RU2249877C2 - Device for producing photoelectronic emission into vacuum - Google Patents

Abstract

FIELD: electrovacuum electronic engineering; semiconductor photoemissive devices operating in visible and close-to-ultraviolet region.

SUBSTANCE: proposed device that can be used to develop new class of effective photoemissive instruments around wide-range semiconductors has semiconductor material layer based on nitrides of third-group elements of electron conductivity with concentration of minimum 5·10 ¹⁶ cm ^-3 , forbidden gap width Eg, electron affinity χ, and electron work function φ on working surface of mentioned semiconductor material turned into vacuum and activated by alkali metal or its oxide meeting relationship E _g > χ, eV; χ > φ, eV.

EFFECT: enlarged spectral range and enhanced reproducing ability of spectral characteristics.

19 cl, 9 dwg

Description

The invention relates to the field of electronic vacuum electronic equipment, namely to photoemission semiconductor devices operating in the visible and near ultraviolet region. Such devices can be used to create a new class of efficient photoelectronic devices (photoemitters, photocathodes) based on wide-gap semiconductors.

Such photoelectronic devices are designed to register weak signals, as well as electron-optical conversion of signals used in optics, nuclear physics, automation, medicine, television and other fields. For practical use, these devices should provide a high quantum yield in combination with a well reproducible spectral characteristic of photoelectron emission.

A device for producing photoelectron emission in the visible and near ultraviolet spectral region, consisting of p-type semiconductor material based on antimony films with an alkali metal coating (see A. Sommer, "Photoemission materials", Moscow, Energia. ", - 1973, pp. 64-103).

The known device provides a high quantum yield η (electron per incident photon), equal to 0.3-0.35 at maximum, close to theoretically obtained estimates. The main disadvantage of this device is the irreproducibility of spectral characteristics, the long-wavelength edge of which can shift uncontrollably in energy by 1 eV to 2 eV, which is associated with the difficulty in controlling the production modes of the semiconductor materials themselves (see A.G. Berkovsky, V.A. Gavanin , I.N. Zaidel. - "Vacuum Photoelectronic Devices", - Moscow, Publishing House "Energy", - 1976, p. 32).

A photocathode is known for an electron tube whose emission surface is made of p-type semiconductor material having an acceptor impurity concentration of at least 10 ¹⁸ cm ^-3 and selected from the group consisting of semiconductor materials A ^III B ^V or their compounds having a forbidden width bands 1.1-1.6 eV. The emission surface of the photocathode is activated by an alkali metal. Material A ^{III is} selected from the group: B, Al, Ga, In. Material B ^{V is} selected from the group: N, P, As, Sb (see US Patent No. 3387161, IPC H 01 J 039/00, published June 4, 1968).

The known device has well reproducible semiconductor properties while providing a high quantum yield in the visible region of the spectrum. The disadvantage of this device is a sharp decrease in the quantum yield in the spectral wavelength range of 620-800 nm and in the near ultraviolet region of 300-400 nm.

A known cathode made of cesium-activated GaAs material (see JJScheer and J. Van Laar. - "GaAs-Cs: A new type of photoemitter". - Sol. State Commun. - V.3, 1965, p. 189 -193). For the known GaAs-Cs photocathode, the quantum yield η in the spectral range 400-600 nm is η = 0.4, and at 800 nm it is η = 0.08.

A disadvantage of the known device is a sharp decrease in the quantum yield in the wavelength range of 620-800 nm. In addition, it is known that cesium coatings on the surface of a GaAs semiconductor are unstable and quickly deteriorate due to the high mobility of cesium atoms on the surface of this material.

The closest set of essential features to the claimed solution is a device for producing photoelectronic emission in a vacuum (see European application No. 1024513, IPC H 01 J 1/34, published 02.08.2000), selected as a prototype. The known device consists of a p-type conductivity layer of a wide-gap semiconductor based on III-nitrides, in particular GaN, activated by an alkali metal or its oxide.

