EP0066926B1 - Semiconductor electron emitting device whose active layer has a doping gradient - Google Patents

Description

The invention relates to a semiconductor device having a surface capable of emitting electrons in response to electronic bombardment or to the impact of light radiation.

The invention relates to electronic or optoelectronic devices, such as photocathodes used in picture tubes and photomultipliers, which convert between photons and electrons, or dynodes used in photomultipliers, which operate by secondary electronic emission.

These various electron emitting devices, generally in a vacuum, are known from the prior art, both in their structure and in their manufacturing process, and the example of the United States patent, number 3,959,038, issued on May 25, 1976 and describing GaAs photocathodes, operating in transmission.

• le premier est l'excitation d'un électron, par exemple sous l'effet du rayonnement dans le cas des photocathodes, le second est la diffusion de cet électron au sein de la couche active, et le troisième est son émission dans le vide. Ces trois phénomènes physiques distincts obéissent à des lois différentes et sont favorisés par des caractéristiques différentes du matériau semiconducteur formant ladite couche active.

These devices very generally comprise an active layer, flush with the emitting surface, which is the seat of three distinct physical phenomena:

• the first is the excitation of an electron, for example under the effect of radiation in the case of photocathodes, the second is the diffusion of this electron within the active layer, and the third is its emission in a vacuum. These three distinct physical phenomena obey different laws and are favored by different characteristics of the semiconductor material forming said active layer.

The object of the invention is to obtain semiconductor devices whose efficiency is improved, that is to say whose excitation, diffusion and electronic emission are simultaneously improved.

The invention is based on the fact that these functions are separable.

The semiconductor device according to the present invention is remarkable in that it comprises a so-called active semiconductor layer, flush with the emitting surface, the doping of which increases when the distance to said emitting surface decreases.

Thus, the active layer of this device has characteristics which vary as a function of the distance from the emitting surface, so that in depth, the diffusion length is high due to low doping, which makes it possible to improve the scattering of excited electrons, and that on the surface, the probability of emission is high due to strong doping.

According to a practical embodiment of the invention, the semiconductor active layer consists of at least two different doping zones, the zone close to the emitting surface being relatively more doped.

In this case, the doping varies "in steps" instead of continuously varying, but the separation of the broadcasting and transmission functions is also ensured.

Finally, according to a particular embodiment of the invention, the active layer is made of a III-V compound, for example gallium arsenide, of type p conductivity, of a thickness less than 10 microns, and has a doping which varies radially between 10 ¹⁸ and 10 ¹⁹ atoms / cm ³ continuously or discontinuously.

The description which follows, with reference to the appended drawings, given without limitation, will make it possible to better understand how the invention is executed and to appreciate its scope.

Electron emitting devices generally fall into two types, depending on their mode of operation in transmission or reflection. The invention is intended to apply to devices of both types, but for simplicity the following description will rather refer to devices of the first type, without being able to draw any limitation therefrom.

Thus, FIG. 1 represents a photocathode with an inverted structure, operating in transmission, capable of emitting electrons as a result of the absorption of light radiation. This photon-electron transducer does indeed belong to the devices targeted by the present invention, and will serve as the basis for its description.

Such a photocathode is constituted by the sealing on a glass substrate 1 (or of corundum) of a complex semiconductor structure, by means of a sealing layer 2, for example in a glass of the short type such as that described in the French patent application, number 2,300,413, filed on February 4, 1975, in the name of the Applicant. The semiconductor structure consists of a semiconductor layer 3, called "active layer", generally made of gallium arsenide, of p conductivity type, and an additional layer 4, called "passivation layer", placed between the glass and the active layer, the function of which is to decrease the rate of recombination at the interface. In the case of an active layer 3 of GaAs (p), it is composed of gallium and aluminum arsenide, Ga _i -y Aly As, also of p conductivity type. This layer 4 also makes it possible to define the lower limit in wavelength of the pass band of the radiation; for example, for y = 0.50, it allows the passage of radiation whose wavelength is greater than 0.60 µm.

Finally, the active semiconductor layer 3 has on its outer face, intended to be subjected to vacuum, a state of negative apparent electronic affinity, obtained for example, by a conventional surface treatment, well known in the prior art, of covering with cesium and oxygen.

Such a glass-semiconductor composite material is not obtained immediately by simple bonding, but initially requires the growth of a double hetero-structure on a substrate, then the subsequent removal by pickling. chemical of the first heterostructure.

According to a now conventional method, the production of this structure therefore requires epitaxial growth, on a GaAs substrate 5, shown in dotted lines in FIG. 1 as destined to disappear, of a first layer 6 of Ga _1-x Al _x A _s , for which x typically equals 0.5, the so-called "chemical stop" layer (or blocking, because it makes it possible to stop the pickling process of the substrate, which would otherwise continue in the active layer 3), a second layer of GaAs called "active layer" 3, of p conductivity type, obtained for example by doping with zinc (Zn) or germanium (Ge), and finally a third layer 4 of Ga _1-y Al _y As, for which varies there between 0.25 and 1, according to the desired characteristics, the so-called passivation layer, and the functions of which have been specified previously.

The growth of these layers can be carried out by epitaxy in the liquid phase or in the vapor phase, for example according to the organometallic method. This structure is then bonded to a substrate 1 of glass (or corundum), which plays a role of mechanical support and optical window, this sealing being able to be carried out by means of a layer of glass 2, layer 4 known as passivation being the closest to the glass substrate 1, which explains in particular the mention of photocathode "with inverted structure".

