RU2474010C2 - Nanocomposite thermoelectric and method of its production - Google Patents

Abstract

FIELD: electricity.

SUBSTANCE: nanocomposite thermoelectric material comprises thermoelectric nanocrystals and alloying fullerene molecules distributed among them. Concentration of charge carriers is controlled by concentration of introduced alloying fullerene molecules that take electrons from nanocrystals of the thermoelectric and are quantum traps for electrons. The volume concentration of alloying fullerene molecules K_alloy, which are added into the nanocomposite thermoelectric, is defined by the difference of K_{nanocomposite} and K_initial, divided into the average number k of electrons taken from nanocrystals per one alloying fullerene molecule, namely; K_alloy=(K_{nanocomposite}-K_initial)/k.

EFFECT: modification of electric properties of materials due to variation of electric charge carrier concentration in nanocomposites.

3 cl, 2 tbl, 2 dwg

Description

1. The technical field to which the invention relates.

The invention relates to the field of nanostructured materials with modified electrical properties. One of the main applications of the invention is to improve the properties of nanostructured thermoelectrics.

2. The level of technology

It is known that with bulk doping of semiconductors, their electrical conductivity varies greatly [Bonch-Bruevich VP, Kalashnikov SG Semiconductor Physics. M .: Nauka, 1977, 679 p .; Ioffe A.F. Semiconductor thermocouples. M. - L., 1960, 188 p.]. For example, 0.0001 atomic percent of phosphorus in silicon reduces its resistance by 100,000 times. Element 5 of the group of the Periodic Table of the Elements of Phosphorus is considered as an example for doping silicon. Volumetric doping is carried out either by adding an alloying element to the melt, or by thermal diffusion, or by ion implantation. For the implementation of paired electronic bonds in the lattice of the element 4 of the silicon group, 4 electrons are needed. When doping silicon, the fifth valence electron of the phosphorus atom easily splits off and passes into the conduction band, and the phosphorus atom turns into a positively charged ion. The element of the 5th group is a donor impurity in the semiconductor of the 4th group. Elements of 3 groups, such as boron, have only 3 valence electrons. When doped with boron, local regions with a deficit of electrons are formed in the silicon lattice, i.e., positively charged regions called holes. Thus, the element of the 3rd group is an acceptor impurity in the semiconductor of the 4th group. Volumetric doping of semiconductors with donor or acceptor impurities leads to a change in the concentration of electrons and holes. This is used to optimize the thermoelectric efficiency of a semiconductor.

The disadvantage of volumetric doping of thermoelectrics for their optimization is that the possibilities of this method are almost exhausted today, since a change in concentration due to volumetric doping changes all the parameters in the formula ZT = S ² σT / k [Thermoelectric Handbook, Macro to Nano, DM. Rowe (editor), CRS, 2005] and the available bulk thermoelectric materials have the optimal parameters S, σ and k, (S is the Seebeck coefficient, σ is the electrical conductivity, k is the thermal conductivity, T is the average temperature of thermoelectric devices) obtained by volume doping materials. Thus, for further optimization of thermoelectrics, a method of changing the concentration of charge carriers without their bulk doping is required.

It is known that electron capture can be carried out by fullerene molecules, in particular, C _60. The fullerene molecule is capable of forming long-lived negative molecular ions (OMI) with bombarding electron energies of up to 10 eV [S. N. Yang, C. L. Pettiette, J. Conceicao , O. Chesnovsky, RESmalley. Chem. Phys. Lett., 139 (1987) 233; T. Jaffke, E. Illenberger, M. Lezius, S. Matejcik, D. Smith and TD Mark. Chem. Phys. Lett., 226 (1994) 213; J. Huang, HSCarman and RNCompton. J.Phys. Chem., 99 (1995) 1719.].

Experimental data are known on the capture of electrons by a C ₆₀ molecule. Only those cases of electron capture are considered in the literature, when the captured electrons are located on the lower unoccupied orbital. In this case, lines (1424, 1468, and 1574 cm ^-1 ) active in Raman scattering (Raman) become active in infrared (IR) absorption, and their position shifts to the low-frequency region by 6 cm ^-1 for each electron captured by the C ₆₀ molecule [VN Denisov et al. Optics and Spectroscopy, Vol. 76. No. 2, pp. 242-253 (1994); P. Rudolf, et al. Report of Brookhaven National Laboratory, contract number DE-AC02-98CH10886, Department of Energy, 2000].

The disadvantage of the described schemes and mechanisms of electron capture by the fullerene molecule is that to realize them, it is necessary to arrange metal atoms in the lattice of the fullerene molecular crystal between the fullerene molecules, which does not lead to the solution of the stated problem of doping a thermoelectric. In addition, they missed the possibility of electron capture by the fullerene molecule by another mechanism that is not accompanied, for example, by changes in the Raman spectrum or IR spectrum described in the literature.

The best traditional thermoelectric materials used in systems for converting heat to electricity have a quality factor (in the literature, quality factor ZT) about ~ 1 (which corresponds to an efficiency of about 4%). This limit limits the practical use of thermoelectrics. In practice, ZT> 2.5 values are required. In nanostructured systems, a ZT of 2.5 to 4 has been demonstrated [W. Kim et al. Phys. Rev. Lett. 96, 045901 (2006); Venkatasubramanian, et al. Nature, 413, 597 (2001); TCHerman, et al. Science 297, 2229 (2002)]. The main goal and effect of nanostructuring was to control the quality factor ZT = S ² σT / k by creating conditions for the effect of phonon blocking and electron transmission.