The known prototype device allows to obtain reproducible spectral characteristics and high quantum yield, but in a narrow spectral range of 200-350 nm. A high quantum yield was achieved due to the use of the negative electron affinity effect and p-type conductivity layers with different doping levels from impurity concentration 10 ¹⁶ cm ^-3 near the layer surface to impurity concentration 5 · ^{18 18} cm ^-3 deep in the semiconductor layer. The introduction of a lightly doped layer near the surface allows one to increase the diffusion length of charge carriers and reduce the number of defects associated with a high concentration of the dopant. The main disadvantage of the known device is the very low quantum yield η << 0.001 for wavelengths greater than 350 nm.

The present invention was to provide a device for producing photoelectron emission in vacuum with a high quantum yield in a wider spectral range and with reproducible spectral characteristics.

This object is achieved in that the device for producing photoelectron emission in vacuum includes a layer of semiconductor material based on nitrides of elements of the third group of n-type conductivity with a carrier concentration of at least 5 · 10 ¹⁶ cm ^-3 , with a band gap E _g , with affinity to the electron χ and the electron work function φ on the vacuum surface of the working surface of the said semiconductor material layer activated by an alkali metal or its oxide, satisfying the relations:

E _g > χ, eV;

χ> φ, eV.

The semiconductor material layer can be grown on a substrate of dielectric or semiconductor material, the refractive index n of which is less than the refractive index n _{p of the} semiconductor material layer.

The substrate can be made of sapphire, gallium arsenide, silicon.

The working surface of the semiconductor material layer can be activated by cesium or cesium oxide, in particular in the form of their monolayer or half-layer.

The working surface of the semiconductor material layer can be activated by barium or barium oxide, as well as potassium or potassium oxide, in particular in the form of their monolayer or half-layer.

The semiconductor material layer may be made of gallium nitride of n-type conductivity, hexagonal modification. For such layers, a typical structural feature is a mosaic (columnar) structure with characteristic domain sizes of 200-800 nm (Fernando A. Ponce - "Structural defects and materials performance of the III-V nitrides in Group III-nitride semiconductor compounds physics and applications". - Ed. By Bernard Gil, Oxford - Clarendon Press, 1998). The tilt and rotation of domains, the relaxation features of this complex system of defects play an important role in the formation of the electrical and optical properties of nitrides (NMShmidt, MNBesyulkin, MSDunaevsky, AGKolmakov, AVSakharov, ASUsicov, EEZavarin. - "Mosaicity and electrical and optical properties of group III-Nitrides ". - Journal of Physics: Condensed Matter, V 14, No. 48, 13025-13030, 2002). It is known that the structural features of nitrides can be characterized by studying the surface roughness using atomic force microscopy (AFM). The roughness reflects the disorientation and degree of relaxation of the domains. Recent studies have shown that structural features can be more fully characterized by processing AFM data by multifractal analysis using a parameter such as the degree of order of the mosaic structure. For epitaxial layers with well-relaxed domains with small tilt and turn angles, this parameter has values of no less than -0.33, which corresponds to a surface roughness of not more than 2 nm.

The semiconductor material layer can be doped with silicon or carbon with an impurity concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 .

The semiconductor material layer can be made of n-type AIGaN solid solution and doped with silicon or carbon with an impurity concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 .

As a semiconductor material in the inventive device, material III-nitrides is used - a semiconductor having an n-type conductivity (doped with donor impurities with a concentration higher than 10 ¹⁷ cm ^-3 ) and characterized by the conditions: Е _g > χ and χ> φ.

The condition χ> φ is achieved by activating the surface of the semiconductor by applying a monolayer or half-layer coating of alkali metal on it. The condition χ> φ is necessary for the existence of a layer of degenerate electron gas near the surface in a thin region 10–20 nm in size in the near-surface region of the semiconductor. The condition Е _g > χ makes it possible to efficiently excite photoelectron emission from the conduction band of the semiconductor. Photoelectron emission is excited by irradiation with light from the side of the emitting surface at an angle to it. The angle between the normal to the surface and the exciting light can be from 0 ° to the Brewster angle.

One of the main distinguishing features of the claimed device is the choice as a semiconductor material of a wide-gap semiconductor of n-type conductivity. It is generally believed that the use of n-type semiconductors as the photocathode material is inefficient due to the presence of a negative charge on the surface, which prevents the escape of electrons from the semiconductor to vacuum. As a result, such photoelectronic emitters have a low quantum yield η <0.01. Surface activation by applying an alkali metal or its oxide to the surface of a wide-gap n-type semiconductor material leads to new conditions: Е _g > χ and χ> φ, which eliminates this drawback and ensures the existence of a high density degenerate electron gas in the surface region of the semiconductor material , and on the surface of a positive space charge, which increases the exit of electrons into the vacuum.

