JPH05503608A - Solid state quantum mechanical electron and hole wave devices - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] Solid-state quantum-mechanical electron and hole wave device l Mamatachi 11 The present invention relates to solid state quantum mechanical electron and hole wave devices and their fabrication. (a) solid-state quantum mechanical electron and hole wave devices; devices, such as low-pass filters, high-pass filters, narrowband and wideband notch filters, Narrowband and wideband bandpass filters and impedance transformers (these include (without limitation): (b) Biased semiconductor superlattice structure, e.g. ballistic transistor For example, a tunable bias can act as a thermionic emitter in the star. As-type electronic wave interference filter/emitter (including but not limited to): and (c ) Regarding semiconductor quantum well electron and hole waveguides.

Recent advances in semiconductor growth technology, particularly molecular beam epitaxy (MBE) and and metal-organic chemical vapor deposition (MOCVD), one skilled in the art can produce precise monolayer compositions. It has been a long time since we were able to grow a multilayer superlattice structure with controlled composition. Narrow and wide bandgap semiconductor materials such as GaAs and Ga+-11A Sequentially grown layers of IIIAs are manufactured and widely used in multi-quantum well structures. has brought about In fact, these superlattice structures can be transformed into superlattice structures exhibiting resonant tunneling. There are references in the prior art for use in quantum/multiple quantum well devices. . Specifically, in such devices, superlattices can be formed by epitaxially growing successive layers of gap semiconductor material. The materials and layer widths of these devices are limited by spatial quantization effects in adjacent wells. The resulting quantum states are selected to be coupled. Furthermore, such devices In this case, the interaction of these bonded states is such that charge carriers exhibit tunneling phenomena. This leads to the formation of mini-bands of allowable energy.

The size of the superlattice device exhibiting the above-mentioned resonant tunneling phenomenon disclosed in the prior art For example, Ja, J, A1. Ph s, T in 26°L1332 (1987), Inata, S, Muto, Y.

Nakata, S. 5asa, T. Fujii and S. Hiyam Article by izu; 'M, A, Reed, J, W, Lee and H-L in 6) Article by Tsai: and A1. Ph s, tett, 52.1 442 (1988), S., Y., Chou and J., S., Harr. Such a device is a high frequency microwave 1, which is quite interesting as an oscillator, for example A1.　Ph　s, Let t,, 43.588 (1983) Nio + T, C.

L, G, 5aliner, L D, Goodhue, P, E, Tannew Articles by ald, C, D, Parker, and Peck: and ``Ni Button F Engineering'' Search 1- is 2, TCIL in 50.333 (1'987), G, 5ollner, E, R, Brown, W, D, Good See the article by Hue and H.Q. Lee, but recently: The resonant tunneling phenomenon that passes through a multilayer structure consisting of three wells and four barriers is aAs layer 0 e.g. A1. demonstrated in the AlGaAs material system.　Ph　s , Lett.

C, J, Summers, K, in 52, 132 (1988) by F. Brennan, A. Torabi and H. M. Harris. In addition, these structures can be used in electroluminescent devices, photodetectors, etc. and potential applications as high-energy injectors for high-velocity ballistic transistors. be.

For example, A1. In Ph.s., Lett., 48.806 (1986) Article by C. J. Summers and F. Brennan: In Muesara-eML, 61.5410 (1987), FBrenn Article by an and C, J, Summers; and’ rEEE J, In Quantum Electron, QE-23, 320 (1987) See article by Brennan, F. and Summers, C.J. It's good to do that.

In addition to the above, the following prior art reference documents, namely Elect ronics Letters, Vol, 21°No, 19.882 ( T, Nakagawa, H. (1985).

Imamoto, T, Sakamoto, T, Kojima, Articles by K, 0hta and N, J, Kawai, as well as Guns and Retrospectives Tomo and Microstructures, Academic Press Inc., Vol.

1, No. 2, 1985. I) r Des in L 187-192 ignPrinciples for CHIRP 5uperlattice T, Nakagava, N, J, entitled s DevicesJ. There are articles by Kawai and 0hta, which also discuss the negative differential resistance effect. barrier by using the superlattice to act as a barrier or low-permeability barrier contact. Obtaining mini-bands and forbidden energy bands with charge carrier energy exceeding the height This is disclosed. In addition to the above, there are currently high Significant interest has been expressed in obtaining devices that exhibit speed computation. in particular, The main factor that affects the speed of semiconductor devices is the electron flow from the input terminal to the output terminal. This is the running time. If there is no scattering phenomenon involved, i.e. "ballistic" or " If it is possible to get electrons to pass through a semiconductor by "collisionless" movement, This means that the running time is minimized and the potential speed in the device is maximized. is expected. The possibility of ballistic movement in semiconductor materials has recently been demonstrated in practice in GaAs. provided by the experimental results. This result is Ej! L positive embryo, 1988. pp, 1 3-16 r Ba1listic Electrons and) Ioles 0bserved in a Sem1conductorJ, M, This is disclosed in a non-prior art article by He1lblueI11.

As such, the length of the area to be traversed is the mean free path of the electron (mfpl When the order of magnitude is the same as that of For example, the mfp in silicon is 100 angstroms (Al , but the mfp for electrons in GaAs is approximately 10 times larger. It is.

In an interest to investigate the utility of fabricating such ballistic electronic devices, GaA The experiment described in the previous paper, in which the s-layer was sandwiched between two AlGaAs alloy layers, was It is. AlGaAs has the same lattice constant as GaAs, and as a result its It is suitable for use in such experiments because it can be epitaxially grown on It is reported that it is a suitable material. In addition, further experimental reports show that ballistic Even in GaAs, the movement of holes is lower than that caused by the movement of electrons. In case.

- It was shown that this occurs due to the unique band structure of the valence band of GaAs.

In view of the above, electron and/or hole filter devices, such as low-pass filters, High-pass filters, notch filters and band-pass filters are commonly used in the art. ing. These are useful for solid-state devices that require energy selectivity, e.g. can act as a high-energy electron injector in devices such as electroluminescent devices, photodetectors, ballistic transistors and filter/emitter structures. and electronic or electronic devices for use in manufacturing equivalent integrated optical devices. can be used to fabricate hole waveguide devices.

Kotan JJ”! Embodiments of the present invention satisfy the above needs in the prior art and provide three types of Embodiments of (1) unbiased device: (2) biased device and (3) relating to waveguides.

1. Biased device Embodiments of the present invention provide a solid solution capable of exhibiting energy selectivity, among other things. by providing Tate quantum mechanical electron and hole wave devices. Specifically, embodiments of the present invention meet the above-mentioned needs for solid-state, low-frequency filters. Filter; high-pass filter, narrowband and wideband notch filter: narrowband and wideband notch filter; Impedance transformer equivalent to optical anti-reflection coating: and Embodiments of the present invention that include high reflectivity devices that correspond to mirrors provide a means for the electrons to be substantially ballistic. Contains a multilayer superlattice structure consisting of materials that help electrons move to into such a structure with energy exceeding the potential energy barrier. When inserted into the dielectric film, the structure is similar to that which occurs during electromagnetic wave propagation in the dielectric film. It exhibits similar electronic interference effects. These embodiments are based on the method of the invention. Therefore, the phase quantity is proportional to the square root of the product of the electron kinetic energy and the electron effective mass. Furthermore, by using the first solid refractive index of The second solid curve with respect to the amplitude is proportional to the square root of the electron effective mass divided by the electron effective mass. It is designed by using the refractive index. These refractive indices are related to quantum mechanical electron waves and It is completely similar to the electromagnetic optical wave, and in this case, the wave number vector of the electron k = (2rs” fE-V) j””/l+ (1) is the optical wave vector Similar to, the electron wave amplitude refractive index ns (amplitude) Q: [fE-Vl/a"]"' '' (2) is similar to the optical refractive index.These refractive indices are: - refraction at the boundary It is used in the formula to calculate the rate and transmittance.

This equation converts existing optical interference filter designs into designs for semiconductor devices of the present invention. For example, optical quantities and fixed Using these mappings between A corresponding solid state with Sheff elliptic function or other well-known filter characteristics In particular, the solid-state electron wave device of the present invention may be transformed into a design for a filter device. One embodiment of the vise is made from an alloy of AlGaAs and GaAs. Additionally, solids of the present invention include solid-state equivalents of Apple-Bello optical interference filters. State filter devices use transistor structures to increase their speed. It can be integrated during construction.

The usefulness of mapping between electromagnetic optical waves and quantum mechanical electron waves is due to the fact that electrons bounce through solid materials. moving in a path, i.e. electrons are scattered due to deviations from crystalline perfection relies on moving through solid materials without

Ballistically moving electrons have energy that exceeds the potential barrier in solid materials. , exhibiting quantum mechanical plane wave behavior, and furthermore, these plane waves are These coherent waves follow the behavior of electromagnetic waves traveling in a dielectric to maintain the phase of refraction, reflection, interference, and diffraction in a manner similar to that of Semiconductor doping is the invention Although not critical to the implementation, to avoid scattering in the material, It is preferable to avoid doping in the active area; doping must be applied. This is because the device of the present invention is easier to manufacture if Additional benefits regarding the service, 4. I will provide a.

Electron wave propagation with energy exceeding the potential barrier is caused by quantum It can be described mathematically by a mapping between mechanical electron waves and electromagnetic optical waves in dielectric materials. However, for example, a semiconductor superlattice interference filter design is simply an optical thin film filter design. It cannot be a copy of the data design, because the light that can be realized Although scientific design is originally limited by the refractive index of the material used, The design of a superlattice interference filter, e.g. λ, depends on the following facts, namely ( a) the thickness of the layers in the superlattice structure is limited to an integral multiple of the monolayer thickness, and (b) the need for substantially collision-free movement of charge carriers in the usable composition of the material; This is because it is limited by limiting the range. The need to move without collision as it often precludes the use of material compositions with indirect band gaps. Therefore, this latter limitation arises. Nevertheless, it is obvious to a person skilled in the art that However, using trial and error methods to determine which designs are physically feasible. However, preferred embodiments of the method of the invention described in the detailed description include , including a systematic method for determining superlattice structures that satisfy appropriate physical constraints.

Bias device Embodiments of the invention serve as thermionic emitters of ballistic transistors, for example. Provides a biased semiconductor superlattice tunable electronic interference filter/emitter that can thereby satisfying the above needs in the art. In particular, one of the invention Embodiments provide near-ballistic energy with electron energy above the superlattice potential barrier. A biased semiconductor superlattice film exhibiting energy selectivity for electron waves propagating in Furthermore, the layers of the biased superlattice structure, including the filter/emitter, are is a multiple of one-quarter or one-half of the electron wavelength, and the width of the quantum well barrier is a multiple of one-quarter or one-half of the electron wavelength. High electron refractive index, which is adjusted in the direction of the Alternating refractive index and low electronic refractive index.

The bias type semiconductor superlattice filter/emitter of the present invention is obtained by using the method of the present invention. Therefore, the optical thin film interference filter determined by the existing optical interference filter design method The design can be applied to a semiconductor device of the present invention as described above with respect to an unbiased device. It is formed by metamorphosing into However, this bias type device In this case, V in the above equations (1) and (2) is It is a function of the position inside.

``Guidance'' E Tribute Embodiments of the invention provide semiconductor quantum well electron or hole slab waveguides. For example, such an electronic waveguide satisfies the above needs in the art by are useful in high-speed circuits and as central components in electron or hole waveguide integrated circuits. It should be for use. Specifically, electronic slab waveguides have semiconductor layers that are substantially elastic. When the thickness and composition of the semiconductor layer is changed according to the method of the present invention, The semiconductor substrate layer, the semiconductor film that provides the potential well layer and a semiconductor cover layer.

In particular, according to the invention, the electron energy in the well and the substrate layer and cover The electron energy that exceeds one or both of each potential energy barrier in the layer On the other hand, there are electronic waveguide modes. Furthermore, due to dispersion, the lower In contrast to the behavior of electromagnetic waveguides, which only block energy, each electronic waveguide mode , the substrate layer and/or cover also blocks higher energies and refracts the electron waves. Put it in one layer.

i ward! star emperor j The invention will be apparent from the following detailed description when considered in conjunction with the accompanying drawings. can be completely understood.