The disadvantage of the method of synthesis of these materials is that they are obtained by the method of homoepitaxial growth, which excludes the possibility of industrial production of these materials.

In the patent application [M. Popov, G. Pivovarov, E. Tat'yanin, V. Blank. Thermoelectric nanocomposite. Patent application WO 2009/110815. Published on September 11, 2009 (priority February 29, 2008)], which describes a nanocomposite material (based on thermoelectrics Bi ₂ Te ₃ , Sb ₂ Te ₃ or their alloys) and the method for its preparation. A nanocomposite material consists of thermoelectric nanoparticles coated with at least one layer of unmodified carbon material. The functional purpose of such layers deposited on nanoparticles was only the effective scattering of phonons, without changing other transport properties, in particular, the concentration of electrons and holes: i.e. in this application, the problem of creating boundaries that affect only thermal conductivity was solved. Meanwhile, the factor of modification of the electrical properties of the nanoparticles themselves by deposition of layers, i.e. no conditions were created for the capture of electrons by the deposited layers. Moreover, the application and the claims particularly emphasize that the deposited carbon layers are intact (substantially intact), which eliminates the charge redistribution between the layers and nanoparticles.

A method of obtaining the specified nanocomposite material is as follows. The nanopowder of the starting material of a thermoelectric is mixed with a carbon material (in particular, it is ground together with a carbon material). Then the resulting mixture is sintered by hot pressing (sintering temperature 350 ° C). The disadvantage of this method is the uncontrolled ingress of oxygen into the nanocomposite material and the absence of conditions for the capture of electrons by the carbon layer, which does not lead to the appearance of an alloying effect of the applied layer.

US 2004/0187905 A1 discloses a nanocomposite thermoelectric consisting of a combination of ceramic nanoparticles Be ₂ Te ₃ - Sb ₂ Te ₃ . (with an average size of less than 100 nm) and the method of its manufacture. The method is as follows. The bulk material is taken and ground to obtain ceramic nanoparticles, which are then sintered. Before starting the grinding process, additional material, for example, fullerene, can be added. During grinding, mechanical fusion of ceramic powder and fullerene occurs. In this case, fullerenes are destroyed in the process of mechanical fusion. The result is a nanocomposite with an indefinite composition, unpredictable features and poorly reproducible thermoelectric properties.

In [N. Gothard, J. E. Swart, and T. M. Tritt, Phys. Status Solidi A 207, 157 (2010)] presents a composite thermoelectric material consisting of (Be, Sb) ₂ Te ₃ microparticles and C ₆₀ fullerene crystals (which, as a material, are called fullerite). The composite consists of fragments of particles of 1-3 microns in size, consisting of fullerite, and a microcrystalline matrix, consisting of (Be, Sb) ₂ Te ₃ . The composite presented in principle cannot be considered as a nanocomposite and its transport properties are due to the additive properties of two components: microcrystalline (Be, Sb) ₂ Te ₃ and C ₆₀ fullerite with a negligible influence of the boundaries between (Be, Sb) ₂ Te ₃ and C ₆₀ fullerite due to the lack of nanofragmentation effect. The property of C ₆₀ molecules to bind electrons cannot change the concentration of charge carriers in a given composite of more than 0.1%, if we take into account the number of boundaries (Be, Sb) ₂ Te ₃ - C ₆₀ per unit volume of the composite. The property of a single molecule to bind electrons does not extend to bulk material consisting of these molecules (fullerite). Moreover, depending on the mechanical processing of the mixture, the concentration of charge carriers can either increase or decrease, which eliminates the possible doping effect of C ₆₀ (if it takes place, then the concentration of p-type charge carriers should increase in all cases). The observed properties indicate the absence of a nanocomposite effect. Indeed, commercially available thermoelectrics (just such a material was used in the work) have a small excess of Te, which is in a free state [Thermoelectric Handbook, Macro to Nano, DMRowe (editor), CRS, 2005]. Depending on the mixing conditions of (Be, Sb) ₂ Te ₃ and fullerite C ₆₀ , Te will dope C ₆₀ crystals, giving them semiconductor properties, as in the case of doping with other metals [VN Denisov et al. Optics and Spectroscopy, Vol. 76. No. 2, pp. 242-253 (1994)]. In the case of simple mixing, mixed C ₆₀ microcrystals turned out to be weakly doped with tellurium (high resistance, low concentration of charge carriers), and in the case of mixing in a ball mill, the doping effect of C _{60 with} tellurium is higher. Accordingly, based on the principle of additivity of the transport properties of the two components in the first case, there is a decrease in the average concentration of charge carriers, and in the second, an increase. Doped fullerite C _{60 is} not an effective thermoelectric and its presence in the composite only leads to a deterioration of thermoelectric properties, as shown in Fig. 4.

Thus, currently known technical solutions offer either a change in the concentration of charge carriers in a thermoelectric due to bulk doping, or the creation of carbon layers at the boundaries of nanoparticles without changing the concentration of charge carriers, or the production of nanocomposites with an undefined composition, unpredictable features, and poorly reproducible thermoelectric properties.

3. Disclosure of invention

The aim of the invention is to obtain a nanocomposite thermoelectric having optimized transport properties by changing the concentration of charge carriers in nanocrystals of a thermoelectric without changing their elemental composition.