In addition, the use of n-type conductivity material provides an increase in the diffusion length of charge carriers in a simpler way than in the prototype. It is well known that the diffusion length in an n-type semiconductor material is always higher than in a p-type material. Therefore, in the proposed device there is no need to obtain several layers of material with different concentrations of impurities and charge carriers. Due to the substantially large diffusion lengths, the quantum yield in the long-wavelength region of the spectrum increases.

Thus, the task of obtaining a high quantum yield is achieved by a simpler constructive solution than in the prototype, while expanding the spectral range of sensitivity. In addition, the range of materials providing a high quantum yield is expanding.

Let us explain this position. It is known that the quantum yield of photoelectron emission η is determined by:

η = B ₁ × B ₂ × B ₃ ;

where: B ₁ - the probability of excitation of the filled electronic states of the semiconductor with light to an energy level with an energy higher than or equal to the work function;

B ₂ is the probability of reaching surface electrons;

In ₃ - the probability of overcoming the surface barrier by electrons.

The probability B _{1 is} proportional to the density ρ (hν), cm ^-3 V ^-1 , of occupied electronic states participating in photoelectron emission, and to the matrix element M ₁ , cm ³ , of the excitation of occupied electronic states of a semiconductor participating in photoelectron emission:

B ₁ = C ₁ × _ρ (hν) × M ₁ ,

where: C ₁ - constant, eV.

For the prototype device, photoelectron emission occurs when electrons are excited from the valence band, where the edge of the band is E _V , eV, and where the density ρ (hν) of occupied electronic states at the edge of the band is zero ρ (hν) = 0 at E = E _V , and then slowly increases according to the law:

ρ (hν) = С ₁ × _ν (Е-Е _v ),

where: C ₁ - constant, cm ^-3 eV ^3/2 ;

E is energy, eV.

In the inventive device when photoelectron emission from the surface region of the semiconductor, where there are occupied electronic states of a degenerate electron gas in the conduction band, the density ρ (hν) of occupied electronic states is high:

ρ (hν) = A;

where A is a constant, cm ^-3 eV ^-1 .

Therefore, in the inventive device there is a sharp increase, compared with the prototype device, the density of occupied electronic states involved in photoelectron emission. The matrix element M ₁ in the inventive device corresponds to the prototype device.

The probability B _{2 is} proportional to the diffusion length of the charge carriers. It is well known that the diffusion length in an n-type semiconductor material is always higher than in a p-type material.

Thus, both conditions lead to an increase in the quantum yield for the claimed device.

The probability of B _{3 is} not worse than that of the prototype device, since in the inventive device it is provided with a positive charge on the surface, which creates a field that draws the electrons into the vacuum.

For the inventive device provides n-type conductivity with a carrier concentration of not lower than 5 · 10 ¹⁶ cm ^-3 , by doping a layer of semiconductor material with donor impurities. It is known that epitaxial III nitride layers with an impurity concentration lower than 5 · 10 ¹⁶ cm ^-3 have depleted space charge regions near the surface, the length of which in the depth of the layer is about 1 μm, which is caused by intrinsic self-compensating defects. Depleted areas interfere with electron emission. The introduction of a dopant, for example silicon, provides a carrier concentration of no lower than 5 · 10 ¹⁶ cm ^-3 and allows the suppression of the formation of these regions.

The inventive device for producing photoelectron emission in vacuum is illustrated by drawings, where:

figure 1 schematically shows a side view in section of one embodiment of the inventive device for obtaining photoelectron emission in a vacuum;

figure 2 schematically shows a top view of the device depicted in figure 1, in a section along aa;

figure 3 schematically shows another embodiment of the inventive device for producing photoelectron emission in a vacuum;

in Fig.4 schematically shows a top view of the device depicted in Fig.3, in section along BB;

figure 5 shows a diagram of the measurement of photoelectron emission in vacuum, obtained using the inventive device,

Fig.6 shows a band diagram of an n-type semiconductor, the surface of which is activated by an alkali metal or its oxide;

Fig. 7 shows the spectral characteristics of the quantum yield for the prototype device (a) and the claimed device, including an n-type GaN layer, the surface of which is activated by cesium (b);

on Fig shows the spectral characteristics of the quantum yield for the device of the prototype (a), the inventive device, including an n-type GaN layer, the surface of which is activated by cesium oxide (b), and the inventive device, including an Al layer of _0.15 Ga _{0, 85} N n-type conductivity, the surface of which is activated by cesium oxide (c).

figure 9 shows the orientation diagram of the electric vectors of s-polarized and p-polarized light waves exciting the inventive device.