FIG. 1 schematically shows an electronic superlattice interference filter manufactured according to the present invention. Figure 2 shows the energy level diagram and diagram of the electron superlattice interference filter in Figure 1. diagrammatically depict the material composition; Figure 3 shows an optical thin film interference filter that is the optical equivalent of the electron superlattice interference filter in Figure 1. It schematically shows a refractive index diagram of a filter; Figure 4 schematically shows the transmittance of the electron superlattice interference filter shown in Figure 1. : FIG. 5 shows a bias type electronic wave superlattice interference filter/element manufactured according to the present invention. This diagram schematically shows the energy level diagram and material composition of the mitter.

Figure 6 schematically shows the transmittance of the electron superlattice interference filter shown in Figure 5. : FIG. 7 shows the energy distribution of the asymmetric quantum well slab waveguide fabricated according to the present invention. – It graphically shows the level diagram and material composition: ゛ Figure 8 plots the electron guided mode propagation constant as a function of total electron energy. Gao, asAlo, +sAS substrate layer, GaAs film layer and Bi Gao? Quantum well slab waveguide consisting of 1llAIO, 1oAs power layer Diverse attenuation mode, waveguide mode, substrate mode and radiation mode regions for is shown along with the mode dispersion curve of the fundamental mode M0 for the film layer thickness; Figure 9 shows a GaAlAs material system with a GaAs film layer as thick as 10 monoatomic layers. The wave function Uv for the M0 mode of the slab waveguide with various electron energies This is a plot of .

For ease of understanding, the same symbols are used to designate elements common to each drawing. Ta.

Emperor Bidan and Emperor Hereinafter, the present invention will be described as (1) a non-biased device, (2) a biased device, and (3) Embodiments belonging to three categories of waveguides will be explained in more detail.

Non-bias device FIG. 1 schematically shows an electronic superlattice interference filter 100 manufactured according to the present invention. This is what is shown. This interference filter 100 according to the invention comprises a layer of superlattice 120. 101 to 109, the potential of the material forming the input M131 and the output layer 132 Narrowband transmission filter for a specific incident electron kinetic energy greater than the barrier height It is a ruta. As explained in detail below, the thickness of eyebrows 101-109 and The composition and height of the potential barrier of the materials forming 9101 to 109 are according to the present invention. According to the law, the mapping between quantum mechanical electron waves in a semiconductor and electromagnetic light waves in a dielectric is According to the method of the present invention, this mapping is used to determine the existing Thin-film optical design techniques and existing optical designs can be combined with solid-state quantum mechanical equivalents. This is advantageous because it can be applied to designing electronic wave devices.

FIG. 2 shows an energy level diagram and an energy level diagram of the electron wave superlattice interference filter 100 of FIG. This shows the material composition. As shown in FIG. 2, the input layer 131 and the output N1 32 consists of Gao, 5J1o, and +JS, respectively. The superlattice 120 has a quarter 1 wavelength GaAs layer 101. 1/4 wavelength Gao, 5sAlo4sAs layer 102 ．． Quarter wavelength GaAs layer 103. Quarter wavelength Gao, 5sA1o, aJ s layer 104, half wavelength GaAs layer 105. quarter wavelength Gao, 1sA1o, 4IAS layer 106.1/4 wavelength GaAs layer l○7.1/4 wavelength Gao, 5 sAlo, 411AS 108, and quarter wavelength GaAs JW109. Ru. The terms quarter wavelength and half wavelength are defined below. Furthermore, in Figure 1 For the interference filter L00t:I), the layers 101.103.107 and 109. Each layer consists of 6 GaAs monolayers, so its thickness is 16.9599A. Ru. On the other hand, layers 102.104.106. and 108 each have nine layers of Ga. It consists of a monolayer of o, 5sAlo, and asAs, so the thickness is 25.439 It is 8A. In this example, GaAs and Gao, 5iAlo, 4 Assume that each monolayer of sAs has a thickness of 2.82665A.

In addition, in the filter 100 of FIG. 1, the thickness of the layer 105 is equal to It's doubled.

FIG. 4 schematically represents the transmittance of the electron wave superlattice interference filter 100 shown in FIG. It is something.

Gao, 5sAl. The kinetic energy of electrons input to the 4sAs layer 131 is The passing electron kinetic energy, determined by the method described below in connection, is 0. 139eV, and the passband has a full width at half maximum (FWHM) of O,QO3eV. , its passband is only 2.2% of the passing electron kinetic energy. moreover, The interference filter of the invention advantageously exhibits a maximum transmission nearly equal to 100%. It is important to pay attention to this.

Below, we will discuss the constants between quantum mechanical electron waves in semiconductor materials and electromagnetic light waves in dielectric materials. The method of the present invention using quantitative mapping will be explained. This mapping is explained below. existing optical design methods and existing optical designs to similar solid-state devices. According to the method of the present invention used for converting into a method and design for a device, The mapping is (1) the square root of the product of the effective mass of the electron and the kinetic energy of the electron, that is, Using the "electronic wave phase refractive index" n, (phase) which is proportional to [m''(E - Vll'') By mapping the optical phase effect that depends on the optical path difference, (2) kinetic energy and It is proportional to the square root of the ratio to the loaded mass, that is, [(E − V)/m’l”,,’ [electron wave amplitude refractive index Jn.

(amplitude) to map optical amplitude effects such as transmittance and reflectance.

This mapping shows that quantum mechanical electron waves in a semiconductor and electromagnetic light waves in a dielectric are mutually exclusive. This method takes advantage of the knowledge that be. The result is an electron and hole wave device that is comparable to existing optical device designs. It will be done. This mapping is derived by recognizing the following: ) The continuity of the quantum wave function across the potential energy boundary is Similar to the continuity of the tangential component of the electric field across the boundary, (b) the potential energy The conservation of the probability flow of electrons perpendicular to the energy boundary is the power perpendicular to the boundary between dielectrics. Similar to saving flows. The mapping of the method of the present invention is similar to that of existing optical design methods. When using an existing optical design, it can be rephrased as follows. ( 1) Electron wave vector k = [2m” (E - V) instead of optical wave vector ]””/fi: (2) In the equation for reflectance and transmittance at the boundary, Instead of the optical refractive index, the electron wave amplitude refractive index [(E − V)/+Il"l"' is used. Ru.

Importantly, the above mapping applies to the case of electron waves in semiconductor materials, In that case, it depends on the following two facts. That is, (1) electron has an energy that exceeds the potential barrier in the semiconductor material (electronic energy in Figure 2). (note energy E), (2) Electrons exhibit ballistic or collisionless motion in semiconductor materials. vinegar. Furthermore, it is necessary to note the following important factors regarding the mapping of the method of the invention: . In other words, (1) electromagnetic light waves have polarization, and the mapping related to the amplitude effect has dimensionality. However, the equations for transmittance and reflectance regarding electron waves include electron Since only the dimensionless ratio of the wave amplitude refractive index appears, there does not seem to be any contradiction, and (2 ) Phase effect and amplitude effect The refractive index of an electronic wave exhibits a "normal" dispersion, i.e. These refractive indices increase with decreasing wavelength or increasing energy. (3) The semiconductor material becomes (a) E pair, that is, energy pair. may have a non-parabolic band structure with respect to momentum, and (b) may have a band structure that changes depending on the fixed electron wave propagation direction, but these effects The result is that by using an energy-dependent anisotropic effective mass 1, the method of the invention can be incorporated. Thus, the allowable wavevector surface is Even if the surface is no longer spherical, the design method of the present invention described here However, if an energy-dependent anisotropic effective mass is used in the analysis, it is still applicable. Available for use.

In light of the above findings, we have proposed an electronic wave superlattice device according to the present invention, for example an interference filter. It was confirmed that the filter has characteristics common to thin film optical interference filters.

Therefore, we will discuss some of the main characteristics of these optical filters, and It is useful to also consider the characteristics of the electron wave superlattice interference filter of the present invention.

A simple form of narrowband optical interference filter is the Fapley-Perot filter.

This filter is a reflective mirror often called a "spacer" in the optical literature. It consists of half wavelength 1 sandwiched between. In the case of an all-dielectric Fabry-Perot filter, , a quarter wavelength wave with a high refractive index called H and a quarter wavelength wave with a low refractive index called L. Consists of long layers. The FWHM of this type of filter's band is the line between the eyebrows. This can be reduced by increasing the reflectance of the field. This is a high refractive index This can be achieved by increasing the ratio of nH to the lower refractive index nL . Furthermore, for a constant calendar number, there is a high-index NH on the outer boundary of the filter. In this case, the reflectance is higher. The central half-wavelength resonance 1 of the filter has a high refractive index It may be a material having a low refractive index nL or a material having a low refractive index nL. like this There are basically two types of all-dielectric Fapley-Perot interference filters. Sentences related to optics In the literature, these filters have H and L made of high and low refractive index materials, respectively. As a quarter-wave layer of refractive index material, and N as a pair of opposites shown in square brackets. Symbolically as repeating numbers [HL] NHH[LH]' and H[LH]' 4LL [HL]>NH.

All-dielectric interference filters have other important characteristics as described below. , these characteristics are well known to those skilled in the art, and are useful in the present invention. In the method, the solid-state electronic wave filter of the present invention similar to these is also applicable. Available for use.

(CHI) The maximum transmittance of the filter is 100%; (CH2) The maximum transmittance is (a) a wavelength such that the thickness of the spacer M is half a wavelength measured in the material; (b) Waves such that the thickness of the reflective layer is one quarter' wavelength measured in those materials. (CH3) FWHM and The finesse and finesse are adjusted by the number of surrounding quarter-wave layers, i.e. Each time one wavelength portion is added, the FWHM decreases and the finesse becomes thicker (CH 4) Transmittance characteristics are measured in the material surrounding the filter as a function of reciprocal wavelength. When plotted, the transmittance characteristics show symmetry with respect to the wavelength passed; (CH3) all Transmittance plotted as a function of reciprocal wavelength as the eyebrow thickness varies proportionally A simple transfer of properties occurs. (CH6) When the thickness of all layers increases by an odd number, the passband occurs at the first passing wavelength, and the FWHM becomes small; (CH7) to the filter As the angle of incidence of increases, the transmitted wavelength tunes to shorter wavelengths; (CH 3) Transmittance characteristics are relatively unaffected by changes in layer reflectance and thickness. normal dispersion of (CH9) narrows the FWHM; (CHIO) sidebands are They always occur on both sides of the overband, so the filter is only useful for a limited range. It is effective.

Here, the special technicalities well known to those skilled in the art that are used when referring to filters of this type are: Explain terminology.

(1) Below the passing wavelength, the closest passband beam to this is above the passing wavelength. The range to the passband peak closest to this is called the free spectral range, FSR. (2) Finesse is equal to FSR/FWHM; (3) Resolution is It is equal to the wavelength divided by half the FWHM.

By using the above mapping, we can determine the properties of a multiboundary semiconductor superlattice system. Wear. This is a chain widely used in the field of electromagnetism, as shown in Appendix I. Using the matrix method, the electron wave obtained by the mapping of the present invention to the phase amount of the optical wave number vector is Substituting the motion vector and the refractive index in the formula for reflectance and transmittance at the boundary This is done by substituting the electron wave amplitude refractive index obtained by the mapping of the present invention into It will be done.

The example below shows how to convert an optical design to a solid-state design. vinegar. In this example, r Phy edited by G, Hass and R, E, Thun. A book called sics of ThinFilmsJ (vol. 5) (Ac academic Press.

A, T listed in the chapter starting from page 47 of New York, 1969 edition) Helen's optical thin film design has been translated into an electronic wave filter design. child The optical design is an 11-layer structure that exhibits a FWHM band of 2.2% of the design passing wavelength. . In the notation of optical thin film design, optical filters are defined as: 1. OHL　HHLHL HL is expressed as L'H1,0.

・Here, 1.0 indicates air in the input region and output region, and H indicates high refractive index (in the medium denotes a layer with a thickness of a quarter wavelength (as measured), and L is the layer with a low refractive index (as measured in the medium) ) indicates the thickness of a quarter wavelength. Therefore, the notation HH means a half wave with a high refractive index. means a layer of long thickness. For this optical design, nH=4.0, nL=1.35, nL=1.83. For the case of a passing wavelength of 1.00 μm, the optical filter The physical thickness of each layer is listed in Table ■.