The task is achieved in that in the patented nanocomposite thermoelectric, consisting of nanocrystals of a thermoelectric and doping fullerene molecules distributed among them, the concentration of charge carriers is controlled by the concentration of introduced doping fullerene molecules, which take electrons from the nanocrystals of the thermoelectric and are quantum traps for electrons. In particular, in the case of using an n-type thermoelectric as a material of nanocrystals, a decrease in the concentration of electrons in nanocrystals occurs by their capture in doping fullerene molecules, and in the case of using a p-type thermoelectric as a material of nanocrystals, a decrease in the electron concentration in the valence band of the nanocrystal material increases the concentration of p-type carriers and, as a result, modifies the electrical conductivity.

The indicated effect of doping thermoelectric nanocrystals with fullerene molecules is achieved mainly for nanoscale objects, since the number of electrons taken (or delivered) by the fullerene doping molecules is proportional to the surface area of the nanocrystal, and the number of charge carriers in the nanocrystal is proportional to its volume. Therefore, ceteris paribus, the change in the concentration of charge carriers due to the specified modification effect will decrease in proportion to the increase in particle size. In particular, to reduce the concentration of electrons in the nanocrystal material, it is proposed to use fullerene molecules (C ₃₆ to C ₈₂ ), in particular C _60, as doping fullerene molecules (quantum electron traps).

A method for the synthesis of a nanocomposite thermoelectric with a given concentration of charge carriers is proposed, in which the doping ability of the added fullerene molecules is taken into account. Accordingly, the concentration of doping fullerene molecules in a nanocomposite thermoelectric is a control parameter that affects the concentration of charge carriers in this material. A change in the concentration of charge carriers in a nanocomposite thermoelectric changes in proportion to the concentration of doping fullerene molecules in it and in proportion to the number of trapped electrons, on average, per doping fullerene molecule (relative resonance concentration of doping fullerene molecules).

A method is also proposed for synthesizing a nanocomposite thermoelectric with a constant concentration of charge carriers with respect to the starting material of the nanocrystal and with a given concentration of doping fullerene molecules in the nanocomposite thermoelectric, in which the electron concentration in the starting material of the nanocrystal of the thermoelectric is preliminarily increased by volume doping to compensate for the doping effect of the added doping fullerene molecules.

and

by free electron capture. Activation energy and effect of the internal energy on lifetime. // Chem. Phys. Lett., 226 (1994) 213; Huang J., Carman H.S. and Compton R.N. Low-Energy electron attachment to C₆₀ // J.Phys. Chem., 99 (1995) 1719; Туктаров Р.Ф., Ахметьянов Р.Ф., Шиховцева E.C., Лебедев Ю.А., Мазунов В.А. Плазменные колебания в молекулах фуллеренов при электронном захвате. // Письма ЖЭТФ 81, №4, 207-211 (2005).)In the claimed invention, the property of fullerene molecules to pick up electrons from nanocrystals is provided by the choice of "resonant" concentrations of fullerene molecules in the composite, at which the effective capture of electrons by molecules occurs. The presence of such resonant concentrations is due to a set of eigenvalues ψ _n (where n = 1, 2, 3 ....) of the wave function of trapped electrons in fullerene molecules. Effective capture occurs when the energy W of the electric field E at the interface between the nanocrystals of the thermoelectric and the fullerene layer coincides in magnitude with the energies W _{n of} the eigenvalues of the wave function of the captured electron. The presence of such resonance values is manifested, for example, in the nonmonotonic dependence of the electron capture cross section for free C ₆₀ fullerene molecules in crossed beams (Jaffke T., Illenbergen E., Lezius M., Matejcik S., Smith D. and Mark TD Formatin of

and

by free electron capture. Activation energy and effect of the internal energy on lifetime. // Chem. Phys. Lett., 226 (1994) 213; Huang J., Carman HS and Compton RN Low-Energy electron attachment to C ₆₀ // J.Phys. Chem., 99 (1995) 1719; Tuktarov R.F., Akhmetyanov R.F., Shikhovtseva EC, Lebedev Yu.A., Mazunov V.A. Plasma vibrations in fullerene molecules during electron capture. // Letters of ZhETF 81, No. 4, 207-211 (2005).)

In turn, the energy W of the electric field E at the interface between the nanocrystals of the thermoelectric and the fullerene layer is determined by the radius of the nanocrystals R, the number N _{C60 of} fullerene molecules on the surface of the nanocrystal, and the average number of electrons k captured by one molecule (capture coefficient).

Namely, in the approximation of the spherical shape of thermoelectric nanoclusters, the electric field E at the interface between thermoelectric nanocrystals and the C ₆₀ fullerene layer is described by Poisson's law:

where ε ₀ = 8.85 10 ^-12 C ² / (J m) is the dielectric constant;

e is the charge of an electron;

N = kN _C60 is the number of electron charges, expressed in terms of the number of N _C60 molecules on the surface of the particle and the number of electrons k captured by each molecule.