After sealing, the substrate 5 and the chemical barrier layer 6 are removed by chemical pickling; an example of a bath used for the chemical attack on the GaAs substrate 1 is a solution of NH ₄ 0H (- 40%) at 5% by volume in H ₂ 0 ₂ (- 30%), a bath which has the advantage of '' a relatively high attack speed and excellent selectivity vis-à-vis the stop layer 6. This last layer 6 is then removed, for example by a commercial dilute hydrofluoric acid (HF) bath (40% V), a bath which practically does not attack the GaAs.

Finally, after all these operations, the active layer 3 is brought to an optimum thickness if necessary, for example by a light chemical pickling, then activated, in an ultra-vacuum frame, in order to obtain a photocathode, if that is its destination. .

The operation of these photocathodes is now well known to the physicist. The absorption of a photon in the semiconductor material causes the excitation of an electron which thus passes from the valence band to the conduction band, and which will diffuse within the material, after thermalization, during the time during which it remains mobile (lifetime τ), over an average distance Lα (diffusion length). The impurity atoms introduce additional energy levels into the forbidden band of the material, the location and density of which modify the lifetime of the charge carriers (traps) and therefore its diffusion length. In general, this diffusion length is a decreasing function of doping, which implies that in order to increase this length, the material should only be lightly doped.

And, when the electron thus excited reaches the GaAs / vacuum interface, it can be emitted in vacuum, provided that the material is placed in a state of apparent negative affinity. The probability of electronic emission depends on several factors, including the crystal orientation, the doping, etc ... and in particular, it is all the greater the higher the doping level.

These two criteria are therefore perfectly opposed, and the traditional solution has so far been to achieve a compromise in the value of doping. The materials produced to date are considered suitable, if the diffusion length reaches 4 μm, for a doping of 1.10 ¹⁹ atoms / cm ³ .

The invention aims to improve the apparent diffusion length by proposing a new structure for the active layer.

According to the present invention, the active layer 3 has a doping which increases when the distance to the emitting surface decreases.

In this way, an electron excited as a result of the absorption of radiation will diffuse over a greater length within the material as a result of a lower doping in depth, while its probability of emission in vacuum will be relatively high, because the doping in the vicinity of the emitting surface will be stronger.

According to an exemplary embodiment, carried out by the Applicant, the active layer of a photoemitter is obtained in the form of two different doping zones, in gallium arsenide of p conductivity type, doped with a low coefficient material diffusion such as germanium (Ge).

Zoned

• e = 4 µm
• N _A ―N _D ~ 1 to 2.10 ¹⁸ at / cm ³
• L _{D, 1} = 8 µm

Zone II

• e = 1 µm
• N _A ―N _D ~ 1 to 2.10 ¹⁹ at / cm ³
• L _{D, 2} = 4 µm

• L_D,app ≃ 7 µm

Such a structure, with a total thickness of 5 μm, has an apparent diffusion length for its operation as a photoemitter:

• L _{D, app} ≃ 7 µm

The continuous variation of doping can be obtained either by a variable dosage of impurities in the process of growth, for example in the epitaxy reactor in the vapor phase, or by diffusion thanks to the choice of an impurity do pante with greater diffusion coefficient, such as zinc (Zn).

This structure can also be obtained with any other semiconductor material, such as binary or pseudo-binary compounds III-V or II-VI, etc., the values of the compositions, dopings and thicknesses of layers then being adapted to each case. and easily calculable by the practitioner, without it being done for that purpose.

It is also understood that the invention is not limited to photocathodes, but also finds its application for the production of dynodes, generally in any semiconductor device emitting electrons.

FIG. 2 is a network of theoretical curves, giving an example of the variation of the sensitivity (on the abscissa, in μ A / lumen) in white light (2854 K), photocathodes with inverted structure, in accordance with the present invention, as a function the thickness of the active layer 3 of GaAs (on the ordinate, in μm), for different values of the apparent electron scattering length (parameter, in μm), and for which P represents the probability of photoelectron emission, and S the photoelectron recombination speed at the GaAs / GaAl As interface.

It is interesting to emphasize that the maximum sensitivity increases with the diffusion length in accordance with the experiment, which shows the perfect agreement between the experiment and the theory, but also that the optimal thickness of the active layer 3 of GaAs increases, which facilitates its realization and that the curves flatten, which makes the choice of this thickness less critical, a non-negligible side effect with respect to the development of these structures.

It is obvious to those skilled in the art that many variants can be envisaged without departing from the scope of the present invention as defined by the claims below appended.

Claims (9)

1. A semiconductor device having a surface capable of emitting electrons in response to an electron bombardment or to the inpact of luminous radiation, characterized in that it has a so- called active semiconductor layer (3) which adjoins the emissive surface and the doping of which increases as the distance to the said emissive surface decreases.

2. A semiconductor device as claimed in Claim 1, characterized in that the active layer (3) is formed by at least two discrete doping zones, the zone near the emissive surface being comparatively highly doped.

3. A semiconductor device as claimed in Claim 1 or 2, characterized in that the semiconductor material is chosen among the binary or pseudo- binary III-V, II-VI compounds.

4. A semiconductor device as claimed in Claim 3, characterized in that the semiconductor material which forms the active layer (3) is of gallium arsenide of the p-conductivity type.

5. A semiconductor device as claimed in Claim 4, characterized in that the doping of the active layer (3), going from the emissive surface varies between 10¹⁹ and 10¹⁸ acceptors/cm³.

6. A photocathode as claimed in any of the Claims 1 to 5.

7. A dynode as claimed in any of the Claims 1 to 5.

8. A display tube comprising at least a photocathode as claimed in Claim 6.

9. A photomultiplier comprising a semiconductor electron emitter as claimed in any of the Claims 6 and 7.

Images