The design of an electronic wave superlattice filter corresponding to this is as follows based on the above mapping of the present invention. can be obtained as explained in . First of all, in the input region of m=o The electron passing kinetic energy (E-V) can be calculated using the desired electron wave passing Calculate 51゛ from the following equation in relation to the wavelength 18me, o: (E　-Vol　-h272motan,ofi+However, m0° is the human power range or is the effective mass of the electron for the region m=o, and the amount of counterfeit m [(E − Lower the magnification to convert V,/m, ``]'''' to n, , (a width) of the same layer. Calculate from the following equation: D"-no/[fE-Vol/ffl."]1/ 2f21. However, no is the refractive index in the manual range of optical design (m=o). Third For each mth layer, calculate the required value of (E-Vm) using the following equation: However, , is the refractive index in the optical design for the m-th region, and m, ° is is the effective mass of electrons for the mth region. Fourth, the mth 1 physical Using the electron wave phase refractive index whose thickness is given by the following formula, the thickness of the quarter wavelength layer is z Count by 1: dra-h/(25/2[u,”(E-1/,l 11/21(’n) Next, Follow these steps to convert optical thin film interference filter design to electronic wave interference filter design. Convert to meter.

The motion obtained when the passing wavelength of this solid state filter is 100A The energy (E-■) and thickness values are shown in Table 11. However, in this example For simplicity, the effective mass of an electron is considered to be the mass of free electrons in all layers. Eggplant.

Is the calculated thickness too small for appropriate and practical fabrication of semiconductor superlattices? , or if the thickness is less than the monolayer thickness, all layers are multiplied by an odd number of thicknesses. can be reduced. This allows us to discuss the case of optical interference filters earlier. Exactly as explained above, if the FWHM of the solid state filter is the same odd number As a result, this solid-state filter is a more narrowband filter. becomes. For example, for the interference filter design above, the thickness of the solid-state layer is 3 By doubling or increasing to 51g, the FWHM increases from 2.2% of the passband wavelength to of 0.73% and 0.44%, respectively. However, the fineness of the filter The ratio of stopband to passbandwidth also decreases when a larger thickness is chosen. becomes smaller.

Furthermore, just like the optical thin film interference filter, the electronic superlattice interference filter of the present invention simply rotate the filter mechanically to increase the angle of incidence from normal incidence. Continuously tuned to lower passing wavelengths and therefore higher energies can be done.

The method described above was used to design the filter 100 of the present invention shown in FIG. In particular, the FIG. 3 shows a similar optical interference filter corresponding to the solid state filter 100 of FIG. The design of the filter is shown. As mentioned above, the filter 100 is made of GaAs, Gao, It consists of a superlattice 100 formed of a layer of 1iAlo4sAs, and the superlattice 100 is , Gao, 5iA1o, surrounded by the input layer 131 and output layer 132 of 46As. It is rare. The effective mass used in the design of filter 100 is For the entangled one, m' (GaAs) = 0.067 m, and m responsible Gao, wsAla, 411AS) = 0.10435mo (however, mo is is the mass of a free electron). The conduction band edge energy used is v(GaAs)= 0.0000eV and V (Gao, 1sAlo4sAS) = 0.3479eV , and the monomolecular layer thickness of GaAs, Gao, and 5sA1o4sAs is It was also set to 2.82665A. As a result, for these materials, six layers of GaAs (d+=16.9599A) monolayer and 9 layers of Gao, 5sA104! As The monolayer of (di = 25, 4398A) is surrounded by Gao, gsAl. Electron wavelength of 101.652 A or 0.139 as measured in the GaAs layer It shows a close agreement with quarter wavelength 1 at an electron kinetic energy of 456 eV.

The structure of the invention described above is well known to those skilled in the art, e.g. by molecular beam epitaxy. It can be assembled using this technique. Furthermore, according to the present invention, for example, the above By appropriate selection of the material composition, such as the ratio of Ga to A1 in the example of The passing kinetic energy can be changed. Therefore, the specified passing kinetic energy The design procedure for obtaining energy is as follows. (1) Fl-8G, H type material system For , an integer representing a monolayer of material is measured in the material and passes through kinetic energy. - Find a value of X that is equal to 1/4 wavelength at (E-V); ( 2) Select values of X that are as far apart as possible so that the reflectance at each boundary is maximized. For example, in the example above, this would be choosing x=0 and 0.45. (3) Then, as in optical systems, materials are layered in alternating layers. By sandwiching a half-wavelength layer between quarter-wavelength layers in the form of can form an elementary passband filter (FWHM and finesse are is adjusted by the number of quarter-wave layers in the envelope. In other words, the 1/4 wavelength part (FWHM decreases and finesse increases with each addition). For those skilled in the art, the book The superlattice device of the invention can be used, for example, as shown in FIG. (consisting of a superlattice formed from layers consisting only of different materials), as well as various material combinations. It will be obvious that it is possible to construct the layer with the following structure. 3 types or more The design with the upper layer of material is a solid-steel equivalent of a specialized optical thin-film device. may be required due to the need to manufacture smart devices.

When designing a solid-state device in accordance with the present invention, a specific filter design An appropriate material composition is required to obtain the required quarter-wavelength and half-wavelength layers. The selection can be made by trial and error. However, in each particular case this A preferred implementation of the method of the invention provides a systematic method for selecting such compositions. The aspect will be explained. Here, the materials forming a continuous set of alloys of the form F, , G, H Think about the fee system. The usable composition range is e.g. from O to Xsaw is determined by the range of X. The reason why this occurs is that Ga+□AI++As Indirect energy from direct energy bandgap materials as seen in the case To band gap materials, 1. This is because a transition may occur due to smell.

In principle, there is no rule that says indirect bandgap material O1 must not be used. , a transition between a direct bandgap material and an indirect bandgap material or two indirect If the transition between bandgap materials requires a change in momentum, the indirect band Gap materials cannot be used. It occurs in virtually collisionless motion. This is to deal with the wave effect that causes

Here we consider a systematic method for selecting material compositions for solid-state materials. It is assumed that the electron wave filter is composed of the following three types of materials.

(1) The material 4 in the i=0 region surrounding the superlattice has a composition X0. (2) Two-material supercase The material in the i=1 region, which is the child's high refractive index region, has a composition x 1; (3) 2 materials ``The material of the i=2 region, which is the low refractive index region of the superlattice, has a composition x2. i=1 territory The monolayer thicknesses of the region and i=2 region are rl and r2, respectively. these three The potential energy of an electron in the region of is given by the following formula: VL-daLEo-^XLi-0,1, 2+5 However, delEe is the energy change of the conduction band edge, and A is a constant It is. Furthermore, the effective mass of electrons in these three types of regions is calculated by the following equation: It is also well known in the art that m − (B ap cx llmo iwo, 1.2 (6) where B and C are constants and ma is free It is the mass of an electron. The electron interlocking energy in the i-th region is (EV + )=b"72m+"laml".F, total electron energy passed by filter - represents the passing kinetic energy measured in each region. - is given by: Ep-V4-h2/2+n white 1affflp) 11-0.1.2 +7 madashi , (lame) l is the passing wavelength measured in the i-th region. Super case The total pass coupled energy of the filter, measured in the material surrounding the filter, is Ep- V. This is the passing kinetic energy specified by the user and This is the starting point for this M1 procedure. Using the above equation, find the passing wavelength. Solving, we get the following equation: ,2 (lawpl, -h/12m(1[-^C×, ÷(CEp-^Blx1+B Ep] 11/2i-0 mountain 2(8) The thickness of the superlattice layer is d for i・1.2. represent. These thicknesses must be integral multiples of the monolayer thickness ri. Sara , these thicknesses are equal to the odd number of quarter wavelengths measured in these regions. Must be g. These constraints can be expressed by the following equation.

'i-Plrl-f2ql-11(lalllpH/4 i-1,2(9 ma However, pl is an integer representing the number of monolayers in the i-th region, and ql is a positive number. integers 1, 2, 3. ,,.

, is. Using the above two equations, the following quadratic equation for the composition can get: ゛゛　This equation can be solved by the well-known method of solving quadratic equations. super lattice dry In order to design a filter, the solution for X in the range from 0 to Xsaw must be There must be at least two. The minimum value of Xl within this range is X+, In other words, the composition is a high refractive index material. The value of p for which this X is obtained is p+, that is, is the number of monolayers of the above-mentioned type 1 material used to make the quarter-wave layer. Ru. Similarly, the maximum value of xl in this range is Xz, that is, the set of low refractive index materials. Becomes. The value of p that yields x2 is 1) i, that is, making a quarter-wavelength fake. The number of monolayers of the above-mentioned type 2 material used in

vo to give the widest range of solutions.

■□, is set. Next, for the specified passing kinetic energy (E-V,) Then, determine the value of E.

Furthermore, in order to make the overall thickness of the filter as small as possible, q is initially Set to unit. Next, until all positive real roots within the range from 0 to X, □ are obtained, Then, the above quadratic method equation is converted to p+ = 1.2, 3. ,,.

Evaluate repeatedly. Only one or no positive root is obtained. If not, the above steps must be repeated from the beginning with the parameters changed to λ. do not have. The amount that can be changed here is the composition xo of the surrounding material that changes the integer Qi, E. , and the crystallographic growth direction that changes rl.

Once you have determined X+, p+, Xit and p2 as above, the other parameters of the filter parameter can be calculated. Potential energy ■1, effective mass m, ・, the magnitude of the electron wave vector is 5, and the electron wave amplitude refractive index no (amplitude) is 1, Calculations can be made for various regions according to the above equation.

For example, consider the following example in the case of the material system Ga (-x A l x A s). Let's take a look. This is because all compositions are lattice matched for these alloys. This is a convenient material system. For growth along the [100] direction, a single molecule The thickness of the layer is r=r+=rx=2, 282665A.

This material is a direct gap semiconductor when X is less than or equal to 0.45, so that If so, it means that the composition is within the usable composition range. Furthermore, Ga+-JI In the case of llAs, A=0.77314eV, B=0.067, C=0.083 be.

- To give an example: Design of Ga1-xAlxAs superlattice interference filter with dynamic energy of 0.20eV To do so, the following calculations are performed: First, let XO=X□g=Q, 45 . Therefore, Vo = '0.347913, and E, -Vc = 0.20eV. Therefore, E,=0.547913eV.

q As l=1, p+ until all positive real roots in the range O-0,45 are found. =1.2,3. , , calculate the composition X. In this example, two real roots , i.e., X,=Q, 3984 and P corresponding to p, ;6.

There is x, =0.2063, which corresponds to =7. The smaller value of X is expressed as X+ , the larger value is expressed as X2. Gaa, 79Alo, 1/4 of z+As The thickness of the wavelength layer is d + = p, r = 19, 7866 A, and the thickness is 7 layers. Consists of molecular layers. Gao, s. The thickness of quarter wavelength 1 of A1゜4゜As is d 2=p2r=16.9599A and consists of 6 monolayers. into three areas The effective mass of the electron at is mo"=o, 10435m0.m,"=0.0841 26mo　, m=”=0.10100O7. The hyperkinetic energy is E, -V. =0.2000eV, E, -V, =0.388 4eV, Ep-Va=0, 2399eV. Surrounding i= The electron wave amplitude refractive index normalized to the O region is no (amplitude), = 1.000000 ．． n, (amplitude), = 1.552027, ne (amplitude) * = 1.118372 becomes. [HL] 13-layer Fapley-Bello interference filter in the form of “HH[LH]” In the case of the router, these calculated material properties yield a passband of Q, 20 eV. A filter with a FWHM of 15.4 meV is obtained.

By repeating the above steps, the most potential in ballistic transistors Energy range considered to be useful from 0.14eV to 0°20e Design a Ga+-JlxAs superlattice filter for passing kinetic energy up to V did. The positive real roots for thicknesses of 6 to 10 monolayers are shown in Table m. The real root is O must be within the range from 0.45 to 0.45.

At the lower end of this energy range of 0.14 eV, there are essentially four real roots. Ru. There are also two real roots at the 0.20 eV upper end of this energy range.

There is considerable flexibility in the design of semiconductor superlattice interference filters. For example, a quarter of Other odd multiples of wavelength include q+=2.3, 4. , , can be used , and by changing the surrounding material, the vo can be changed, and other crystallographic growth The direction can be used to change r.