A free electron in such a field at a characteristic size h of the C ₆₀ molecule acquires energy:

From (1) and (2) we obtain the relation for a set of (N _C60 ) _n resonance values of the number of molecules on the surface of the nanoparticle, which provides the resonance capture energy W = W _n :

These values allow us to determine the set of resonant values of z _n volumetric concentration of fullerene (for example, C ₆₀ ) in the composite:

here V _C60 is the volume of the fullerene layer on the surface of a thermoelectric nanocrystal with a volume of V _TE ; v is the volume of 1 molecule of fullerene, equal to 1 nm ³ for fullerene C ₆₀ . Substituting expression (3) in (4), we obtain the relation:

At these values of the volume concentration of fullerene in the nanocomposite, the volume concentration of free electrons in semiconductors with the electronic type of conductivity decreases by kz _n / v100 per 1 cm ³ . And, conversely, in semiconductors with hole type conductivity, the volume concentration of holes increases by kz _n / v100 per 1 cm ³ . Thus, the "resonant" alloying of composite thermoelectric alloys is carried out.

4. A brief description of the drawings and the implementation of the invention

The Figure 1 presents the Raman spectra of light (Raman) of an n-type sample containing 2 vol.% C ₆₀ (upper spectrum) and the spectrum of the original C ₆₀ (lower spectrum) for comparison.

The Figure 2 presents the Raman spectra of light (Raman) of the p-type sample containing 2 vol.% C ₆₀ (upper spectrum) and the spectrum of the original C ₆₀ (lower spectrum) for comparison.

The Figure 3 presents the dependence of the resistivity r (a) and hole concentration (b) on z - the concentration of fullerene C ₆₀ in Bi _0.5 Sb _1.5 Te ₃ samples.

Table 1 presents the results of measuring the concentration of charge carriers in n-type samples.

Table 2 presents the results of measuring the concentration of charge carriers in p-type samples.

Table 3. shows the resonance values of the volume concentration z _n (%) of C ₆₀ fullerene in the nanocomposite at various resonance values of the capture energy W _n and the average number k of captured electrons per C ₆₀ molecule.

Table 4 shows the results of measuring the relative change in the concentration of holes p at resonant concentrations of C ₆₀ and the values of the capture coefficient k of the average number of electrons of each C ₆₀ molecule corresponding to these values.

Example 1

For the manufacture of nanocomposite thermoelectric take the following materials. As a material for the manufacture of nanocrystals using a thermoelectric Bi ₂ Te ₃ n-type, the purity of the material is not worse than 10 ^-4 . The material is treated with the addition of C ₆₀ (2% by volume). The fullerene C ₆₀ (99.98% purity), whose molecules have an electron affinity of 2.65 eV, which is significantly higher than the nanocrystal material component (Bi - 0.95 eV; Te - 1.97 eV), is taken as the material for the doping fullerene molecules. ; Sb - 1.07 eV).

In a fullerene molecular crystal, the distance between the centers of the molecules is 10 Å. Therefore, the volume occupied by one C ₆₀ molecule is ~ 10 ^-21 cm ^-3 and, accordingly, 2 volume percent (vol.%) Of the added C ₆₀ corresponds to a concentration of 2 × 10 ¹⁹ cm ^-3 molecules. This means that, depending on the relative resonant concentration of doping fullerene molecules k (the number of trapped electrons per average doping fullerene molecule), the electron concentration will decrease by k × 2 × 10 ¹⁹ cm ^-3 .

The starting materials (thermoelectric and fullerene) are preliminarily crushed to a particle size of less than 1 mm, then loaded into the AGO-ZU planetary mill. The materials are loaded into the mill and all subsequent operations with the processed materials up to the hot pressing operation are carried out in an Ar atmosphere (99.999% purity) at an oxygen concentration of less than 0.2 ppm in a glove box with a vacuum lock. Materials are processed in a mill under the following conditions. The carrier speed was 545 rpm, the drum speed was 1110 rpm. As grinding media use steel balls ⌀9 mm. Material processing time 60 min. The weight of the loaded components is: 11.47 g of an n-type thermoelectric and 0.05 g of C ₆₀ (2% by volume). The ratio of the weight of the balls to the weight of the processed materials is 1:10. Then, part of the resulting mixture in the amount of 2.5 g is loaded into a piston-cylinder chamber with an inner diameter of 14 mm, pressed at 3.5 kbar and sintered at a temperature of 395 ° C for 15 min under a pressure of 3.5 kbar. In this way, disc-shaped material samples are obtained. After sintering the samples, they are analyzed.

The state of C ₆₀ molecules was monitored by Raman scattering (Raman) as follows. Raman spectra were recorded on a setup with a microscopic attachment based on a TRIAX 552 spectrometer (Jobin Yvon) and a CCD Spec-10 detector, 2KBUV (2048 × 512) (Princeton Instruments), with a cut-off filter system for suppressing exciting laser lines. The source of exciting light is the STABILITE 2017 (Spectra) lasers.

Spectral range 200-1100 nm Spectral resolution 1 cm ^-1 Raman excitation laser wavelength 514 nm Spatial resolution 1-2 microns

In the Raman spectrum of fullerene C ₆₀ there are ten strong lines: 273 cm ^-1 , 431 cm ^-1 , 497 cm ^-1 , 708 cm ^-1 , 773 cm ^-1 , 1099 cm ^-1 , 1248 cm ^-1 , 1426 cm ^-1 , 1469 cm ^-1 , 1572 cm ^-1 . The lines in the region of 273 cm ^-1 and 497 cm ^-1 are respiratory; their appearance is associated with the simultaneous radial vibration of all carbon atoms in the molecule. The line in the region of 773 cm ^-1 is associated with the radial vibration of carbon atoms in the fragment of the molecule. A line in the region of 1469 cm ^-1 is associated with the presence of pentagons in the structure of the fullerene molecule [Science of Fullerenes and Carbon Nanotubes. MSDresselhaus, G. Dresselhaus and PCEklund. Elsevier Science, 1996].

