the contact itself maintains a coherent electron system, the transition can occur within the contact. This process accumulates electronic charge at the bottom contact and partially removes electrons from the top contact. This process accumulates electronic charge at the bottom contact and partially removes electrons from the top contact. This difference in electron density between the two contacts creates an electrochemical potential difference. That is, a finite voltage is generated between the top and bottom contacts.

[0035] FIG. 10 illustrates an example of a quantum device according to the first aspect. The quantum device 10 includes a semiconductor sample 10A, e.g., a silicon sample 10A, to which a magnetic field perpendicular to a major surface of the semiconductor sample is applied. The semiconductor sample 10A can thus exhibit the quantum Hall effect and is an example of a topological material. In this embodiment, the quantum device 10 further includes a nonreciprocal transmitter structure 1, which is an incompressible stripe 1 as described above. The quantum device 10 further includes two terminals 5, which are contacts 5 connected to opposite ends of the nonreciprocal transmitter structure 1. The nonreciprocal transmitter structure 1 is configured to generate a current of electrons between at least two contacts 5, and in this embodiment, the two contacts are configured to inelastically scatter at least a portion of the current dissipatively.

[0036] In the embodiment shown in Figure 10, two compressive stripes 2 are provided on either side of the non-compressive stripe, although in principle the compressive stripes 2 are not essential for the actual function of the part. The device could, of course, be constructed to include only one compressive stripe 2 (the stripe near the edge of the sample), although the presence of this compressive stripe is essentially an incidental effect.

[0037] If the two contacts 5 are connected to each other via a load, for example a lamp, the following circuit is obtained: the incompressible stripe 1 "produces" a current that flows only within the incompressible stripe 1 without dissipation. This current enters the first contact 5, where the electrons are inelastically scattered. The current then flows across the lamp to the second contact 5 and finally back to the incompressible stripe 1. It is also possible to intentionally introduce a small amount of dissipation into the incompressible stripe 1, preventing the current from flowing completely without dissipation.

[0038] It should be noted that the incompressible stripe can be induced by the gate electrode even inside the sample, i.e., at a location far from the sample's edge or surface, as shown in Figures 13 and 14. Furthermore, portions of the semiconductor stack hosting the 2D electron system can be intentionally destroyed locally by etching. Therefore, by utilizing photolithographically defined gate electrodes or chemically etched structures, it is possible to freely position the incompressible stripe throughout the sample. This allows for a larger sample area to be utilized for power generation.

[0039] It is also noted that the efficiency of electron transport into the contact material can be increased by utilizing a more transparent tunnel barrier or by increasing the density of states in the contact material.If the contact is left open, i.e., no further connection is made, a finite voltage can again affect the charge distribution in the incompressible stripe, causing the drift current to disappear.

[0040] However, the contact is formed by the compressive stripe and the normal phase and is in contact with the ohmic part of the sample. In this case, electrons flow from the bottom contact to the top contact through the ohmic part of the sample, resulting in an internal ohmic current I shunt This results in: 5. Achieving output voltage or output current

[0041] When the contacts, preferably the two outermost contacts 8 (see FIG. 12), are connected to an external load, a portion of the flowing charge, determined by Kirchhoff's law, passes through the external load, thereby providing the output power of the device. This output power is I shunt It is clear that this can be enhanced by choosing a sample design that minimizes . The simplest way to do this is to cut multiple slits into the sample, as shown in Figure 11.

[0042] The polarity of the induced voltage is B z and E y Since the polarity of the stripes is determined by the polarity of the sample, any point on a particular edge of the sample will have the same polarity, which makes it easy to fabricate arrays of many stripes connected with the correct polarity (see quantum devices 20, 30 in Figures 12 and 14). 6. Further Embodiments

[0043] The devices described above utilize one or more incompressible stripes 1 in combination with contacts 5 as shown in FIGS.

[0044] The combination of current flowing in the ground state of the electron system and incoherent processes is obviously applicable to further device configurations different from the example shown in Figure 10. Two further examples are given below. I) Devices characterized by voltages and / or currents generated in conductors in the vicinity of the incompressible stripe

[0045] 15 is a cross-sectional view of a quantum device 20 comprising a semiconductor sample 20A, an electronic system forming an incompressible stripe 21 through which current flows in the ground state of the system, a barrier layer 22, and a further conductor 23. The barrier layer 22 couples the electronic systems of the incompressible stripe 21 and the further conductor 23, for example by electron tunneling or phonon exchange.