It will be apparent to those skilled in the art that other embodiments can be constructed without departing from the teachings of the present invention. It is clear that it is possible to do so. For example, similar to electronic wave devices, It is also within the scope of the present invention to provide a wide variety of hole wave devices. Sara Within the scope of the invention, potential barriers similar to electromagnetic light wave devices are used. A wide variety of electron wave or hole wave devices using higher electron wave or hole wave propagation It is also possible to manufacture. Furthermore, according to the method of the present invention, such devices using existing methods for determining optical design and using the mapping of the present invention described above. It can be manufactured by using More specifically, such devices Examples of filters include low-pass filters, high-pass filters, and notch filters (narrowband and wideband filters). ), bandpass filters (narrowband and wideband), impedance transformers (antireflection, vice), and highly reflective surfaces (mirrors). Furthermore, these filters Putterworth characteristics (maximum flatness), Chebyshev characteristics well known in the field, It is also possible to have elliptic function properties or other properties. Furthermore, the present invention Devices include conventional devices such as electroluminescent devices, photodetectors, and ballistic transistors. It can also be used as a source of monoenergetic electrons for all types of devices.

Additionally, the device of the invention is useful for electronic spectroscopy, electronic lithography, and crystal folding. Used as an aid to control, shape, and filter free-space electron beams for analysis It is also possible to

“In addition to the above, according to the present invention, for example, in order to obtain total internal reflection by the above method, Electrons or holes are injected into the semiconductor layer sandwiched between the designed superlattices. These electrons or holes can be guided by entering the electrons or holes.

Regarding terminology, if the electron energy is higher than the potential barrier, then It is natural that it refers to energy higher than the conduction band, as shown in Figure 2. This will be obvious to businesses. Furthermore, when referring to holes, valence electrons It will also be clear to those skilled in the art that it refers to energies lower than the band.

Bias device Figure 5 shows a voltage biased electronic superlattice interference filter/emitter fabricated according to the present invention. 2 schematically shows an energy level diagram of the material composition of the cutter 100; FIG. This filter/effect The transmitter 100 consists of M layers 2001 to 2oo11. These layers 2001 ~200. are bulk semiconductor layers 1100+ and 1100. Surrounded by Nyota, Fi Router/emitter 100i: is a voltage source (not shown) between M2ooI and 2ooM. ), a predetermined bias potential V□/q is applied. . q is the charge of the electron.

As shown in FIG. 5, the filter/emitter 1o○ contains 250 electrons from the layer 1100+. is injected. Furthermore, the passing (band) energy E, (band energy E, is the layer 200I~200. a narrow spectrum centered at Only electrons with band kinetic energy (KE) are filter/emitter 100 and is released into 1111002.

Furthermore, these electrons have an output kinetic energy (KE) greater than the input kinetic energy (KE), fi. force kinetic energy (KE). , is released into the layer 1100□.

According to the invention, (KE). Since ut > (KE), ,, the present invention According to the filter/emitter 100, (KE) has an incident kinetic energy of In. An emitter function for electrons can be obtained. For easy understanding, the input kinetic energy The emission of electrons from the device of the present invention with kinetic energies higher than bias potential Vb, -/q caused by its application. The filter characteristics are affected by the energy ■ and □, and the This is achieved by In other words, the filter/emitter 100 A pot of material that has incident kinetic energy and forms layers 200 to 20011 of the superlattice. Narrowband transmission filter for electrons with total energy greater than the height of the mechanical barrier It provides a data/emitter.

As described in detail below, layer 200. ~200° thick, these layers 200. Composition of materials forming ~200 layers and potential barrier of layers 2001-200Il According to the method of the present invention, the height of the quantum mechanical electron wave in the semiconductor and the dielectric The method of the present invention determines this mapping using the mapping between the electromagnetic light waves in Using existing optical filter design techniques, thin-film optical designs designed using similar Application to the design of solid-state quantum mechanical electron wave filter/emitter devices Therefore, it is convenient.

The basic structure of the filter/emitter 100 of the present invention is based on the above-mentioned “unbiased device”. The structure of an unbiased superlattice electron interference filter as described in the section on It is the same as the construction. Specifically, for example, the above-mentioned narrowband optical interference A lattice consisting of successive layers of odd and even times of a quarter wavelength of electrons, such as a ruther. Electron wave interference filters can be viewed by analogy with optical filters.

Furthermore, as shown in FIG. 5, the j-th barrier or quantum well has a thickness denoted by d, , at zero bias, it has potential energy indicated by VJ. Here, the main To make it easier to understand the operation of this implementation, we will make the following assumptions (these (the assumption of N200 . does not limit the invention): (a) the surrounding N200. and 20 0 is the potential energy of the same zero bias■. selected to have; (b) The layers 2oO3~200M are alternately coated with a low potential such as V, energy and high potential energy such as ■2. selected. The result of choosing a particular value in this way, and n, is (EV) or The above equation (2) shows that the kinetic energy of the electron is proportional to the square root of the kinetic energy of the electron. By pinging, layers 200, ~200M have alternating high and low refractive indices. Ru.

In the embodiment of the invention shown in FIG. When applying the potential ■1.8, the layers 2001 to 200. ．． is reflector R 1 and in its reflector R1, each layer has a band energy E in each layer. , and the layer 200□,1 has a thickness of one-fourth the wavelength of the electron, measured at layer 2 with a thickness of one-half the wavelength of the electrons measured at band energy E2 in the layer; 0OR,. 2-200. is the electron wave measured with band energy E, in that layer. It has a thickness of one quarter of its length. Furthermore, the wavelength of electrons in a layer can be determined by the equation ( 1) and given by the following equation: Wavelength of electron = 2TTfi/[2♂(E-Vl11/2(sco)) Book shown in Figure 5 Embodiments of the invention are formed of a material system consisting of a continuous collection of alloys of the form F, G, H. It will be done. Generally, such a material system is expressed in the range O≦X≦Xmaw, for example. It has a limited usable composition range. This is G41-3 (AlxA The indirect energy from the direct energy gap, such as occurs when x = 0.45 in the s material system, This is because a transition to an energy gap can occur at X□.

Additionally, in the embodiment of the invention shown in FIG. It consists of a layer 11001 surrounding the. As is well known to those skilled in the art, Fl-oG, H The electron potential energy in a layer of material is given by: Vj-dalEc-AXj(N4) Here, delEc indicates the change in the energy of the conduction band edge in the material, and A is the constant It is a number. According to equation (N4), a given The potential energy range for a material system is given by O≦V≦V□. It will be done. The usable range of this potential energy is shown in Figure 5 for a given material system. MIlj50 representing zero electron potential energy and its V for a given material system 31. The potential energy between M[170 representing the electron potential energy of indicated by the spread of energy.

Furthermore, for the filter/emitter lOO to be physically realizable, the 200 ．． ~200. The thickness d of each layer in is the uniformity of the monoatomic layer thickness r of the material composition of each layer. It must be several times p.

The method of the present invention for obtaining embodiments of the present invention comprises, as a first step, A suitable unbiased superlattice as disclosed in the section on unbiased devices in It consists of selecting an electronic wave interference filter. The specific 9-eyebrow filter, i.e. An M=9 interference filter is disclosed. Here, we will explain the remaining steps of the method of the invention. Depends on how embodiments of the present invention are designed, the specific material composition and often How the thickness is determined will be explained. In particular, the filter of the invention disclosed herein / emitter 100 embodiment has an input kinetic energy of 0.10 eV (i.e. KE, n = 0.10eV), with a kinetic energy of 0.20eV (i.e., KEout = 0.20 eV). in this case ,Vゎ0. is equal to the difference between the output kinetic energy and the input kinetic energy, i.e. Output kinetic energy (KE). ut is V1. , +(KE), since it is equal to n ,■01. is equal to 0.10eV.

In optical thin film design notation, H is the fourth quarter of the high refractive index (as measured in the medium). , and L is the quarter-wavelength thickness of the low refractive index (as measured in the medium). Indicates the layer of sa. The symbol HH therefore means a half-wavelength thick layer of high refractive index. here An embodiment of the filter/emitter 100 of the present invention disclosed in (1) Layer l is a quarter wavelength layer (H) with a high refractive index; (2) Layer 2 is a quarter wavelength layer (L) with a low refractive index; (3) Layer 3 is a quarter wavelength layer with a high refractive index. (4) Layer 4 is a quarter wavelength layer (L) with a low refractive index. (5) Layer 5 is a high refractive index half-wave layer (HH); (6) Layer 6 is a low refractive index layer; (7) Layer 7 is a quarter wavelength layer (L) with a high refractive index. H); (8) Part 8 is a quarter wavelength layer (L) with a low refractive index; (9) Layer 9 is a high refractive index quarter wavelength layer (H).

Here, the design method of the present invention will be explained. First, how can a given electronic The thickness of the semiconductor layer, which is a multiple of a quarter of the wavelength of the electron at superenergy E, is Explain how it is determined. Thickness of the jth counterfeit of filter/emitter 100 In order to make 1/4 of the wavelength of the electron at the passing energy EI, the jth The input boundary to the layer, i.e., z　J−, in Fig. 5, and the output boundary of the j-th fake, i.e. In other words, the phase difference of the electron wave between Z in Fig. 5 is given by (2qJ 1)l ' must be an odd multiple of 11/2. This condition is expressed as: The potential energy VJ(Z) in the jth layer is given by: vj (zl = Vbla, (1-z/Ll + Vl ([16 + However, L is the total length of the superlattice 100, and q is a positive integer.

The energy passed through the filter/emitter 100 is calculated using the following formula: (]) However, (KE), , is the passing kinetic energy in the input layer 1100+.

The effective mass of the j-th layer is given by the following formula: ) However, B and C are constants, and mo is the mass of free electrons.

By using V, =Ax, we obtain the following "quarter wavelength" condition: Solve this equation (B9) to find the composition X of the j-th layer.

Determine.

For that, (1) i represents the number of monoatomic layers of the material forming the filter/emitter 100 of the present invention. (2) iJ is the j-th layer of the filter/emitter 100 of the present invention; shall indicate the number of the rightmost monolayer; (3) where i- indicates the total number of monoatomic layers in the filter/emitter 100 of the present invention; To be.

By using this notation: (1) The thickness of the j-th fake is given by dj=pJrJ. p J=L J I J −+ is the number of monoatomic layers in the j-th layer, and r is the number of monoatomic layers in the j-th layer. The material composition of the counterfeit X.

is the monoatomic layer thickness of; (2) The total thickness of the filter/emitter 100 of the present invention is determined by the sum of L=pyrJ. Given. This means that in a material system where r is the same value r for all layers, Equals lur; (3) the jth layer from one end along the filter/emitter 100 of the present invention; The distance to the beginning is ZJ-1=N'-'/iJ*' if r is the same for all given as; (4) the end of the jth layer from one end along the filter/emitter 100 of the present invention; If all r is the same, the distance to is given by

Described using this notation, the method of the invention consists of the following steps: (1) For the first layer, let 10 equal ○ and set j:0.

(2) For the next value of j, increase the value of 11 by 1, for example, set i,=1. 4 minutes For the one-wavelength layer, set ql equal to 1, and for the half-wavelength layer, set ql equal to π/2. Use π instead. Here, 2qJ-1) is the number of quarter wavelengths in the j-th layer. be.

(3) Using j, the previously set value of f J-+, and the setting value of iJ, set Xl and raise it. Solve equation (B9).

X, is the composition of the jth layer. For the high electronic refractive index layer, the value of X, which is closest to 0, is Choose a positive real value and for the low electronic refractive index layer, directly bandgap to indirect Real value of x, closest to and less than the value at which the transition to the bandgap occurs Select. For example, this value is 0.45 for the GaAlAs material system.

If you need to do this for more layers, return to step 2; If not, proceed to step 4.

(4) After the last step is completed, the total number of monolayers determined is the maximum number of monolayers. If the first predicted value i is larger or smaller than the first predicted value, then modify the first predicted value. , corresponding to the value of The optimal thickness of the last counterfeit that corresponds naturally matches the initial predicted value of iM used in the design. The above process is repeated until the total thickness of the filter/emitter 100 of the present invention is achieved. repeat.