The main lines in the Raman spectrum when excited by light with a wavelength of 514 nm are in the region of 273, 497 and 1469 cm ^-1 . It is these lines that are visible on the spectra (Figure 1), which suggests that C ₆₀ molecules did not collapse during processing. The absence of a line displacement of 1469 cm ^-1 allows us to conclude that charge transfer by a known mechanism, accompanied by a frequency shift (by 6 cm ^-1 for each electron captured by a C ₆₀ molecule), did not occur from a Bi ₂ Te ₃ nanocrystal to a fullerene molecule. The low intensity of the fullerene lines in the spectra is associated with its low concentration in the composite. The spectrum shown in FIG. 1 was averaged over 10 points.

When an electron is captured by a C ₆₀ molecule, the electron enters the internal cavity of the molecule, which is not accompanied by a noticeable change in the Raman spectra.

The average nanocrystal size in a nanocomposite thermoelectric is determined by X-ray diffraction by the Hall-Williamson method [G.K. Williamson and W.H Hall. Acta Metallurgies 1, 1953, 22-31]. The average nanocrystal size in a nanocomposite thermoelectric sample is 30 nm.

On the X-ray diffraction patterns, only lines of the initial Bi ₂ Te ₃ n-type are present and there are no additional lines. In particular, this indicates that there are no crystalline phases C ₆₀ and C ₆₀ located in the form of isolated fullerene molecules.

The concentration of carriers in the samples is determined by measuring the Hall coefficient.

Measurements are made by the 4-pin Van der Pauw method in a constant magnetic field with a strength of 2 Tesla. Contacts in the form of copper conductors with a thickness of 0.1 mm are welded along the edges of the thermoelectric sample in the form of a disk with a diameter of 14 mm and a thickness of 1 mm so that the distance between the contacts is equal, i.e. the contacts are welded at the ends of two diameters of the disk and these diameters make an angle of 90 °. The plane of the thermoelectric sample with the contacts is perpendicular to the direction of the magnetic field vector. The measurements are carried out at constant current through a sample of 0.1 A. The concentration of charge carriers is calculated by the formula:

n = 1 / (q R _Hav );

where q is the electron charge;

R _Hav - the value of the Hall coefficient of the material, calculated according to the measurement results by averaging the values of R _HC and R _HD : obtained with two opposite directions of the magnetic field vector:

where t is the thickness of the samples (in meters), B is the magnetic field (in tesla), V _{ij, kl} is the potential difference (in volts) between the i and j contacts when current flows through the k and l contacts. The indices i, j, k, l mean the number of the contact (4 contacts are used in total). The numbering of contacts is sequential in a circle. The signs "+" and "-" in the designations determine the direction of the current and magnetic field.

The results of measuring the concentration of charge carriers are shown in table 1.

As can be seen from table 1, the concentration of electrons (charge carriers) decreased by 7 × 10 ¹⁹ cm ^-3 . The concentration of positive charge carriers is denoted by the letter p, and the negative by n. In the general case, the concentration is denoted by K, implying a positive sign for positive charges and a negative sign for negative ones. This is due to the possibility of capturing by a single C ₆₀ molecule at the same time 2 or more electrons. From table 1, the resonant concentration of the doping fullerene molecules is determined: on average, 3.5 electrons (k = 3.5) fall on one fullerene molecule.

When an electron is captured by a C ₆₀ molecule, the electron enters the internal cavity of the molecule, which is not accompanied by a noticeable change in the Raman spectra.

Once in the inner region of the spherical C ₆₀ molecule, the electron can no longer leave the closed molecule if its energy is in the range of 0.3 eV ≤E≤18 eV. Indeed, the characteristic cell size L C ₆₀ - (pentagon or hexagon) is not more than 2.8 Å. This corresponds to the minimum electron energy (in the case of a de Broglie plane wave), capable of leaving the internal volume C ₆₀ - 18 eV. E = p ² / 2m, p = (h / L), where h is the Planck constant, m is the mass of the electron.

The minimum energy of electrons that can be in the internal volume of a C ₆₀ molecule should be more than 0.3 eV, which is determined from the principle of uncertainty with the diameter of the C ₆₀ -D≈7 Å molecule and the curvature of the free electron path inside the C ₆₀ molecule.

In resonant electron capture in the specified energy range, a C ₆₀ molecule is sequentially affected by the following factors:

1) significant polarization of the molecule, due to its convex shape, when interacting with an incident electron;

2) acceleration of an electron by the field of the ion of the carbon atom closest to the approaching electron;

3) the lack of significant polarization of the molecule (due to its concavity) after the electron enters the C ₆₀ molecule.

Accordingly, after polarization of the molecule by an incident electron, the electron accelerates in the field of the carbon ion nearest to it and acquires the kinetic energy necessary for penetration into the internal volume of C ₆₀ . After the rounding of the nucleus of the ion, the electron leaves the accelerating ion, already being in the internal volume of the C ₆₀ molecule. According to the geometric structural features of the C ₆₀ molecule, the electron inside the molecule cannot substantially polarize it. And therefore, this “classic” electron cannot get out of the molecule.