[0046] The further conductor 23 is an ohmic conductor 23, such as an n-doped semiconductor, in this example, with a mean free path shorter than the magnetic length, and does not form Landau levels or exhibit the quantum Hall effect. The coupling between the incompressible stripe 21 and the further conductor 23 tends to induce electron drift in the further conductor 23, due to electrons moving in the incompressible stripe 21 with a velocity in the x-direction given by the value of the current flowing through it. This induction occurs either by electron tunneling with a finite drift velocity from the incompressible stripe 21 to the further conductor 23, or by electrons in the incompressible stripe 21 first inducing a finite drift velocity in the phonon system, which then induces a finite drift velocity in the further conductor 23. By the same principle, the electron system in the further conductor 23 has the effect of slowing down the electron drift in the incompressible stripe 21.

[0047] The electron current thus induced in the further conductor 23 induces a voltage between two contacts 24 arranged at either end of the further conductor 23. In steady state, this voltage suppresses the flow of current in the further conductor 23. However, this output voltage can also be used to drive an electrical device. A load on this device reduces the output voltage, which again induces a current in the further conductor 23. This current flows in a loop as it passes through the electrical device.

[0048] Of course, additional conductors 23 may be placed adjacent to the sides of the non-compressible stripes 21, and multiple additional conductors may be used to increase output power. The additional conductors 23 may also be placed in close proximity to the non-compressible stripes 21 without the intentional fabrication of a barrier layer 22. Indeed, compressive stripes may also be used as additional conductors. The barrier layer 22 itself may also be configured as compressive stripes located at the edge of the semiconductor device. II) Devices characterized by a temporary interruption of current flow in an incompressible stripe

[0049] To understand the operation of quantum devices that function by the temporary interruption of current flow in an incompressible stripe, we restate the following two points. a) The carrier density of the 2D electron system can be easily adjusted up to full depletion by using a field-effect transistor, in which a charge applied to the gate electrode changes the carrier concentration of the 2D electron system. Such a field-effect transistor operates with power amplification: very little power is needed to turn a large current on or off. b) A current flowing through an incompressible stripe induces a magnetic field, and therefore the current is accompanied by an inductance L. For a straight stripe, L increases proportionally to the length of the stripe, and for a looped stripe, L increases proportionally to the diameter of the loop.

[0050] The operation of each further quantum device according to the first aspect will be explained with reference to Figures 16 and 17. In the example shown, the quantum device 30 comprises an incompressible stripe 31 forming a loop in the xy plane in a layer near the surface of a semiconductor sample 30A, for example a silicon sample 30A. The device 30 further comprises a gate electrode (G) 32 having a diameter d much smaller than the width w of the incompressible stripe 31. The gate electrode 32 can be electrically charged to completely deplete an area of ​​lateral dimension s > w, rendering it insulating with a potential barrier so that charge carriers cannot traverse said area. This area is further shunted by a load resistor 33, which has an ohmic resistance R L Conductive electrode layers 34 are provided on the inner and outer surfaces of the incompressible stripes 31. The electrode layers 34 generate an electric field, which can induce the Hall effect.

[0051] Figure 17 shows the time course of the current I through the device. The gate voltage is applied for a period t during which the gate is charged, causing complete depletion, as described above. on <t<t off is zero except for t <t on In the case of t, a constant current I0 flows in the ring given by the incompressible stripe, according to the equilibrium condition. on However, for a given inductance L, the circulating current can only change with a time constant τ due to the device inductance. This allows the current to decrease in value as it passes through the ohmic resistance R. L 33 and bypass the gate area. L Due to the dissipation induced by L It decays with t off Then the gate is turned off again and the current flows again entirely in the incompressible stripes 31 and returns to its original value I0.

[0052] The maximum gate voltage required to block current flow under the gate is I c R L, which corresponds to the maximum voltage that can occur along resistor 33 or across the depletion region under gate 32. This limit is independent of the inductance L, and therefore the diameter D of the ring. The square of that voltage is the maximum energy E that can be charged to the gate capacitor to block current flow. gate , the above energy also does not depend on L or D.