We apply the inventive method to the design of the inventive filter/emitter 100. Ta. As mentioned above, in this embodiment,! 200. ~200Il and Surrounding layer 1100. and 11002 are made of Ga+-xAIJs material system. There is. This is because all compositions are lattice matched for these alloys, and [1 For growth along the 001 direction, the monolayer thickness is the same, i.e. r,=r= 2.282665A, it is a convenient material system. Also, if X is 0645 or more In the case below, the composition in this material system is a direct gap semiconductor, resulting in This indicates the usable composition range. Furthermore, in Ga+-JIJs, A= 0.77314e·V, B=0.067, and C=0.083. Furthermore, this fruit In an embodiment, Gao with the same composition, i.e. x0=0.45, 5sA1o , a@As surrounding layers 1100° and 11002 are used.

0.10 eV input, useful as emitter in fast ballistic transistors Ga+ -Jlx with kinetic energy and 0.20eV kinetic energy The design of a 9N implementation of As-based filter/emitter 100 is listed in Table B-A. . The effective masses used in the design of filter/emitter 100 are obtained from prior art documents. m'(GaAs) = 0.067. m　o　and m”(Gao , 1sA1o, 4aAS) = 0, 10435mo.

mo is the mass of free electrons. Also, the conduction band edge energy used is V (GaAs) = 0.0OOOeV and V (Gao, 5sA1o4sAS) = 0, 3479eV. In this design, the filter of the present invention /The total thickness of the emitter 100 is the thickness of 71 monoatomic fi, L=20.06 This corresponds to a length of 92 nm. This means that the filter/emitter 100 of the present invention A length of such an embodiment can be formed from a single piece of ballistic semiconductor material. This shows that the length is short enough to

In order to verify the above method of the present invention, the biased superlattice design shown in Table B-A was used. The electron flow transmittance was calculated. Does the bias result in a linear potential decrease? Then, the electron wave function in one of the filter/emitters 100 is the Airy function A It can be expressed as a linear combination of i(ρ) and residual Airy function Bi(ρ). just Then, the new variable ρ is defined in the j-th layer by: )ya, -(2+a",vbla,742t,)1/3(z+(z-vbla , -Vl)L/Vb1. , l+9101M layers shown in Figure 5 are laminated. In the case, the electron transmittance T, can be obtained by multiplying the following coefficient by the square of the electron amplitude transmittance: can get: [+E -Vo-vbiasl/”o11/2/[+’-v01/M”M+l ]1/2(alxlT, is for the biased superlattice design listed in Table B-A. This is shown in FIG. ■. , wa, = 0.1 oeV design bias, i.e. For curve 300, the device emits electrons at (KE), u, = 0.20 eV. Release into the force layer 1100□. The full width at half maximum (FWHM) is 30.7 meV, That is, it is 15.35% of the central energy.

Furthermore, the output kinetic energy from the filter/emitter 100 of the present invention is continuous. Tunable, i.e. the peak of the curve can be shifted, and curve filtering function Brings maintain shape. In particular, the peak of the output kinetic energy is Bias potential energy V applied to filter/emitter 100 It can be moved by changing I, im *. The HH resonance layer Since the peak of the output kinetic energy is at the center of the device as measured by the optical path length of the The movement of is equal to one-half the change in bias potential energy. Sunawa Therefore, when the bias potential energy changes by 50 meV, the output kinetic energy changes. The energy peak changes by 25 meV. Therefore, as shown in Figure 6, ±5 For a change in V Ili□ of 0 meV, curves 301 and 302 show that the output motion This shows that the peak energy changes by ±25 meV. deer However, note that different curves have different transmittances. This continuous tunable The characteristics of As a result, an assembly with output characteristics that deviate from closely matching a given design Vary the bias potential energy to synchronize the built embodiment It offers considerable advantages in that it can be

The all-dielectric optical interference filter described in the section on unbiased devices above. (Some important characteristics are that the solid-state electronic wave filter/emitter of the present invention , that is, (CH2); (CH3); (CH4); (CH3); (CH 6); (CH7); (CH3); (CH9): and (CHIO) also apply. circle.

It is possible to form the superlattice device of the present invention with layers having different material compositions. This will be clear to those skilled in the art. Further, for example, layer 200 of FIG. By providing electrodes at 200M and connecting a power supply between these electrodes, a bias voltage can be generated. It will be apparent to those skilled in the art how to apply the pressure Vblam to the filter/emitter of the present invention. It is very well known.

It is understood that other embodiments of the invention may be made without departing from the disclosure of this application. will be clear to those skilled in the art. For example, a wide variety of hole wave devices and It is also within the scope of the present invention to provide an electronic wave device. Furthermore, the scope of the present invention A wide variety of electron wave or hole wave filter/emitter devices can be used within the can be made. Furthermore, an electroluminescent device can be produced using such a device of the present invention. Thermal for all kinds of devices like chairs, photodetectors, and ballistic transistors Obtaining a narrowband semiconductor superlattice interference filter/emitter for use as an electron emitter Both are possible. Furthermore, such a device of the invention can be used in electronic spectrometers, electronic Controlling free-space electron beams for lithography and electron diffraction analysis of crystals, etc. It can also be used as an auxiliary means for , shaping and filtering.

To what extent are the embodiments of the present invention designed for both non-biased devices and biased devices? The degree of ballistic movement that occurs in the material forming the device depends on It is important to note that This means that the electrons in the material If the movement is approximately ballistic, the mode of operation of the device of the invention becomes more desirable. This means that the characteristics will be similar to those of the meter.

However, the device of the present invention can be used even if the electron movement is not very ballistic. Operates according to design characteristics in a degraded manner, i.e. performance degrades slowly It is also important to keep this in mind. Regardless, if you are a person skilled in the art, Current molecular beam epitaxy (MBE) and metal organic chemical vapor deposition (MOCVD) By using manufacturing technology, precise monolayer composition control and near-ballistic electron Multilayer superlattice structures can be grown with materials that ensure mobility. Sara However, doping of the semiconductor is not a critical factor for embodiments of the present invention; Doping within the active region of the device is avoided to avoid scattering within the device. It is preferable. This makes manufacturing easier by not doping. This is another advantage of the device of the present invention.

In principle, there is no prohibition against using indirect bandgap materials, but between a direct bandgap material and an indirect bandgap material, or between two indirect bandgap materials. When transitions between bandgap materials require a change in momentum, indirect bandgap It is not possible to use plastic. In the present invention, this results in almost collisionless motion. This is to deal with the effects of waves that occur in

Semiconductor superlattice interference filter in the case of no bias and in the case of bias type These filters/emitters, like their thin-film optical filter counterparts, require a design assembly. It is relatively unaffected by fluctuations from the closing price. Furthermore, (1) the semiconductor material is (a) A case with a non-parabolic band structure in terms of E vs k, i.e. energy vs momentum. (b) Band structure that changes depending on the specific direction of electron wave propagation in the material , that is, it may have an anisotropic energy band structure, but these effects If present, these effects can be calculated using the energy-dependent anisotropic effective mass m°. The results can be incorporated into the design method of the present invention. In this way, the allowable wave vector The surface ceases to be spherical in the presence of anisotropy, but energy-dependent anisotropy If the effective mass is used for analysis, all of the design methods of the present invention maintain effectiveness. do.

Finally, both non-biased and biased devices are suitable for fabrication of embodiments of the present invention. Suitable solid-state materials for use in construction include, inter alia, elements of groups Semiconductor materials such as binary, ternary and quaternary compositions of elements in groups ■~■ It is clear to a person skilled in the art that there are Dew.

l hill 1 FIG. 7 shows the energy level diagram and structure of the asymmetric quantum well slab waveguide 2100. show. In this application, this embodiment of the invention is described using the following notation: (a)1! ! 2200 is the substrate 1NS, layer 2201 is the film layer f, layer 2202 is referred to as a covering layer; (b) Direction perpendicular to the surfaces of the waveguide layers 2200 to 2202 indicated by arrow 2100 is denoted by XW; (c) layer 2201, i.e. at the bottom of the quantum well in the film Nf; The potential energy of the electron at is represented by ■; (d) Counterfeit 2200, i.e., substrate layer S, and M2202, i.e., substrate layer S; The potential energy barrier of electrons associated with the covering layer C are ■ and ■, respectively.

Expressed as: (e) N2200 to 2202 are materials in the form of F, -x, G, x H. These l112200-2202, i.e. substrate The material compositions of layer S, film layer f, and coating layer C are X, , X, and Xc, respectively. Represented by; (f) The guided mode propagation direction shown by arrow 2110 is represented by zw; (g) Let d be the thickness of the waveguide layer 2201, that is, the film layer f; (h) Express the incident angle of the two plane wave components forming the electronically guided wave as a zigzag angle of 0: (i) The size of the electron wave vector in any of layers 2200 to 2202 is To + = [2m”.

(EV,)]’/”/h. However, i if, substrate aS, all ie @2200, membrane Nf, ie M2201. and coating IWc, i.e. layer Corresponding to 22o2, i=s, f, c, m''l are the effective masses of the electrons, and Vl is The potential energy of the electron, E is the total kinetic energy.

More specifically, in the following description, layers 22oO to 2202 are Ga+-X The waveguide 2100 is made of materials selected from the AIIIAS material system. The electron potential energy between 12200 and 2202 of V, , V, , and ve are given by the following equation by the conduction band edge.

V, = AX + i = s, f, c, (S3) Furthermore, the layer 220 of the waveguide 2100 The effective mass of electrons in 0 to 22o2 is given by the following formula = m” = (B+ Cx+)m,, i=s, f, c (S4) where mo is the mass of the free electron Ru.

Before explaining specific embodiments of the slab waveguide of the present invention, the slab waveguide 2 of the present invention The operation of 100 will be explained qualitatively. This explanation will help you better understand the method of the invention It is thought that it will be possible to do so. Furthermore, the concept of critical angle used in the present invention To better understand how the slab waveguide 2100 operates by understanding Can be done.

In particular, the component of the electron wavevector parallel to the boundary between the two layers is Make it the same later, i.e. the transmitted electron wave along the border between the two eyebrows and by making the phase of the reflected electron wave the same as the phase of the incident electron wave, A law equivalent to Snell's law for electronic waves is applied to the slab waveguide 2100 of the present invention. It can be obtained in continuous form. According to this law, the starting point of total internal reflection is the incident This occurs when the angle, ie the zigzag angle defined above, is equal to the critical angle. The critical angle is ( a) The boundary between layers 2200 and 2201, i.e., i=s.

(b) The boundary between 112201 and 2202, i.e. between the membrane layer f and the coating NO For the critical angle of the boundary, i=c.

(c) E+t=(m”+V+-m’tVr)/(m”t-m”r)This is , which can be physically interpreted as follows. Boundary at an angle greater than θ If An incident electron wave will be totally reflected if the layer on the opposite side of the boundary is infinitely thick.

Therefore, in steady state, for example an infinitely thick substrate layer or an infinitely thick coating layer. An incident electron stream from the membrane layer that is incident on the membrane layer is reflected back into the membrane layer. interesting here In particular, the kinetic energy of the electron wave is less than or equal to O, i.e. (EV,)≦0. , total internal reflection can occur at all angles of incidence, including normal incidence. This means that the refractive index is What is the case of electromagnetic waves in which total internal reflection at normal incidence never occurs because it does not become O? different.

- In FIG. 8, the substrate layer 2200 of the slab waveguide 2100 is G a a, siA 42 o, +sA S, film @2201 is made of GaAs , the coating layer is G ao,t. Aβ. In the case of 3°As, the electron propagation constant Figure 2 is a graph plotting the number against the total electron energy. field of infinitely thick medium In this case, the electron propagation (E-V,)’/”/h, where i is the substrate layer S, i.e., the layer 2200, a membrane layer f, i.e., layer 2201, and a cover layer C1, i.e., layer 2202. i=s, f, and c, respectively. In Figure 8: (a) The song $1500 is about the substrate layer S, i.e. layer 2200, plotted. (b) The curve 501 is for the film layer f, that is, the layer 2201; is a curve plotted with; (c) The curve 502 is for the cover layer C1, i.e. @2202. It is something that The meaning of the information obtained from the curves 500 to 502 shown in FIG. 8 is as follows. That's right. For a given electron energy E, the propagation constant of the guided mode is As a result, the region 600 to the left of the curve 501 is attenuated. mode or non-physical mode, and the allowable guided modes in this electronic waveguide are , must be on the right side of curve 501. However, the permissible guidance mode is The gzag angle is the critical angle θ' at the boundary between the cover eyebrow and the membrane 1. , and the substrate layer-film layer interface The critical angle θ′ at the field.

