Light-emitting thyristor and self-scanning light-emitting device

ABSTRACT

A light-emitting thyristor having an improved luminous efficiency is provided. According to the light-emitting thyristor, a p-type AlGaAs layer and an n-type AlGaAs layer are alternately stacked to form a pnpn structure on a GaAs buffer layer formed on a GaAs substrate, and Al composition of the AlGaAs layer just above the GaAs buffer layer is increased in steps or continuously.

INFORMATION

DETAILED DESCRIPTION OF THE INVENTION

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a schematic cross-sectional view of a conventional light-emitting thyristor having a buffer layer.

FIG. 2 is a schematic cross-sectional view of a conventional light-emitting thyristor using a GaAs layer as a topmost layer.

FIG. 3 is a graph showing a light absorption spectrum of an n-type GaAs layer.

FIG. 4 is a schematic cross-sectional view of a conventional light-emitting thyristor.

FIG. 5 is an equivalent circuit diagram of the thyristor shown in FIG. .

FIG. 6 is a schematic diagram showing a first embodiment of the present invention.

FIG. 7 is a schematic diagram showing a second embodiment of the present invention.

FIG. 8 is a circuit diagram of a characteristic estimating circuit for a light-emitting thyristor.

FIG. 9 is a graph showing a measured threshold current.

FIG. 10 is a graph showing a measured holding current.

FIG. 11 is a schematic diagram showing a third embodiment of the present invention.

FIGS. 12 and 13 are schematic diagrams showing a fourth embodiment of the present invention.

FIG. 14 is a schematic diagram showing a fifth embodiment of the present invention.

FIG. 15 is a graph showing a photoluminescence intensity of InGaP.

FIG. 16 is a graph showing a light absorption spectrum of In0.5Ga0.5P layer.

FIG. 17 is a circuit diagram of a light output measuring circuit.

FIG. 18 is a composition diagram of InGaAs.

FIG. 19 is a graph illustrating the relationship between a lattice constant and an energy gap of AlGaInP.

FIG. 20 is a graph showing current-light output characteristics.

FIG. 21 is a schematic diagram showing a seventh embodiment of the present invention.

FIG. 22 is an equivalent circuit diagram of a first fundamental structure of the self-scanning light-emitting device.

FIG. 23 is an equivalent circuit diagram of a second fundamental structure of the self-scanning light-emitting device.

FIG. 24 is an equivalent circuit diagram of a third fundamental structure of the self-scanning light-emitting device.

BEST MODE FOR CARRYING OUT THE INVENTION

The embodiments of the present invention will now be described with reference to drawings.

First Embodiment

Referring to FIG. 6, there is shown an embodiment of the present invention wherein the problems in the conventional thyristor shown in FIG. 1 may be resolved. In FIG. 6, AlxGa1−xAs layer is epitaxially grown on a GaAs substrate in such a manner that Al composition x is increased in steps from 0 (GaAs) to 0.35. Since the method of epitaxial growth is identical independent of the types of conductivity (n-type or p-type) of GaAs and AlGaAs, the embodiment will be illustrated without distinguishing the conductivity type.

AlGaAs layer is epitaxially grown on the GaAs buffer layer by varying the supply flow rate of Al in such a manner that Al composition is varied at 0, 0.1, 0.2, 0.3, 0.35, for example. That is, a GaAs layer having Al composition of 0, an AlGaAs layer - having Al composition of 0.1, an AlGaAs layer - having Al composition of 0.2, an AlGaAs layer - having Al composition of 0.3, and an AlGaAs layer - having Al composition of 0.35 are epitaxially grown successively.

Four AlGaAs layers -, -, - and - that Al compositions are increased in steps correspond to the layer in FIG. . In this case, the total film thickness from the GaAs buffer layer to AlGaAs layer - is determined by a containment efficiency of carriers.

The processing steps hereinafter is the same as the conventional steps in FIG. . Then, AlGaAs layers having Al composition of 0.35, respectively, are epitaxially grown hereinafter.

Second Embodiment

Referring to FIG. 7, there is shown a second embodiment of the present invention wherein the problems in the conventional thyristor shown in FIG. 1 is also resolved. In FIG. 7, AlxGa1−xAs layer is epitaxially grown on a GaAs buffer layer in such a manner that the composition s of Al is continuously increased from 0 to 0.35. Such modification of Al composition is implemented by varying continuously at least one of supply flow rates of Al and Ga.

In this manner, on the GaAs buffer layer just above the GaAs substrate , formed is an AlGaAs layer - having an Al composition varied from 0 to 0.35, and then an AlGaAs layer - having an Al composition of 0.35.

These two AlGaAs layers - and - are corresponding to the AlGaAs layer in FIG. . In this case, the total of film thickness from the GaAs buffer layer to AlGaAs layer - is determined by a containment efficiency of carriers.

The steps hereinafter is the same as the conventional steps in FIG. . Then, AlGaAs layers having Al composition of 0.35, respectively, are epitaxially grown hereinafter.

According to the present embodiments wherein an Al composition is continuously varied, the lattice defects such as dislocation due to lattice-mismatching at the interface between the GaAs buffer layer and the AlGaAs layer may be decreased, and the extreme deformation of an energy band at the interface may be softened. As a result, an adverse effect with respect to the device characteristic is reduced.