Stabilization of long-lived molecular negative ions, therefore, occurs due to the discharge of incident electrons by carbon atoms into a trap - the internal volume of the molecule. As a result, electron traps are formed that provide an alloying effect.

Therefore, in the considered example, the conditions for electron capture by quantum traps (doping fullerene molecules) were realized, the doping effect of introducing quantum traps (doping fullerene molecules) was shown, and the resonance concentration of doping fullerene molecules was determined.

Example 2

For the manufacture of nanocomposite thermoelectric take the following materials. As the starting material for the manufacture of nanocrystals, a p-type thermoelectric (with hole conductivity) Bi ₂ Te ₃ containing 26 atomic% Sb ₂ Te _{3 is} taken; the purity of the material is not lower than 10 ^-4 . The concentration of the main charge carriers (holes) in the initial material (p _initial ) was 1.63 × 10 ¹⁹ cm ^-3 according to the Hall measurements. To improve the electrical properties, it is necessary to increase the concentration of holes in the nanocomposite thermoelectric (p _{nanocomposite} ) to 4 × 10 ¹⁹ cm ^-3 . This requires preliminary determination of the resonant concentration k.

As in example 1, C ₆₀ fullerene (99.98% purity) is taken as the material for the doping fullerene molecules, 4 volume percent (vol.%) Of added C ₆₀ corresponds to a concentration of molecules of 4 × 10 ¹⁹ cm ^-3 . This means that in the case of capture by C ₆₀ molecules on average of electrons (minority charge carriers), the hole concentration will increase by 4k × 10 ¹⁹ cm ^-3 .

The starting materials (thermoelectric and fullerene) are preliminarily crushed to a particle size of less than 1 mm, then loaded into the AGO-ZU planetary mill. The materials are loaded into the mill and all subsequent operations with the processed materials up to the hot pressing operation are carried out in an Ar atmosphere (99.999% purity) at an oxygen concentration of less than 0.2 ppm in a glove box with a vacuum lock. Materials are processed in a mill under the following conditions. The carrier speed was 545 rpm, the drum speed was 1110 rpm. As grinding media use steel balls ⌀9 mm. Material processing time 60 min. The weight of the loaded components is: 11.47 g of a p-type thermoelectric and 0.1 g of C ₆₀ (4 vol%). The ratio of the weight of the balls to the weight of the processed materials is 1:10. Then, part of the resulting mixture in the amount of 2.5 g is loaded into a piston-cylinder chamber with an inner diameter of 14 mm, pressed at 3.5 kbar and sintered at a temperature of 395 ° C for 15 min under a pressure of 3.5 kbar. In this way, disc-shaped material samples are obtained. After sintering the samples, they are analyzed.

The state of C ₆₀ molecules was monitored, as in Example 1, by Raman spectroscopy using a Raman spectroscopy setup with a microscopic attachment based on a TRIAX 552 spectrometer (Jobin Yvon) and a CCD Spec-10 detector, 2KBUV (2048 × 512) (Princeton Instruments ), with a system of cut-off filters to suppress exciting laser lines. The source of exciting light is the STABILITE 2017 (Spectra) lasers.

Spectral range 200-1100 nm Spectral resolution 1 cm ^-1 Raman excitation laser wavelength 514 nm Spatial resolution 1-2 microns

The main lines when excited by light with a wavelength of 514 nm are in the region of 273, 497 and 1469 cm ^-1 (among them the most intense mode is the pentagonal). It is these lines that are visible on the spectra (Figure 2), which suggests that C ₆₀ molecules did not collapse during processing. The absence of a line displacement of 1469 cm ⁻¹ allows us to conclude that charge transfer by a known mechanism, accompanied by a frequency shift (by 6 cm ⁻¹ for each electron captured by a C ₆₀ molecule), did not occur from a Bi ₂ Te ₃ nanocrystal to a fullerene molecule. The low intensity of the fullerene lines in the spectra is associated with its low concentration in the composite. The spectrum shown in FIG. 2 was averaged over 10 points.

As in example 1, when an electron is captured by a C ₆₀ molecule, the electron enters the internal cavity of the molecule, which is not accompanied by a noticeable change in the Raman spectra.

The average nanocrystal size in a nanocomposite thermoelectric is determined by X-ray diffraction by the Hall-Williamson method [G.K. Williamson and W.H Hall. Acta Metallurgica 1, 1953, 22-31]. The average nanocrystal size in a nanocomposite semiconductor sample is 30 nm.

On the X-ray diffraction patterns, only lines of the initial Bi ₂ Te ₃ n-type are present and there are no additional lines. In particular, this indicates that there are no phases of C ₆₀ and C ₆₀ is in the form of isolated fullerene molecules.

The concentration of carriers in the samples is determined by the Hall effect, similarly to example 1. The measurements are carried out using the 4-pin Van der Pauw method in a constant magnetic field with a strength of 2 Tesla. Contacts in the form of copper conductors with a thickness of 0.1 mm are welded along the edges of the thermoelectric sample in the form of a disk with a diameter of 14 mm and a thickness of 1 mm so that the distance between the contacts is equal, i.e. the contacts are welded at the ends of two diameters of the disk and these diameters make an angle of 90 °. The plane of the thermoelectric sample with the contacts is perpendicular to the direction of the magnetic field vector.

The results of measuring the concentration of charge carriers are shown in table 2.