[0053] Also, the period during which the current flows through RL is t on <t<t off The electrical energy flows through the resistance R L The energy is dissipated in the ohmic resistor 33, generating heat. If this ohmic resistor 33 is replaced with an electrical device with ohmic properties, the electrical device can perform work determined by the energy. This energy is expressed as cI c 2 R L R is given by L and the time constant τ=L / R L It is scaled by the product of E out =cI c 2 R L xL / R L =cI c 2 L Here, c≒1 / e 2 where e is Euler's constant, a constant coefficient. By increasing D and thereby increasing L, E out can be arbitrarily large, but E gate It is important to note that does not grow with D. When D is large, E out is E gate Exceeds.

[0054] Thus, when operated by repeatedly switching the gate, the device generates a time-averaged R LA net power is delivered to the gate. In particular, if an ohmic resistor is connected in series with the bias voltage, repeated gate switching can also be achieved by thermal noise induced by the ohmic resistor. Furthermore, the device can be operated at high T, where the depletion region is formed by the charge accumulation process of kT-driven point or line defects, as shown in Figure 16.

[0055] 16 thus shows a further example of a quantum device according to the first aspect, the quantum device 30 comprising a semiconductor sample 30A. For example, a magnetic field perpendicular to a major surface is applied to the semiconductor sample 30A, which allows the semiconductor sample 30A to exhibit the quantum Hall effect and is therefore an example of a topological material. Furthermore, in this embodiment, the quantum device 30 comprises a nonreciprocal transmitting structure 31, which is an incompressible stripe 31 provided as described above. Furthermore, the quantum device 30 comprises a further element 33, namely a load resistor R connected to the nonreciprocal transmitting structure 31 by two terminals 34 located on opposite sides of the gate electrode 32. L In this embodiment, the non-reciprocal transmission structure 31 includes a load resistor R L 33, which in this embodiment is configured to induce a current of electrons through a load resistor R L 33 is arranged to dissipatively and inelastically scatter at least some of the current. Thus, in this embodiment, it is not the lossless current itself that flows through the incompressible stripe 31 that is dissipatively scattered by that current, but rather the load resistance R L Only the current induced in 33 is utilized to generate an external voltage or current. 7. Expansion to other systems

[0056] The above discussion has been based on the example of an integer QHE. However, this disclosure is not limited to integer QHE devices and is generally applicable to any sample or phenomenon in which a dissipationless macroscopic surface current is induced, as will be explained below using examples of fractional QHE, anomalous QHE, and the quantum spin Hall effect. Additionally, other systems exist, such as Chern insulators.

[0057] First, it should be noted that the present disclosure is applicable without modification to fractional QHE, since the above discussion is also fully valid in this case.

[0058] In the case of the anomalous QHE (Figure 1C), the present disclosure functions similarly to the method described above, except that the internal magnetization of the sample obviates the need for the application of a magnetic field B. In the case of the quantum spin Hall effect (Figure 1B), the surface current consists of two countercurrent flows of electrons with opposite spin orientations, so-called chiral edge modes. The application of a magnetic field B is not required. Fabricating and operating the device as described above (without a magnetic field B) results in electrons with one spin direction accumulating at one contact and electrons with the opposite spin direction accumulating at the other contact. This behavior is useful for separating electrons with different spin directions and creating a magnetic dipole. Net voltage and current are generated when the symmetry between the electron populations of the two spin directions is broken, for example, by applying a magnetic field B, introducing spin-dependent scattering into the sample, or using an additional spin polarizer or spin analyzer.

[0059] A further embodiment of the present disclosure utilizes chiral surface currents that flow along the edges of topological insulators. In the typical case, two chiral currents with opposite spin directions circulate in opposite directions through the sample, as shown in Figure 1B.

[0060] Figure 18 shows an embodiment of a quantum device 40 based on a disk-shaped topological insulator 40A (TI) with two counter-circulating chiral currents at its edges. The direction of electron flow in the chiral channel is indicated by the triangle symbols. The clockwise-flowing chiral current is formed by electrons with a spin-down direction in the plane of the topological insulator 41, while the counterclockwise-flowing current is formed by electrons with a spin-up direction. Device 40 comprises two regions A and B that generate spin-flip scattering. Region A preferentially scatters spin-up electrons downward, while region B preferentially scatters spin-down electrons upward. Device (C) has its top half connected by contact 42 and its bottom half connected by contact 43.