The condition that it must be thicker must be met, i.e. The range of zigzag angle and total energy in FIG. 8 is θ>wax [θ ef + θ′, f] must be satisfied. As a result, the allowable induction mode is curve 5 It must be within the area 601 to the left of 00.

Next, we will qualitatively explain the cutoff phenomenon related to the slab waveguide 2100 of the present invention. Explain. Electron guided waves are cut by reducing the electron energy can be cut off, and the energy at which this cutoff occurs is set as the lower limit energy cap. This is called cut-off.

The zigzag angle of the plane wave component of the electron wave decreases as the energy decreases, and the zigzag angle When the angle θ=0, a lower energy cutoff occurs. Induction mode v number The propagation constant bv of the eye (where v is an integer starting from 0) is given by the following formula: bV-[2♂[(E-v)74211/2sine (56) As a result, the lower limit Energy cutoff occurs when =0. In this respect, the wave function is In the layer f, that is, the layer 2201, it is a sine function, and in the substrate layer S, that is, the layer 22 00 and the cover layer C1, that is, the layer 2202, decays exponentially. this In the sense, the lower energy cutoff is a hollow metal waveguide with finite conducting walls. It is similar to the cutoff of the dielectric mode of electromagnetic waves in a tube. hollow gold like this In waveguides, the plane wave component of the N magnetic dielectric liquid is incident at right angles to the boundary of the waveguide. Wherever it does, it repeats reflections back and forth.

As the electron energy of the guided mode increases, an upper energy cutoff occurs. Ru. The upper energy cutoff can occur in three forms: (1) Substrate Asymmetric dielectric conduction of electromagnetic waves when the refractive index of the layer is larger than the refractive index of the single cover layer Substrate mode in region 602 of FIG. 8 similar to the cutoff of a wave tube. (2) Electromagnetic asymmetric conductor with the same refractive index of the substrate layer and the cover layer; Radiation mode in region 603 of FIG. 8 similar to cutoff in electric waveguide Cutoff for: (3) When the refractive index of the cover layer is larger than the refractive index of the substrate layer, Cuts in sheathing modes similar to cutoffs in magnetically asymmetric dielectric waveguides OFF, the type of upper energy cutoff that occurs in the slab waveguide 2100 of the present invention. The class is determined by the intersection of the propagation constants s, b, and be. To be more specific, The intersection of b and Or occurs at energies E and t given by the following equation: Esf　’-sVg　-”[vf)/(+a”,-♂[)　(S))This E In terms of energy, the film layer f, i.e. layer 2201 and the substrate layer S, i.e. layer 220 When the electron wave phase refractive index of 0 is equal and reaches this energy, the slab waveguide of the present invention Tube 2100 no longer induces electron beams even at grazing incidence along its walls. will no longer be able to do so. This energy E, is the critical energy at the boundary between the substrate layer and the film layer. It is equivalent to the angle θ', ,=90°.゛Similarly, the intersection of bf and bc is given by the following formula Occurs with energy E at given by: “cf　-(ゞCvC-m”ev, l/(ゞ.-Ill”, (581 This energy In the Lugi, the membrane layer f, i.e., layer 2201, and the cover layer C1, i.e., layer 22. The electron wave phase refractive index of 02 is equal, this energy Ect is This is equivalent to the critical angle θ'c of the boundary between the two areas, =90°.

Similarly, the intersection of bl and bc is the energy E C1l given by the following equation. happen: At this energy, the substrate ms, layer 2200 and the cover layer C1 That is, the electron wave phase refractive indexes of 2202 are equal.

゛ In general, the type of upper energy cutoff that occurs depends on the material parameters. Depends on the data. In particular, in the embodiment shown in FIG. The plotted bl appears at lower energy than the be plotted as curve 502. Therefore, the upper energy cutoff is the cutoff for the substrate mode. Ru.

In short, when the electron energy increases in the electron induction mode, the zigzag angle θ also increases do. Furthermore, when the zigzag angle θ reaches the critical angle θ gt, the electronic guided wave S, i.e., rather than decaying exponentially in M2200, refraction into this layer Begin to. Then, the zigzag angle θ is the critical angle θ′1. From the energy equal to In the case of large electron energy, the electron wave propagates through the substrate layer S as well as through the film layer f.

This state is called substrate mode. Finally, the zigzag angle reaches the critical angle θ ef When the electron energy is further increased to 2202 and begins to refract into the substrate layer S. In this respect, electron waves are divided into three layers: In other words, it becomes a state where it propagates through all the layers from 2200 to 2202, and this state is called a radiation mode. It is called de. Furthermore, for a different set of material parameters, the electronic energy As the energy increases, it may occur that the electron waves are refracted into the cover layer C, This state is called coating mode.

The thickness and composition of a specific embodiment of the slab waveguide 2100 of the present invention are determined below. The methods used to do this are described below.

As mentioned earlier, the electronic wave vector of the electron waves in layers 2200-2202 and the electron wave phase refractive index ns (amplitude) are given by equations (1) and (2) above. Select. Furthermore, the wave function of the two-dimensional (X, , Z,) quantum well-induced electron wave is 2 It has a unidirectional sinusoidal dependence and is expressed by: UV(x,,z,l+υy(N)axp(jbvzwr(S101) Here, bv is a guided mode propagation constant.

Using equation (S10), Schrödinger's time position i-dynamic equation is as follows: It is expressed as: d2tJV(x, l/dx,'+12in''ih21[ E, -V(X,l]-b,,'IU,(x,l　-0(5111 Therefore, the induction mode For the case of the board, the wave function amplitude of the substrate layer, i.e. layer 2200, is given by: :'vs (Xw' -4^, sxp (gll ~ l (5121 membrane layer f, That is, the wave function amplitude of layer 2201 is given by the following equation: 'vf(xwl-^tlaxP (j') (Kv'+^, axp(-jkfx , ) (5131 Also, the wave function/amplitude of the cover layer C1, that is, the layer 2202 is Given by: 11vc(～)-^caXP(-9o[Xv-dll　(514 but below) Assume that the condition is satisfied: How the guided mode is distributed in the film layer f, that is, M2201 The equation is, U and (1/m') (du/dx) are the cover layer-film layer boundary and substrate layer. - Determined by using the boundary condition of continuity across the membrane layer boundary. . The dispersion equation of the guided mode of this film layer f is expressed by the following equation: kfd-Hian [+q, /mllll/+kf/♂f1, where ■ is an integer mode number . Further, the guided electron wave is expressed as Mr.

energy cutoff As mentioned above, the lower limit energy cutoff for waveguide mode MV is The energy decreases and the waveguide propagation constant bv becomes 0, so that the mode no longer propagates. Occurs when not transported. Figure 8 shows that this is the lower limit cut of the slab waveguide 2100. 1 only for electron energies below the off-energy. In other words, if E<V, This shows that this occurs only when

The electron energy when the lower limit energy cutoff occurs is determined as E Le6. , the lower limit energy cutoff condition is by substituting by=0 into the dispersion equation (S16) Determined by As a result, the lower energy cutoff condition is Mv It is determined by solving the following transcendental equation with E L + and o corresponding to the mode. Ru.

−””　”　fCvc　-”LCo)/lll11c(ELco-vfth 1/ 2- vTT (s171 energy cutoff As explained in detail, the upper limit energy cutoff for the conductive electron wave Mv is The energy increases so that total internal reflection no longer occurs, for example at the substrate-antinode interface. It can occur when there is no Therefore, through the upper energy cutoff As the electron energy increases, the electron wave is refracted and enters the substrate. sand That is, as the electron energy increases through the upper energy cutoff, the fundamental The amplitude of the electronic wave function in the plate propagates from a state where it decays exponentially. , that is, it changes to a sine state. This is the lower cutoff of the slab waveguide 2100. Only for electron energy exceeding the f energy, that is, when E>V, can only occur in As a result, modes “leak” into the substrate. Ru. The electron energy when the upper limit energy cutoff occurs is defined as Euca. The conditions for [Saturday energy cut-off] are as follows: Substituting g, 20 into the dispersion formula (S16) determined by As a result, the upper cutoff for the substrate mode The conditions for the limited energy cutoff are as follows, including E uco, which corresponds to Mv mode. It is determined by solving the transcendental equation.

(2 [♂svs　-”evf　-’♂5-♂r　l　EUc.]l　d/4 mode Assuming that the energy composition and potential energy of 7, which is Then, as the waveguide thickness d increases, i.e. Mf or the thickness of layer 2201 As d increases, the guided mode Mv first propagates at energy E=Vs. Start. This energy is the cutoff for the lower energy cutoff. Cutoff for the highest possible value of energy and upper energy cutoff corresponds to the lowest possible value of energy. E=V, using the dispersion formula (S16) 4 times By substituting, the thickness d when the mode Mv starts propagating becomes d-(4/2+m', (V, -Vf)] l"(tan [+s), (VC-V , l/♂c(vs-Vf)] 2+vnl (obtained as SL91).

Also, the range of thicknesses that produce waveguides that support only modes that include the Vth mode is obtained by the following formula.

K1 * [K2 + VTT] < d < K1 meeting [K2 ÷ (v + IITTI ('201 Therefore, using equation (S20) as V=O, the lowest mode, In other words, the thickness range of the layer 2201 can be determined so that only Mo is guided. Wear. As can be easily understood from this, regarding electromagnetic asymmetric dielectric slab waveguides, There is a minimum mode required for any mode to propagate. .

In Figure 8, songs 4!2300-2302 are the mode dispersion curves The propagation constant b0 for the low electron waveguide mode M0 is determined by the slab waveguide 210 of the present invention. Plotted as a function of total electron energy for specific embodiments of It is. Specifically, the (a) layer 2200, that is, the substrate layer S, is composed of 'cao, *5Alo, +sAS, that is, XI = Q, 15; (b) The layer 2201, ie, the film layer f, is made of GaAs, ie, x, 20; (C) Layer 2202, that is, the cover layer C is Gao, t. A1゜. From As Become.

That is, X. =0.30. We set A=,0.7331eV. Using equation (S3), we find that the conduction band discontinuity is about 60% of the energy gap change. Assuming that Vm=0.115971eV for layer 2200 and Vm=0.115971eV for layer 2201, For that, V, = 0.000eV, and for layer 2202, ■.

=0.231942eV. In addition, B=O1067 and C Using equation (S4) as =0.083, for layer 2200 m0□= 0.07945m, m”t=0.067mo for layer 2201, and eyebrow For 2202, it was determined that m''e = 0.0919mo. Additionally, in this embodiment, we assume that the eyebrow growth is along the [100] direction. , so that each monoatomic layer of material of waveguide 2100 has 0.28267no+ It becomes thick.

We use equation (S19) to determine that the film layer f has a thickness d equal to six monoatomic layers. It was found that the fundamental mode M0 starts propagating when Furthermore, again Using equation (S19), the following mode M is expressed as follows at a thickness of 31 monoatomic layers: It was discovered that propagation begins. As a result, in this embodiment, the slug The waveguide 2100 is made of a GaAs layer 2201 having a thickness of N5 to 30 monoatomic layers. It acts as a single mode waveguide. Curves 2300 to 2302 shown in FIG. is the value of bo as a function of the total electron energy for various thicknesses d of the film layer f. This is the result of solving the dispersion equation (S16). Specifically, (a) curve 2300 is It has a thickness of 10 monoatomic layers of GaAs (i.e. d=2.82665 nm ) corresponds to the film 11f, that is, the layer 2201; (b) the curve 2301 corresponds to the GaAs (c) curve ! 2302 is the thickness of 30 monoatomic layers of GaAs, that is, d=8.47995 nm). As can be easily understood from FIG. As the thickness of 201 increases, the mode dispersion curves 2300-2302 shift toward the left upper limit. Move to. Furthermore, as the thickness d of the film layer f increases, the lower limit energy cut off is reduced, i.e. curve 2301 has lower energy than curve 2300. intersects the energy axis at -, the upper energy cutoff increases, i.e. Chi song! I! 2301 has a higher energy than the song afA 2300. It intersects with the song AIl 500. Even at a thickness of 1110 single atoms, the waveguide module for example Mo at energies above any potential barrier. , that is, E>V and E>VC, propagation is possible. This is a song At the point where the energy of line 2300 exceeds both vc and V, curve 5 This is understood by the fact that it intersects with 00.