As can be seen from table 2, the hole concentration increases by about 5 × 10 ¹⁹ cm ^-3 , for 4% concentration of C ₆₀ . From table 2, the resonant concentration of dopant fullerene molecules is determined: 1.3 electrons per average fullerene molecule (k = 1.3). Knowing k, determine the concentration (K _C60 ) C ₆₀ necessary to obtain a nanocomposite material with a hole concentration p _{nanocomposite} = 4 × 10 ¹⁹ cm ^-3 as follows: K _C60 = (p _{nanocomposite} -p _initial ) / k = 2 × 10 ¹⁹ cm ^-3 , which corresponds to 2 volume% C ₆₀ .

The above procedure for producing a nanocomposite thermoelectric is repeated with a concentration of 2 volume% C ₆₀ . In the material thus obtained, the concentration of charge carriers is 3.5 × 10 ¹⁹ cm ^-3 (Table 2), which is close to the given parameters.

Therefore, in the considered example, the conditions for electron capture by quantum traps (doping fullerene molecules) were realized, the doping effect of introducing quantum traps (doping fullerene molecules) was shown, and the resonance concentration of doping fullerene molecules was determined. On the basis of k, the concentration (K _C60 ) of С ₆₀ is determined to obtain a nanocomposite material with a hole concentration p of the _{nanocomposite} as follows K _C60 = (p _{nanocomposite is} p is the _original ) / k, where p _{is the} concentration of the main charge carriers (holes) in the original thermoelectricity.

Example 3

For the manufacture of nanocomposite thermoelectric take the following materials. As starting materials for the manufacture of nanocrystals take p-type thermoelectric Bi _0.5 Sb _1.5 Te ₃ ; the purity of the material is not lower than 10 ^-4 . The concentration of the main charge carriers (holes) in the initial material (p _initial ) was 1.5 × 10 ¹⁹ cm ^-3 according to the Hall measurements. To reduce thermal conductivity without increasing electrical resistance, it is necessary to increase the concentration of holes in a nanocomposite thermoelectric (p _{nanocomposite} ) to 6 × 10 ¹⁹ cm ^-3 . This requires preliminary determination of the resonant concentration z _{n of} fullerene C ₆₀ .

As in example 1, fullerene C ₆₀ (99.98% purity) is taken as the material for the doping fullerene molecules, 0.6 volume percent (vol.%) Of added C ₆₀ corresponds to a molecular concentration of 0.75 × 10 ¹⁹ cm ^-3 . This means that if C ₆₀ molecules capture an average of k = 6 electrons, the hole concentration will increase by 6 × 10 ¹⁹ cm ^-3 .

As in example 1, the starting materials (thermoelectric and fullerene) are pre-crushed to a particle size of less than 1 mm, then loaded into the AGO-ZU planetary mill. The materials are loaded into the mill and all subsequent operations with the processed materials up to the hot pressing operation are carried out in an Ar atmosphere (99.999% purity) at an oxygen concentration of less than 0.2 ppm in a glove box with a vacuum lock. Materials are processed in a mill under the following conditions. The carrier speed was 545 rpm, the drum speed was 1110 rpm. As grinding media use steel balls ⌀9 mm. Material processing time 60 min. The weight of the loaded components is: 11.47 g of a p-type thermoelectric and 0.1 g of C ₆₀ (4 vol%). The ratio of the weight of the balls to the weight of the processed materials is 1:10. Then, part of the resulting mixture in the amount of 2.5 g is loaded into a piston-cylinder chamber with an internal diameter of 14 mm, pressed at 3.5 kbar and sintered at a temperature of 395 ° C for 15 min under a pressure of 3.5 kbar. In this way, disc-shaped material samples are obtained. After sintering the samples, they are analyzed.

The state of C ₆₀ molecules was monitored, as in Example 1, by Raman spectroscopy using a Raman spectroscopy setup with a microscopic attachment based on a TRIAX 552 spectrometer (Jobin Yvon) and a CCD Spec-10 detector, 2KBUV (2048 × 512) (Princeton Instruments ), with a system of cut-off filters to suppress exciting laser lines. The source of exciting light is the STABILITE 2017 (Spectra) lasers.

Spectral range 200-1100 nm Spectral resolution 1 cm ^-1 Raman excitation laser wavelength 514 nm Spatial resolution 1-2 microns

The main lines when excited by light with a wavelength of 514 nm are in the region of 273, 497 and 1469 cm ^-1 (among them the most intense mode is the pentagonal). It is these lines that are visible on the spectra (Figure 2), which suggests that C ₆₀ molecules did not collapse during processing. The absence of a line displacement of 1469 cm ^-1 allows us to conclude that charge transfer according to the known mechanism, accompanied by a frequency shift (by 6 cm ^-1 for each electron captured by a C ₆ molecule), from a Bi-Sb-Te nanocrystal to a fullerene molecule does not occurred. The low intensity of the fullerene lines in the spectra is associated with its low concentration in the composite. The spectrum shown in FIG. 2 was averaged over 10 points.

As in example 1, when an electron is captured by a C ₆₀ molecule, the electron enters the internal cavity of the molecule, which is not accompanied by a noticeable change in the Raman spectra.

The average nanocrystal size in a nanocomposite thermoelectric is determined by X-ray diffraction by the Hall-Williamson method. The average nanocrystal size in a nanocomposite semiconductor sample is 30 nm.