[0061] Therefore, in this embodiment of the present disclosure, these chiral surface currents are locally coupled to a spatial region containing inelastic scattering centers A and B, which cause spin-flip scattering. Such inelastic scattering centers can be realized, for example, by impurity states or quantum dots with magnetic moments. If the density of spin-flip scattering centers is high, the scattering can cause undesirable Anderson localization. Therefore, the scattering center density is preferably selected to be smaller than the material-dependent value required for Anderson localization.

[0062] In this embodiment, at least one such scattering region is provided, and the scattering process is designed to have a specific directionality. For example, the scattering is designed to favor spin-up to spin-down scattering over spin-down to spin-up scattering. The directionality of the scattering process can be achieved by selectively aligning impurity states or spin magnetic moments with a locally applied magnetic field (e.g., a magnetic field using a permanent magnet).

[0063] As shown in Figure 18, device 40 comprises at least two such scattering regions. In the example shown in Figure 18, one region (A) preferentially scatters from spin-up to spin-down, while the other region (B) preferentially scatters from spin-down to spin-up. Furthermore, the device comprises two contacts 1 and 2 located in two spatial sections C and D of the chiral surface current connecting A and B, respectively.

[0064] The chiral current flow, shown schematically in Figure 18, corresponds to a conventional equal current density. In this case, the two contacts 42 and 43 are at the same electrochemical potential, and the electron density in the current path is constant. Due to the action of spin-flip regions A and B, the state in Figure 18 is not stable. However, A and B deflect the chiral current, as shown in Figure 19. As a result, starting from a constant electron density, the electron flow rate from region C to D is faster than the electron flow rate from D to C. Therefore, a difference in electron density between regions C and D, i.e., a difference in electrochemical potential between regions C and D, is created, inducing a dissipative backflow of electrons from D to C, and a new steady state is established, where a new balance of the current occurs. As a result, compared to the beginning of the process, the action of A and B can pump electrons from C to D. This electrochemical potential difference is transmitted to contacts 42 and 43 and can be used to drive an external load.

[0065] 19 thus shows the result of the action of the spin-flip scattering process occurring at A and B. At A, counterclockwise circulating electrons are flipped into clockwise circulating electrons (see curved arrows), preventing them from entering region C. At B, clockwise circulating electrons are flipped into counterclockwise circulating electrons (see curved arrows), preventing them from entering region C. As a result, the electron density decreases at C and increases at D, producing a voltage across contacts 42 and 43.

[0066] Naturally, the electrochemical potential difference of such a device, or the output current when a load is connected, will be favored by the larger surface currents flowing in the sample before A and B are added. Therefore, to maximize the beneficial effects described above, it is useful to use topological insulators that have large chiral currents or a large increase in the free energy associated with these chiral currents.

[0067] 18 shows a further example of a quantum device according to the first aspect, where quantum device 40 comprises a topological insulator 40A, which is an example of a topological material. Quantum device 40 further comprises a transmitter structure 41, which in this embodiment comprises chiral channels for counter-propagating electrons. Quantum device 40 further comprises regions of topological material, i.e., inelastic scattering centers A and B, adapted to inelastically scatter at least a portion of the current dissipatively.

[0068] The example shown in Figure 18 is merely one specific example of the general principle of this embodiment of the present disclosure. For example, the device will work even if either region A or B does not have a specific spin scattering direction. In that case, in addition to the electrochemical potential difference described above, a net chiral circulating current will arise surrounding the entire sample, which will generate a magnetic field.

[0069] Furthermore, as shown in Figure 20, the operation of this device can be further enhanced by adding some samples of the AB region and contacts 52, 53. Quantum device 50 comprises a topological insulator 50A, such as topological insulator 50A(TI) shown and described for quantum device 40 of Figures 18 and 19, several scattering centers A and B, and several contacts 52 and 53. Quantum device 50 is thus based on the principles of quantum device 40 shown in Figure 18. In the configuration shown in Figure 20, the presence of a large number of scattering centers A and B and contacts 52, 53 results in a larger device output.

[0070] Also, multiple devices such as those shown in Figures 18 and 20 can be combined in series, parallel, or networks, including stacking in the third spatial dimension.