The inventive device for producing photoelectron emission in vacuum includes a layer 1 of semiconductor material (see figure 1, figure 2) based on nitrides of elements of the third group of n-type conductivity, doped with a donor impurity with a concentration higher than 10 ¹⁷ cm ^-3 , with a band gap Е _g , with an affinity for the electron χ and the work function φ on the working surface 2 turned into vacuum, satisfying the relations:

E _g > χ, eV;

χ> φ, eV.

As such a semiconductor material, for example, n-type gallium nitride and n-type AlGaN solid solution can be taken. Layer 1 may also have a hexagonal modification with a mosaic structure with a degree of ordering of at least -0.33. Layer 1 can be doped, for example, with silicon or carbon with a concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 ,

The working surface 2 is activated by an alkali metal or its oxide, for example cesium or cesium oxide, barium or barium oxide, potassium or potassium oxide, preferably in the form of a monolayer or half-layer. Layer 1 can be deposited on a substrate 3 (see FIG. 3, FIG. 4) of a dielectric or semiconductor material, the refractive index n of which is different from the refractive index n _{p of} layer 1. The substrate 3 can be made, for example, of sapphire, arsenide gallium, silicon.

The zone diagram of the n-type semiconductor, the surface of which is activated by an alkali metal or its oxide, is shown in Fig.6, where E _c is the conduction band, E _V is the valence band, e _f is the Fermi level, E _cs is the position of the edge of the conduction band at surfaces, E _vac - vacuum level.

When using the inventive device in a photoelectronic emitter (see FIG. 1 and FIG. 3), the device is mounted on a metal plate 4 provided with a terminal 5 and placed in a vacuum chamber 6. Opposite the plate 4, an electron collector 7 is provided with a terminal 8. Conclusions 5 and 8 are passed through insulators 9.

The measurement of photoelectron emission in vacuum was carried out on the installation, schematically depicted in figure 5. In addition to the above elements 1–9, the installation includes a measuring circuit in the form of a series-connected electrometer 10, a direct current source 11 connected to pins 5 and 8, and a light source 12.

Example. The inventive device was manufactured on the basis of an epitaxial layer of gallium nitride GaN (0001) n-type conductivity doped with silicon (2 · 10 ¹⁷ cm ^-3 ), 4 μm thick, grown on a sapphire substrate (0001) by epitaxy from MOCVD organometallic compounds. Directly in vacuum, the sample was purified by thermal annealing at a temperature of 800 ° C. Activation by cesium was carried out by spraying atomically pure cesium on the GaN (0001) surface from a standard source (IPO-34-33 type evaporator). The concentration of cesium atoms on the surface ranged from 3 · 10 ¹⁴ cm ^-2 to 6 · 10 ¹⁴ cm ^-2 . Light from the visible and near ultraviolet spectral regions with an intensity of ~ 1 · 10 ^-3 W / cm ² was supplied to the GaN sample from the side of the surface activated by cesium. The wavelength of light was varied in the range from 250 nm to 950 nm. The sample was irradiated with p-polarized light at an angle γ = 45 ° to the normal to the surface, since for GaN the Brewster angle is 75 °. Irradiation of the sample can be carried out by unpolarized light, as well as light of s-polarization or p-polarization. Irradiation of the sample with p-polarized light provides an increase in the quantum yield of photoelectron emission by a factor of 2–3. The orientation pattern of the electric vectors of s- and p-polarized light waves that can excite photoelectron emission in the inventive device is shown in Fig. 9 (γ is an angle of 45 °, E _s is the electric vector of the s-polarized light wave, E _p is the electric vector p-polarized light wave, E $\genfrac{}{}{0ex}{}{\text{S}}{\text{II}}$ -tangential component of the electric vector of the s-polarized light wave, Е⊥ ^p - normal component of the electric vector of the p-polarized light wave, Е $\genfrac{}{}{0ex}{}{\text{p}}{\text{II}}$ - tangential component of the electric vector of the p-polarized light wave). The integrated photoemissive current was recorded in the sample – electron collector – electrometer circuit. The sample was grounded, a voltage of +150 V from a stabilized DC source was applied to the collector. The intensity of the irradiating light was measured using a calibrated photoelectronic multiplier, which made it possible to estimate the quantum yield with a relative error of 10%.