On the X-ray diffraction patterns there are only lines of the original Bi _0.5 Sb _1.5 Te ₃ and there are no additional lines. In particular, this indicates that there are no phases of C ₆₀ and C ₆₀ is in the form of isolated fullerene molecules.

The concentration of carriers in the samples is determined by the Hall effect, similarly to example 1. The measurements are carried out using the 4-pin Van der Pauw method in a constant magnetic field with a strength of 2 Tesla. Contacts in the form of copper conductors with a thickness of 0.1 mm are welded along the edges of the thermoelectric sample in the form of a disk with a diameter of 14 mm and a thickness of 1 mm so that the distance between the contacts is equal, i.e. the contacts are welded at the ends of two diameters of the disk and these diameters make an angle of 90 °. The plane of the thermoelectric sample with the contacts is perpendicular to the direction of the magnetic field vector.

For the average nanocrystal radius R = 30 nm, the effective diameter of fullerene molecules C ₆₀ h = 1 nm, we have:

Accordingly, the calculated values of the resonance values of the volume concentration of fullerene C ₆₀ in the nanocomposite at different resonance values of the capture energy W _n and the average number k of trapped electrons per C ₆₀ molecule are presented in Table 3.

In figure 3. experimental dependences of the resistivity r, hole concentration p, and their Hall mobility on z — fullerene concentration C ₆₀ in Bi _0.5 Sb _1.5 Te ₃ samples are presented.

Table 4 shows the results of measuring the relative change in the concentration of holes p at resonant concentrations of C ₆₀ and the values of the capture coefficient k of the average number of electrons of each C ₆₀ molecule corresponding to these values.

As follows from Table 3 and Table 4, in these materials a resonant change in the concentration of free charge carriers is observed in the concentration range z _n = 0.6 and 1.3%. Thus, alloying of said alloys is carried out with the capture coefficient values k = 6 and k = 2, respectively.

Thus, in order to manifest the property of fullerene molecules described in claim 1, to take electrons from thermoelectric nanocrystals in the amount of k pieces per 1 molecule, it is necessary to use relation (5) to select the volume concentration of fullerene molecules in a nanocomposite of a thermoelectric with fullerene. In this case, fullerene molecules can be different (for example, C ₆₀ , C ₇₀ , C ₈₀ , C ₈₂ , etc.). Claim 2 specifies that in this case, C _{60 is} used as fullerene molecules.

Table 1 Charge carrier concentration measurement results Conductivity type Concentration C ₆₀
(volume%) Carrier concentration
charges × 10 ¹⁹ cm ^-3 n - type (electronic) 0 9.5 n - type (electronic) 2% 2.5

table 2 Charge carrier concentration measurement results Conductivity type Concentration C ₆₀
(volume%) Carrier concentration
charges × 10 ¹⁹ cm ^-3 p - type (hole) 0 1.6 p - type (hole) 2% 3.5 p - type (hole) four% 7

Table 3 Resonant values of the volume concentration z _n (%) of C ₆₀ fullerene in a nanocomposite at different resonance values of the capture energy W _n and the average number k of captured electrons per C ₆₀ molecule w _n , eV 1,5 four 5 6 k one 0.8 2.2 2,8 3.3 2 0.4 1,1 1.4 1.65 6 0.13 0.36 0.47 0.55

Table 4 The relative change in the concentration of hole electrons Δр at resonance concentrations of C ₆₀ and the corresponding values of the capture coefficient k of the average number of electrons by each C ₆₀ molecule in nanocomposites with Bi _0.5 Sb _1.5 Te ₃ . Concentration C ₆₀ , vol.% N _C60 , concentration
C ₆₀ , cm ^-3 Δn (or Δр) The number of attached electrons at C ₆₀ k = Δn / N _C60 0.6 0.75 × 10 ¹⁹ 4.5 × 10 ¹⁹ 6 1.3 1.6 × 10 ¹⁹ 3.2 × 10 ¹⁹ 2

Claims (3)

1. Bi-Sb-Te-based _{nanocomposite} thermoelectric with a concentration of charge carriers K is a _{nanocomposite} , characterized in that it is formed by a combination of thermoelectric nanocrystals and doping fullerene molecules distributed among them, which take electrons from these nanocrystals and are quantum traps for electrons, while the concentration of doping fullerene molecules K _legir , which are added to the thermoelectric, is determined by the difference K of the _{nanocomposite} and the concentration of charge carriers in the nanocrystals of the thermoelectric K _initial , divided by the average number k of electrons taken from nanocrystals per one doping fullerene molecule, namely: K _legir = (K _{nanocomposite} -K _initial ) / k , where k = 3.5. 2. Nanocomposite thermoelectric according to claim 1, characterized in that the doping fullerene molecules are composed of C ₆₀ fullerene molecules. 3. A method of producing a Bi-Sb-Te-based nanocomposite thermoelectric with a charge carrier concentration K; a _{nanocomposite} consisting of a combination of thermoelectric nanocrystals with a charge carrier concentration K is the _{initial one} and doping fullerene molecules distributed among the thermoelectric nanocrystals, which is that the volume concentration of doping fullerene molecules K _legier , which is added to the nanocomposite thermoelectric, is determined by the difference between K _{nanocomposites} and K _initial , divided by the average number k taken from nanocrysts of electrons per one doping fullerene molecule, namely: K _legir = (K _{nanocomposite} -K _initial ) / k, where k = 3.5.

Images