[0071] Furthermore, the principles described here for 2D topological insulators are naturally applicable to 3D and other topological materials, resulting in improved output. In the case of 3D topological materials, regions A and B and contacts 42 and 43 may be stripes that intersect a plane perpendicular to the surface current flow direction. Current backflow along the surface, which provides unwanted current shunting, can be minimized by reducing conduction along such surfaces by adding additional inelastic or magnetic scattering centers.

[0072] We also emphasize that the discussion presented here is not limited to edge or surface currents composed of moving electrons. The mechanism uncovered is equally applicable to moving holes. It is also applicable when these currents involve other quasiparticles, such as spinons, magnons, phonons, polarons, or polaritons, whether these quasiparticles are yet to be discovered or have already been discovered.

[0073] Figure 21 shows a further example of a quantum device according to the first embodiment. The device utilizes a topological insulator with two contacts 1 and 2 and embedded in a thermal bath. The topological insulator forms chiral edge states at its top and bottom edges, as shown, carrying spin-polarized electrons. Electrons are depicted as small open circles, their spins indicated by up- or down-pointing arrows, and the direction of electron motion indicated by left- or right-pointing arrows. Inelastic scattering centers are spatially grouped into two regions, A and B, and are sketched as open circles. The spin-flip direction of these scattering centers is controlled by applying a magnetic field in the direction indicated by the thick arrows in regions A and B.

[0074] Here, AMBozkurt et al., Phys. Rev. B97, 245414 (2018) and WO2018 / 027243A1 show that spin-flip scattering of electrons moving within the chiral edge channels of a topological insulator can be utilized to implement a battery function, i.e., a device can be obtained that is charged by a charging current and then supplies an output current by discharging.

[0075] Although similarities exist with the embodiment shown in FIG. 21 , the devices described in AMBozkurt et al., Phys. Rev. B97, 245414 (2018) and WO 2018 / 027243 A1 function based on fundamentally different operating principles than the present disclosure. This difference is evident in the fact that the devices in AMBozkurt et al., Phys. Rev. B97, 245414 (2018) and WO 2018 / 027243 A1 are batteries, whose primary function is a storage mechanism using a memory resource whose information changes through spin flips. Battery operation requires an initial drive phase in which the device is "charged." During this phase, the memory resource is charged with an electric current to store energy. Furthermore, the battery does not require thermal excitation, inelastic scattering processes, or a heat bath. Furthermore, the battery does not utilize spatially inhomogeneous distributions of nuclear spins or dopant spins, nor does it utilize a specific, predefined spin flip scattering direction.

[0076] In contrast, the device disclosed here is a thermal power generator, not a battery, and has a different architecture. It does not require memory resources to store information or a charging current. However, it does require thermal excitation, inelastic scattering in a specific direction, and a heat sink.

[0077] These differences become clearer with reference to the embodiment of the present disclosure shown in Figure 21, which is shown for easy comparison with Figure 1c of A.M. Bozkurt et al., Phys. Rev. B97, 245414 (2018). This embodiment utilizes a spatially distinct scattering region A. This region contains a scattering center characterized by preferred spin-down to spin-up scattering. The preferred spin scattering direction is obtained by locally breaking spatial symmetry by applying a magnetic field, for example, with a permanent magnet. The scattering center can be composed of, for example, atoms that provide two empty states for electrons (one for each spin direction), one state for each spin direction, which are energetically Zeeman-split by a locally applied magnetic field. The unoccupied atoms do not require nuclear spin or total electron spin. These atoms are coupled to a thermal bath in which the device is embedded. To enhance the output signal, a scattering region B may be applied to the lower channel, but region B is required to have different spin-flip characteristics than region A. In the illustrated example, region B has a preferred spin-up to spin-down scattering direction, which is achieved by applying a magnetic field with an opposite polarity to that of region A.

[0078] As is evident from the above discussion, preferential spin-flip scattering causes a preferred direction of electron migration in the upper channel, which in turn creates an electrochemical potential difference between the two contacts.