Figure 7 shows the spectral characteristic of the quantum yield for the inventive device, including the n-type GaN layer, activated by cesium at a surface concentration of cesium 4 · 10 ¹⁴ cm ^-2 (curve b). For comparison, the curve of the spectral characteristics of the quantum yield for the device prototype (curve a).

On Fig shows the spectral characteristics of the quantum yield for the device of the prototype (a) and for the inventive device, which includes an n-type GaN layer activated by cesium oxide (b), and the inventive device, including an Al layer of _0.15 Ga _0.85 N n-type conductivity activated by cesium oxide (c).

The above spectral characteristics show that the quantum yield of the claimed device and the quantum yield of the prototype in the spectral range of 250-350 nm coincide, and the claimed device has a high quantum yield in a wide spectral range up to 750 nm, and then it drops to 10 ^-4 at 900 nm, while the red border of photosensitivity lies near 950 nm.

Thus, the claimed device has a high quantum yield in a wide spectral range of 250-750 nm;

expands the range of materials for the creation of highly efficient photoelectronic emitters in the near ultraviolet and visible part of the spectrum;

simplifies and reduces the cost of technology for producing photoelectronic emitters, since the inventive device does not require the creation of multilayer structures;

the reproducibility of spectral characteristics increases in comparison with those analogs that provide the same spectral range, but have low reproducibility of characteristics.

Claims (19)

1. A device for producing photoelectronic emission in a vacuum, comprising a layer of semiconductor material based on nitrides of elements of the third group of n-type conductivity with a carrier concentration of at least 5 · 10 ¹⁶ cm ^-3 , with a band gap E _g , with electron affinity χ and the electron work function φ on the vacuum-facing working surface of the layer of said semiconductor material activated by an alkali metal or its oxide, satisfying the relations: Eg> χ, eV; χ> φ, eV; 2. The device according to claim 1, characterized in that said layer of semiconductor material is deposited on a substrate of dielectric or semiconductor material, the refractive index n of which is different from the refractive index n _{p of the} said layer of semiconductor material. 3. The device according to claim 2, characterized in that said refractive index n is less than said refractive index n _p . 4. The device according to claim 2, characterized in that said substrate is made of sapphire. 5. The device according to claim 2, characterized in that said substrate is made of gallium arsenide. 6. The device according to claim 2, characterized in that the said substrate is made of silicon. 7. The device according to claim 1, characterized in that said working surface of said layer of semiconductor material is activated by cesium or cesium oxide. 8. The device according to claim 7, characterized in that said working surface of said layer of semiconductor material is activated by a monolayer or half-layer of cesium or cesium oxide. 9. The device according to claim 1, characterized in that said working surface of said semiconductor material layer is activated by barium or barium oxide. 10. The device according to claim 1, characterized in that said working surface of said semiconductor material layer is activated by potassium or potassium oxide. 11. The device according to claim 1, characterized in that said layer is made of gallium nitride of n-type conductivity. 12. The device according to claim 11, characterized in that the said layer is made of a thickness of 2-4 microns. 13. The device according to claim 11, characterized in that said layer has a mosaic structure. 14. The device according to item 13, wherein the mosaic structure has a degree of ordering of at least 0.33. 15. The device according to claim 11, characterized in that the said layer is doped with silicon with a concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 . 16. The device according to claim 11, characterized in that the said layer is doped with carbon with a concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 . 17. The device according to claim 1, characterized in that said layer is made of n-type AlGaN solid solution of conductivity. 18. The device according to 17, characterized in that the said layer is doped with silicon with a concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 . 19. The device according to 17, characterized in that the said layer is doped with carbon with a concentration of 10 ¹⁷ -10 ¹⁸ cm ^-3 .

Images