The energy E l1co of the upper limit energy cutoff, that is, the equation (S1 8), and the intersection of the M0 curves 2300 to 2302 and the curve 500 in FIG. The cutoff for the substrate mode and the lower limit energy cutoff corresponding to the point Energy ELc6, determined from equation (S17), M0 curve 230 The cutoff at bv = 0, which corresponds to the intersection of 0 to 2302 and the energy axis, The energy difference ΔE between Listed below.

Jiang Kai plate! Elle Embodiments of the slab waveguide 2100 of the present invention that are symmetrical, i.e., Vc = V. In this case, the substrate is distributed! ! b, and the cover dispersion curve bc are the substrate dispersion curve b @: [2m, ″(EV,)l″″/h, and the cover dispersion curve be=[2 mc Responsibility E-VC)] ""/h match at a certain point. In this case, the lower limit The energy cutoff is the point at which the electron energy decreases and the propagation constant bV becomes 0. That is, it occurs again when BV20. This is due to the lower electron energy barrier. This can only occur when the wall energy is below, i.e. when E<V, = Ve. can. If b　v　　　O, then the zigzag angle θ is also equal to 0 and the waveguide flat The surface wave component is reflected back and forth and is directly incident on the boundary of the waveguide.

For symmetric waveguides, the upper energy cutoff means that the energy increases and the waveguide This occurs when the mode becomes a radiation mode. At the upper energy cutoff , total internal reflection occurs at both the substrate-film interface and the cover-film interface. Does not occur. Then, the electron wave is refracted to the substrate layer S, that is, the 1112200 and the 2202. Upper limit energy cutoff As the electron energy increases through the substrate layer S, i.e. layers 2200 and The amplitude of the electronic wave function in the layer C1 of the cover 22o2 changes from the attenuated state. Changes to transport state. This is true if the electron energy exceeds the barrier energy. That is, it can only occur if E>V, =Vc. g,: If gc=Q , the mode is substrate/layer S, i.e., layer 2200 and cover Nc, i.e., layer 220. It leaks into 2. This is the cutting of electromagnetic waveguide modes in symmetric dielectric slab waveguides. Similar to off. For this type of cutoff, the zigzag angle is the critical angle between the substrate and the film. and the critical angle between the cover and the membrane are both simultaneously equal.

Electron guided wave MV appears for the first time when electron energy E=V,=Vc. this As a result, the electron waveguide M is such that the thickness d of the film layer f, that is, the layer 2201, is d - m TT/+2m” (+vs-Vf)] 1/2 (S21) and start propagation.

Also, from equation (S20), mode ■ is the lowest order, that is, v=0. of The range of thicknesses in which to fabricate waveguides that support waveguides is given by the equation:

o < a </+■/(2m”, (vs-Vfll”/2 (szz) equation As can be understood from (S22), the symmetric electron slab waveguide has a minimum There is no minimum thickness required for a mode of the order to propagate, i.e. , t6ii symmetric slab in that any thickness supports the M0 mode. Similar to a waveguide. However, for very thin electron slab waveguides, the wavefunction The exponentially decaying tail extends far into the substrate and cover layers. Maybe.

FIG. 9 shows a GaAs film layer f having a thickness of 10 monoatomic layers, that is, M of layer 2201.

The wave function Uv for the mode is shown diagrammatically for various electron energies. It is. In the figure, a curve 401 shows (a) b=o cutoff electron energy y, <E Corresponds to X<Ve; (C) Corresponds to electron energy Ve<Ej<Eue. ; (d) corresponds to the electron energy E 4” E U c o; and (e) One curve 401 to 404 corresponding to electron energy E, >Euco falls within these ranges. The curve 405 represents the behavior of the waveguide mode at an electron energy of Show the wave function for energies above the to-off energy to determine the substrate mode. Represents behavior.

Obviously, those skilled in the art will appreciate that further modifications of the invention can be made without departing from the teachings of the invention. It is recognized that implementations may vary. For example, not only electronic slab waveguides It is also within the spirit of the invention to provide a hole slab waveguide.

Regarding technical terminology, when we talk about electron energy exceeding the potential barrier, It is clear to those skilled in the art that this refers to energy exceeding the conduction band shown in Figure 7. It should be. Furthermore, a similar mention of holes can be found at energies below the valence band. It should be obvious to those in the painting industry that this refers to

Additionally, methods for injecting electrons and/or holes into the membrane layers of a slab waveguide are within the skill of those skilled in the art. It is well known.

Still further, embodiments of the present invention provide a method in which the membrane layer is comprised of a substantially ballistic material, and the substrate layer and It will be clear to those skilled in the art that the cover layer and/or cover layer may be configured otherwise. It should be white. However, in such embodiments, the substrate layer and/or The doping of the cover layers is small, and the losses in these layers account for the doping. Therefore, it is advantageous to avoid overdoing it.

Furthermore, the thickness of the substrate layer and the thickness of the cover layer can be substantially any numerical value. It should be clear to those skilled in the art that this may be the case. However, in reality, this The thickness of these fakes is large enough to support the upper tail of the waveguiding index in the membrane layer. Normally such thickness will be much less than the thickness of the membrane layer. Generally, several monoatomic layers of the material constituting the substrate layer and/or cover layer are combined. It's probably just a little bit. In practice, the substrate layer is typically grown on or The thickness of the substrate layer is not relevant since it is significantly thicker than the film being deposited. thing In theory, these requirements mean that the waveguiding in the membrane layer 1 has a boundary on top of the cover 1, for example. Understood as establishing the necessary condition that the existence cannot be "sensed" be done.

Finally, h is the blank constant, and h is the blank constant divided by 2π. , suitable fastening materials used in constructing embodiments of the present invention include semiconductor materials. materials, such as (but not limited to) binary compounds of group m-V elements and group elements It will be obvious to those skilled in the art that there are split, three-component and four-component compositions. It is.

Refractive index Thick Al Input range 1.00・ Area 1 (formerly 4.00 625.000 Area 2 (Ll 1.35 1851. 852 area 3 (L’l 1.83 1366.120 output area 1.00・ Kinetic energy (eVl thickness fAl Input area 0.015037 Area l (old area 0.240592 6.25000 area 2 fLl o.02 7405 18.51852 Area 3 fL’l 0.050357 13.66 120 output range 0.015037 on a plate 0, +40 0.0015 0.2902 0.3923 0.4513 0. 48980.145 0.0278 0.2996 0.4004 0.458 8 0.49700.150 0.0502 0.3088 G, 4084 0 ．． 4663 0.50420.160 0.08g5 0.3272 0.42 44 0.4813 0.51870.165 0.1057 0.3363 0.4324 0.4888 0.52590.170 0.1219 0.3 453 0.4403 0.4963 0.53310.175 0.1373 0.3543 0.4482 0.5037 0.54030.180 0. 1520 0.3632 0.4561 0.5112 0.54750.18 5 0.1662 0.3721 0.4640 0.5186 0.5547 ・0.190 0.1800 0.3809 0.4718 0.5260 0.56190.195 0.1933 0.3897 0.4797 0.5 334 0.56900.200 0.2063 0.3984 0.4875 0.5408 0.5762 Table B-A I H0770, 22220, 17180, 08542L 7 16 9 0. 4151 0.32G9 0.10153 H’ 16 23 7 0.266 3 0.2059 0.08914 L 23 32 9 0.4493 0. 3473 0.10435 HH3244120,06390,0494ro, 0 7236 L 44 52 8 0.4364 0.3374 0.10327 H525860, 14420, 11150, 07908L 58 65 7 0.3748 0.2898 0.09! l119 H657160, 1951 0,15080,0832 and juni top Waveguide film thickness: GaAs1 d (nm) 2.8267 5.6533 8.4800d (monoatomic layer) 1 0 20 30 With "L" The following description describes the design of electromagnetic-optical wave devices using solid-state quantum mechanical electronics. We explain how to map waves into the design of devices.

The present invention uses the mapping of the present invention between the quantum mechanical electron wave and the electromagnetic optical wave described above. According to the description, the techniques used in electromagnetic methods, in particular the chain Directly applying matrix design methods to designing devices such as interference filters Therefore, a solid-state superlattice system consisting of many boundaries can be determined.

The method for doing this is described below using a solid superlattice structure consisting of 6M layers. Assume that This solid superlattice structure has an input medium, denoted as layer O, in its -plane. The other side is bounded by the output medium represented as layer M+1. are doing. In the M-1 layer of the superlattice, the electric current moving in the positive direction and incident on layer m In the 1M-1 layer where the amplitude of the child wave is defined as Ull, it moves in the negative direction and the layer m The amplitude of the electron wave reflected from is defined as Ur,a. These complex wave amplitudes are The corresponding amplitude U1.m incident on the boundary of layer M+1 and reflected from there. Ml l and U7. It can also be expressed as follows as w and l.

(Henny 1): However, to, is the amplitude transmittance of the interface between layer m-1 and layer m, and ra, r- are is the amplitude reflectance of the boundary surface, 1 (0, is the magnitude of the electron wave vector in layer n1 , d is the thickness of layer n, and @7 is the angle in the wavenumber bevel direction in layer m. In the case of a superlattice consisting of 0M layers, the total normalized transmitted electron wave in the output FjM+1 is The amplitude U t, Mll and the total normalized reflected electron wave amplitude U r, o in the human layer O are M For each of the layers and one of the output layers, all the sums of the Mll form of equation AI-1 are It is determined by chain multiplication. The results are as follows.

Then, solve this directly and calculate the electron wave amplitude transmittance Ut, v*1 and the electron wave amplitude reflection. The rate tJr,o can be calculated.

° Formulas in the form of formulas Ar-1 and AI-2 are useful in the design of thin optical coatings and filters. It has been widely used for many years. The type of device being processed and If there is a wide range of settings Δl for , notch filters (narrowband and wideband), impedance transformers (anti-reflection coating) ) and high reflectivity surfaces (dielectric mirrors).

Electronic superlattice interference filter Electron kinetic energy j-v (eV) FIG. 5 Output kinetic energy ρτノ. , / / eV) distance @ x, (nm) FIG.9 Transfer constant 7nm international search report ll11#+-+IIA teacher----Nizlδ891052

Claims (41)