[0079] Further embodiments or examples of the present disclosure are given below. (a) The device according to the first embodiment comprises an additional sample geometry, for example, which causes the incompressible loop to be interrupted in only one place. (b) To obtain a large output signal, optimize the shape of the sample to have a large area acting as a stripe. Example embodiments: 1) Use surface currents instead of edge currents. 2) Place an incompressible channel over the entire surface area of ​​the sample (compare Figures 13 and 14). (c) The contacts (5) are not in contact with the normal phase (3) but are connected only to adjacent stripe elements. Example of embodiment: An insulating film is disposed under at least some of the contacts. (d) Stripe elements can be connected in series or parallel, allowing for arbitrary circuit designs. (e) By stacking semiconductors in N layers, for example by using a heterostructure, the output is amplified N times. (f) The use of multiple gate electrodes suppresses the formation of multiple incompressible stripes. (g) Using impurity diffusion or defect formation to create normal phases for devices based on the quantum Hall effect. (h) In devices based on the quantum Hall effect, irradiation of the sample with particle beams or electromagnetic radiation is used to create normal phases. (i) Examples f, g, and h are combined with patterning processes to form geometric structures. (j) Using non-uniform magnetic fields to create desired structures, e.g., generating stripes with specific shapes. (k) Creating samples with non-uniform electronic or magnetic properties, for example, by using non-uniform carrier density, doping, permittivity, magnetic susceptibility, electric polarization, magnetization, or surface structure to form desired structures (e.g., stripes of a specific shape). (l) Adopt a sample design that suppresses shunt current and appropriately select electrical characteristics and geometric structure. (m) Use a gate voltage to turn the effect on / off or invert the polarity. (n) Use light to turn effects on / off or reverse polarity. (o) Materials for use in quantum devices: Si (silicon), GaAs (gallium arsenide), other standard elemental semiconductors, III-V and II-VI semiconductors, ZnO (zinc oxide), graphene, other 2D materials, topological materials (Bi2Se3, Heusler and semi-Heusler alloys), semimetals, metals, and materials containing or consisting of organic molecules. (p) Use bulk single crystal, bulk polycrystalline, bulk amorphous, thin film, heterostructure, and multilayer. (q) Replacing the tunnel contact with a Schottky or ohmic contact. (r) The normal phase (3) is designed to be in a high resistance state to minimize the shunt current. (s) A magnetic field is generated using a permanent magnet, nuclear spin polarization, or an electromagnet. (t) Use of uniform electric and / or non-uniform magnetic fields, allowing their application locally in any direction. (u) Adjust the operation of the device using optimized operating temperatures (below room temperature, room temperature, above room temperature).

[0080] The present disclosure also relates to the use of a quantum device according to the first aspect or an array of two or more quantum devices according to the second aspect in an electrical device by connecting two terminals of the quantum device or two external terminals of the array of quantum devices to a load of the electrical device.

[0081] The present disclosure also relates to a method according to a third aspect for converting thermal energy into one or more of a voltage, current, magnetic field, or temperature inhomogeneity or electron spin distribution, the method comprising fabricating a quantum device according to the first aspect or an array of quantum devices according to the second aspect, and connecting the quantum devices to a load or other device configured to generate one or more of the voltage, current, magnetic field, or temperature inhomogeneity or electron spin distribution.

[0082] According to an embodiment of the method according to the fourth aspect, in the case of a quantum device functioning based on the quantum Hall effect, the further step of activating the quantum device comprises applying and / or modifying a magnetic field to the transmitting structure and adjusting the strength of the magnetic field such that an incompressible stripe is formed in the at least one first region.

[0083] While the present disclosure has been described and illustrated with respect to one or more embodiments, changes and / or modifications can be made to the illustrated examples without departing from the spirit and scope of the appended claims. In particular, with respect to the various functions performed by the above-described components or structures (assemblies, devices, circuits, systems, etc.), the terms used to describe these components (including "means") are intended, unless otherwise specified, to correspond to any component or structure (e.g., functional equivalent) that performs the specified function of the described component, even if it is not structurally equivalent to the disclosed structure that performs that function in an exemplary embodiment of the present disclosure.

Claims

1. a transmitting structure (1; 1a, 1b; 21; 31; 41; 51; 61) made of a topological material (10A; 20A; 30A; 40A; 50A; 60A); at least two terminals (5; 8; 24; 34; 42, 43; 52, 53; 62, 63) connected to said transmission structure, the transmitting structure functions to generate a current of electrons, holes, or other quasiparticles between the at least two terminals; at least one or more of the regions (A, B) of said topological material (40A, 50A; 60A), one or both of said at least two terminals (5; 8), and further elements or materials (23; 33) have the function of inelastically scattering at least a portion of said current, Quantum devices (10; 20; 30; 40; 50; 60).