[Claims] 1. An electronic wave semiconductor device including a superlattice structure, The superlattice structure also has energy exceeding the potential energy barrier of the superlattice. A device that supports electrons to move almost ballistically. 2. The superlattice structure has a kinetic energy that exceeds the potential energy barrier of the superlattice. The device according to claim 1, which exhibits kinetic energy selectivity for electrons having Vice. 3. The superlattice structure has a kinetic energy that exceeds the potential energy barrier of the superlattice. The method according to claim 1, which exhibits a predetermined transmittance and reflectance for electrons having - device. 4. The thickness and composition of each layer of superlattice-structured material with predetermined transmittance and reflectance are , exhibiting predetermined transmittance and reflectance for electromagnetic optical waves in the optical wave thin film device; Claim No. 1 determined by analogy with optical wave thin film devices made of dielectric materials. The device according to item 3. 5. A superlattice structure is a predetermined arrangement of many layers of semiconductor material; Each layer has a thickness approximately equal to a predetermined number of quarter wavelengths of electron waves in the material of that layer. 4. A device according to claim 3, comprising: 6. The superlattice structure is a plurality of layers of a first semiconductor material, each layer containing one quarter of the electron wave in the material; having a thickness approximately equal to a predetermined number of wavelengths; a number of layers of a second semiconductor material, each layer containing one quarter of the electron wave in the material; having a thickness approximately equal to a predetermined number of wavelengths; 6. The device according to claim 5, comprising a predetermined configuration. 7. Claims that the semiconductor material is a two-component, three-component, or four-component semiconductor composition The device according to item 5. 8. A superlattice structure is surrounded by a semiconductor material with a conduction band of a predetermined height. 6. The device according to claim 5. 9. An electronic wave semiconductor device comprising a semiconductor layer for waveguiding electrons, the device comprising: The semiconductor layer also has energy that exceeds the potential energy barrier of its superlattice. at least one that completely reflects the electrons. A device characterized in that a superlattice structure is also surrounded by one superlattice structure. 10. Electrons with energy exceeding the potential energy barrier of the superlattice An electronic wave semiconductor device containing a superlattice structure that supports near-ballistic movement for manufacturing, forming an epitaxial superlattice structure consisting of a predetermined number of layers of semiconductor material; A manufacturing method that uses electromagnetic optical wave design, the parameters of which are determined according to an optical design method. By using a vise design, the refractive index of the optical phase can be adjusted to the electron kinetic energy and electron effective quality. Mapping into the first solid refractive index for a phase quantity proportional to the square root of the product of the quantities, the light The refractive index of the optical amplitude is the square root of the electron kinetic energy divided by the electron effective mass. Optical data by mapping into a second solid refractive index for proportional amplitude quantities. Converting the device design parameters into the thickness and composition of each layer of the superlattice structure A method characterized by selecting the thickness and composition of each layer. 11. At least one thickness obtained by the transformation is an adjustment of the thickness of a monolayer of the material of the layer. The thickness is multiplied by an odd integer multiple of the quarter wavelength of the electron wave until it becomes approximately equal to the number. 11. The method according to claim 10, wherein the method is modified. 12. At least one composition is changed to obtain a direct bandgap material composition. 11. The method of claim 10 further comprising: 13. Is the superlattice structure of the first material A1-xBxC and the second material A1-yByC? Ranari, The height of the conduction band in the material is approximately directly proportional to x and y, i.e. V1 = Ax and V2=Ay; the effective electron mass in the material is approximately linearly related to x and y; That is, m*1=(B+Cxi)m0 and m*1=(B+CXi)m0, respectively. And; Let X and y be the following equation ▲Contains mathematical formulas, chemical formulas, tables, etc.▼ ▲Contains mathematical formulas, chemical formulas, tables, etc.▼ (However, Ep is the energy of the electron incident on the superlattice, and h is the blank constant. where m0 is the free electron mass and r1 and r2 are integers of the monolayer thickness. is the thickness of the superlattice layer that must be doubled, and q1 and q2 each start from 1. If it is a whole integer, then d1=P1r1 and d2=p2r2) from the square root of The method of claim 10 for determining. 14. A hole wave semiconductor device including a superlattice structure, The superlattice structure has energy exceeding the potential energy barrier of the superlattice. A device characterized in that it assists holes to move almost ballistically. 15. A predetermined behavior is achieved by applying a predetermined bias potential energy to the superlattice structure. The device of claim 1 further comprising means for filtering dynamically energetic electrons. vinegar. 16. A superlattice structure is a structure made up of many layers of semiconductor material in a predetermined manner. Each layer is approximately equal to a predetermined number of quarter wavelengths of electron waves in the material of that layer under bias. 16. The device of claim 15, including one having a new thickness. 17. The superlattice structure includes a first reflector, a spacer and a second anti-reflector: The first reflector comprises a predetermined number of layer pairs, each layer being under a bias to Including those having a thickness approximately equal to a predetermined number of quarter wavelengths of the subwave; , approximately equal to a predetermined number of quarter wavelengths of electron waves in the material constituting the layer under bias. comprising a layer having a large thickness; The second reflector comprises a predetermined number of layer pairs, each layer being biased to Claim No. 1 having a thickness approximately equal to a predetermined number of quarter wavelengths of subwaves 16. The device according to 16. 18. each pair of layers of the first reflector exhibits a different electron wave amplitude refractive index; Claim 17: Each pair of layers of the second reflector has a different electron wave amplitude refractive index. device. 19. As we proceed from the first end of the device to the second end of the device, the The layers in each pair of layers of the first reflector are configured such that the layer is followed by a layer of lower refractive index. is; As we proceed from the first end to the second end, lower refractive index layers are followed by higher refractive index layers. The layers in each pair of layers of the second reflector are configured such that; 19. The device of claim 18, wherein the spacer comprises a higher refractive index layer. 20. As we proceed from the first end of the device to the second end of the device, the lower refractive index The layers in each pair of layers of the first reflector are configured such that the layer is followed by a layer of higher refractive index. is; As we proceed from the first end to the second end, higher refractive index layers are followed by lower refractive index layers. The layers in each pair of layers of the second reflector are configured such that; the spacer comprises a lower refractive index layer; a first higher refractive index layer a second higher refractive index layer is placed in front of the projector and a second higher refractive index layer is placed after the second reflector. 19. The device of claim 18, wherein: 21. the superlattice structure consists of a first semiconductor material and a second semiconductor material, A superlattice structure is surrounded by a semiconductor material with a conduction band of a given height. The device according to claim 15. 22. The first and second semiconductor materials are group III to V elements or group II to VI elements. Claim 21, which is a component-based, ternary-component, or quaternary-component semiconductor composition device. 23. The layers of the superlattice structure alternately include layers with a high electronic refractive index and layers with a low electronic refractive index, Each layer has a thickness approximately equal to a predetermined number of quarter wavelengths of electron waves in the material of the layer under bias. Has a certain quality; The width of the quantum well barrier in the superlattice structure is predetermined in the direction of electron emission. 16. The device of claim 15 for filtering predetermined kinetic energy. 24. A predetermined behavior is achieved by applying a predetermined bias potential energy to the superlattice structure. 16. The device of claim 15 further comprising means for filtering dynamic energy electrons. chair. 25. Electrons with energy exceeding the potential energy barrier of the superlattice Biased electronic wave semiconductor containing a superlattice structure that supports near-ballistic movement Epitaphrase consisting of a predetermined number of layers of semiconductor material for producing filters/emitters The manufacturing method includes a step of forming a axial superlattice structure, and the parameters are based on optical design. Use the electromagnetic optical wave device design determined according to the law as an initial estimate According to the iterative method, the refractive index of the optical phase can be expressed as the electron kinetic energy and the electron effective quality. The light is mapped into the first solid refractive index for a phase quantity proportional to the square root of the product of the quantities. The refractive index of the optical amplitude is the square root of the electron kinetic energy divided by the electron effective mass. Optical data by mapping into a second solid refractive index for proportional amplitude quantities. Converting the device design parameters into the thickness and composition of each layer of the superlattice structure A method characterized in that determining the thickness and composition of each layer. 26. The iterative method reflects the wavelength of at least one layer of the superlattice structure under a bias. At each step of the process, we measure a bilayer that approximately corresponds to an integer monolayer thickness of the material in that layer. The thickness of the superlattice structure can be determined by determining the thickness of the layer under the ass. 26. The method of claim 25, further comprising: 27. The iterative method is such that the thickness of at least one layer is approximately an integer of the thickness of a monolayer of the material of the layer. Try changing the thickness in odd integer multiples of a quarter wavelength until they match. 27. The method of claim 26, comprising: 28. The iterative method changes the composition of the material of at least one layer to directly form the bandgap. 28. The method of claim 27, further comprising providing a cup material composition. 29. (a) the layer of superlattice structure is made of material of the form A1-xBxC; (b) The conduction band height of the j-th layer is approximately directly proportional to xj, i.e. Vj = Axj And; (c) The effective electron mass in the material is approximately linearly related to xj, i.e. m*j=(B +Cxj)m0; (d) the layer under bias has a slow energy (KE)in has a thickness corresponding to a multiple of a quarter wavelength of electrons entering the filter/emitter with The conditions for doing this are the following equation ▲Contains mathematical formulas, chemical formulas, tables, etc.▼ (where L is the thickness of the superlattice structure and Zj-1 is the thickness measured from the beginning of the superlattice structure. Zj is the distance from the start of the superlattice structure to the start of the jth layer. is the distance to the end of the eye layer, and Vbiss is the bias potential energy , qj is an integer, h is the blank constant divided by 2π, and m0 is the free electron mass and V0 is the conduction to the material adjacent to the entrance of the superlattice structure. 28. The method of claim 27, wherein the height of the band is 30. a semiconductor substrate layer; a semiconductor film layer adjacent to the semiconductor substrate layer; and a semiconductor film layer adjacent to the semiconductor film layer; An electronic waveguide comprising an abutting semiconductor cover layer, A portion of the semiconductor substrate layer adjacent to the semiconductor film layer, the semiconductor film layer, and the semiconductor cover The portion of the layer adjacent to the semiconductor film layer provides nearly ballistic electron transfer; The thickness of the body membrane layer and the composition of the semiconductor layer are predetermined to provide a potential well; As a result, the electron energy in the potential well or the semiconductor substrate layer and Electrons exceeding either or both of the potential energy barriers of the conductor cover layer An electronic waveguide characterized by the existence of an electronic waveguide mode for energy. 31. Height of electronic potential energy barrier to semiconductor cover layer and semiconductor 31. The waveguide of claim 30, wherein the semiconductor substrate layers of the device are substantially the same. tube. 32. The semiconductor cover layer and the semiconductor substrate layer of a semiconductor device are made of the same material. The waveguide according to item 31. 33. The semiconductor cover layer, the semiconductor film layer, and the semiconductor substrate layer are made of Ga1-xAlxAs. 31. The waveguide of claim 30, comprising a composition of the form. 34. 34. The waveguide according to claim 33, wherein the semiconductor film layer is made of GaAs. 35. At least the vth electronic waveguide mode (v is an integer) propagates, and d ▲Contains mathematical formulas, chemical formulas, tables, etc.▼ (However, m*c, m*f and m*s are semiconductor cover layer, semiconductor film, respectively. are the effective masses of electrons in the layer and the semiconductor substrate layer, and Vc, Vf and Vs are respectively electronic vocatives related to the semiconductor cover layer, semiconductor film layer and semiconductor substrate layer. is the height of the energy barrier, where h is the blank constant divided by 2π). The thickness d of the semiconductor film layer is determined to be approximately equal to or greater than that. The waveguide according to claim 30. 36. The waveguide only supports modes that do not exceed the vth mode, K1*[K2+ vΠ]<d<K1*[K2+(v+1)Π] (where K1=h/[2m*f( Vs-Vf)]1/2K2=tan-1[m*f(Vc-Vs)/m*c(Vs -Vf)]1/2. 37. At least the vth electronic waveguide mode (v is an integer) propagates, and d d=vhΠ/[2m*f(Vs-Vf)]1/2 (however, m*c, m*f and m*s are the electrons in the semiconductor cover layer, semiconductor film layer, and semiconductor substrate layer, respectively. Vc, Vf and Vs are the effective masses of the semiconductor cover layer and the semiconductor film, respectively. is the height of the electronic potential energy barrier associated with the layer and the semiconductor substrate layer; h is the blank constant divided by 2π), or The waveguide according to claim 31, wherein the thickness d of the semiconductor film layer is determined so as to exceed tube. 38. the waveguide supports only the lowest order mode, i.e. v=0; Claim 37, where 0<d<hΠ/[2m*f(Vs-Vf)]1/2. mounted waveguide. 39. Semiconductor material is a binary or ternary system of group II to VI elements or group III to V elements. 31. The device according to claim 30, which is a four-component semiconductor composition. 40. a semiconductor substrate layer; a semiconductor film layer adjacent to the semiconductor substrate layer; and a semiconductor film layer adjacent to the semiconductor film layer; An electronic waveguide comprising an abutting semiconductor cover layer, The semiconductor film layer provides nearly ballistic electron transfer; the thickness of the semiconductor film layer and the semiconductor layer provides a potential well with a predetermined composition such that the electrons in the potential well energy or each potency of the semiconductor substrate layer and semiconductor cover layer. electron conductivity for electron energies exceeding either or both of the physical energy barriers. An electronic waveguide characterized by the existence of a wave tube mode. 41. Semiconductor substrate layer: a semiconductor film layer adjacent to a semiconductor substrate layer; and a semiconductor film layer adjacent to a semiconductor film layer; A hole waveguide including a semiconductor cover layer in contact with the hole waveguide, the hole waveguide comprising: A portion of the semiconductor substrate layer adjacent to the semiconductor film layer, the semiconductor film layer, and the semiconductor cover The portion of the layer adjacent to the semiconductor film layer provides approximately ballistic hole movement; The thickness of the body membrane layer and the composition of the semiconductor layer are predetermined to provide a potential well; As a result, the hole energy in the potential well or the semiconductor substrate layer and half Holes exceeding either or both of the potential energy barriers of the conductor cover layer A hole waveguide characterized by the existence of a hole waveguide mode for energy.