2. Regions (A, B) of the topological material (40A, 50A; 60A); one or both of said at least two terminals (5; 8); said further element or material (23; 33), At least one of the above is connected to a heat bath. Quantum device (10; 20; 30; 40; 50; 60) according to claim 1.

3. the transmission structure (1; 1a, 1b; 21; 31; 41; 51; 61) is a non-reciprocal transmission structure (1; 1a, 1b; 21; 31); Quantum device (10; 20; 30; 40; 50; 60) according to claim 1 or 2.

4. said transmitting structure (1; 1a, 1b; 21; 31; 41; 51; 61) being capable of generating said current when said at least two terminals operate at the same temperature; A quantum device (10; 20; 30; 40; 50; 60) according to any one of claims 1 to 3.

5. said transmitting structure (1; 1a, 1b; 21; 31; 41; 51; 61) extracts thermal energy from the surrounding thermal bath and converts said thermal energy into the generation of said electric current; A quantum device (10; 20; 30; 40; 50; 60) according to any one of claims 1 to 4.

6. The functional principle of the quantum device is based on either the quantum Hall effect, the quantum spin Hall effect, the quantum anomalous Hall effect, or surface currents on 3D topological materials; A quantum device (10; 20; 30; 40; 50; 60) according to any one of claims 1 to 5.

7. the functional principle of the quantum device is based on the quantum Hall effect, the at least one first region (1) of the topological material (10A) is formed between the at least two terminals (5; 8) such that, by applying a magnetic field of a predetermined intensity, the at least one first region (1) forms at least one incompressible stripe (1); In the incompressible stripe (1), a current of electrons is generated unidirectionally and without dissipation, 7. The quantum device (10) of claim 6.

8. The at least one first region (1) is surrounded by at least one second region (3), the at least one first region (1) has a first mean free path of electrons, and the at least one second region (3) has a second mean free path of electrons, the first mean free path being higher than the second mean free path; Quantum device (10) according to claim 7.

9. The first mean free path is at least 10 times higher than the second mean free path.

9. The quantum device (10) of claim 8.

10. The first region (1) has a width greater than 50 nm; A quantum device (10) according to any one of claims 7 to 9.

11. and further comprising at least one electrode (35) connected to the transmitting structure (31) and configured to generate an electric field for inducing the Hall effect or to direct the electrons in a desired direction. A quantum device (30) according to any one of claims 7 to 10.

12. the functional principle of the quantum device is based on the quantum spin Hall effect, In the transmitting structure (41; 51), at least one first region of the topological material (40A; 50A) is arranged so that two chiral currents of electrons having opposite spin directions circulate in opposite directions on a path connected to two opposite terminals (42, 43; 52, 53), and the first region has at least two inelastic scattering centers (A, B) arranged on the path, which cause spin flip scattering with different spin scattering directions and cause an electrochemical potential difference between the two terminals. Quantum device (40, 50) according to claim 6.

13. The functional principle of the quantum device is based on the anomalous quantum Hall effect, In the transmitting structure, at least one first region of the topological material is arranged so that two chiral currents of electrons having opposite spin directions circulate in opposite directions on a path connected to two opposite terminals, and at least two inelastic scattering centers are arranged in the first region on the path, and the inelastic scattering centers cause spin-flip scattering having different spin-flip scattering directions and different spin-flip scattering intensities, causing an electrochemical potential difference and a magnetic field between the two terminals. The quantum device of claim 6.

14. An array (100; 200) of two or more quantum devices according to any one of claims 1 to 13, The quantum devices are connected in series and / or in parallel or anti-parallel to each other. array.

15. Use of a quantum device according to any one of claims 1 to 13, or an array of two or more quantum devices according to claim 14, in an electrical device, comprising: connecting two terminals of the quantum device or two external terminals of the array of quantum devices to the electrical device; use.

16. 1. A method for converting thermal energy into one or more of a voltage, a current, a magnetic field, or a temperature inhomogeneity or an electron spin distribution, comprising: Manufacturing a quantum device according to any one of claims 1 to 13 or an array of quantum devices according to claim 14; connecting the quantum device to a load or other device configured to generate one or more of a voltage, a current, a magnetic field, or a temperature inhomogeneity or electron spin distribution; method.

Images