A threshold current, holding current, and light output of the light-emitting thyristors of the first and second embodiments are measured by the following method as illustrated in FIG. . An anode electrode , a cathode electrode , and gate electrode of a light-emitting thyristor are connected to a constant current source and a constant voltage source as shown in FIG. . In this characteristic estimating circuit, a cathode voltage Vk and a gate current Ig were measured by varying an output current Ik of the constant current source . A typical example of measurements is shown in a graph of FIG. 9. A maximum current just before the gate current Ig was reversed from up to down was considered as a threshold current. A cathode voltage Vk was also measured by varying the output current Ik which is equal to a cathode current of the thyristor . A typical Ik-Vk characteristic is shown in a graph of FIG. . The holding current was defined as a current wherein the cathode voltage was over a constant value (e.g., 0.2V). The light output in the case when the gate electrode was connected to the anode electrode through a resistor and the output current Ik was set to a suitable value (e.g., 13 mA), was measured by a photodiode.

Fifteen-twenty thyristors of the first or second embodiment were measured as to the threshold current, holding current, and light output. Compared with the conventional thyristor shown FIG. 1, the threshold current was decreased by about 20% in average, the holding current was decreased by about 15% in average, and the light output was increased by about 10% in average.

Third Embodiment

Referring to FIG. 11, there is shown a third embodiment of the present invention wherein the problems in the conventional thyristor shown in FIG. 1 is also resolved. In FIG. 11, a quantum well layer is formed on the GaAs buffer layer just above the GaAs substrate . On the layer , the AlGaAs layer , AlGaAs layer , . . . are epitaxially grown in the same manner as in FIG. . The quantum well layer serves as same as the AlGaAs layer wherein Al composition is increased in steps in the first embodiment or the AlGaAs layer wherein Al composition is continuously increased in the second embodiment. As a result, the lattice defects such as dislocation due to lattice-mismatching at the interface between the GaAS buffer layer and the AlGaAs layer may be decreased, and the extreme deformation of an energy band at the interface may be softened.

Alternatively, a quantum well layer may be inserted into the AlGaAs layer in place of providing the quantum well layer between the GaAs buffer layer and AlGaAs layer . Also, a strained superlattice structure may be used in place of the quantum well layer, resulting in the same effect as that in the quantum well layer.

Fourth Embodiment

In the AlGaAs layer wherein Al composition is varied in steps or continuously, there is a problem in that misfit dislocation caused by lattice-mismatching propagates through the AlGaAs layer and reaches the upper layer, thus adversely affecting the characteristic of the thyristor. Embodiments will now be described in which the propagation of such misfit dislocation is reduced or blocked.

In an embodiment shown in FIG. 12, a quantum layer or strained superlattice structure is inserted into the AlGaAs layer - in FIG. . The propagation of the misfit dislocation may be blocked by means of the quantum layer or strained superlattice structure .

In an embodiment shown in FIG. 13, a quantum layer or strained superlattice structure is inserted into the AlGaAs layer - in FIG. . The propagation of the misfit dislocation may be blocked by means of the quantum layer or strained superlattice structure .

Fifth Embodiment

FIG. 14 is a schematic cross sectional view of a light-emitting thyristor of the present invention, in which the problems in the conventional thyristor using a GaAs layer for a topmost layer as shown in FIG. 2 is resolved. The structure of the embodiment is substantially the same as that of the conventional thyristor shown in FIG. 2, except that the topmost GaAs layer is replaced by a layer consisting of InGaP which is lattice-matched with the GaAs substrate.

In1−xGaxP is lattice-matched with GaAs, when the composition x is about 0.5. When InGaP is grown by means of MOCVD, trimetylindium (TMI) is used for In component, trimetylgallium (TMG) for Ga component, and phosphine for P component. Since the growth conditions of InGaP depend upon the structure of a reactor, the setting of growth conditions is required so as to obtain the desired composition x, i.e., 0.5. A growth temperature is in a range of 600-700° C., when a low pressure growth method is used. Mole ratio (TMG/TMI) of TMG and TMI both of them being material for III Group components is determined, assuming that said mole ratio is proportional to a mixed crystal ratio(x/l−x). Selenium is used as a dopant for obtaining an n-type InGaP, hydrogen selenide being utilized for selenium component.

A sample in which a single layer of InGaP was grown on a GaAs substrate was prepared for estimating an optical characteristic. FIG. 15 shows a photoluminescence intensity of In0.5Ga0.5P layer measured at a room temperature. A central wave length of emitted light is about 660 nm. FIG. 16 shows a light absorption spectrum of said In0.5Ga0.5P layer in comparison with that of the GaAs layer shown in FIG. . The absorption edge wave length of In0.5Ga0.5P is about 650 nm (0.9 ev) and the absorption coefficient for the light of 780 nm wave length is lower than 10 cm−1, which is sufficiently smaller than the absorption coefficient 1.5×104 cm−1 of GaAs.

Said InGaP layer is used as the uppermost cathode layer when a light-emitting thyristor is fabricated. The growth method of the InGaP layer is the same way as described above, and the other processes are substantially the same as that has been disclosed as to the thyristor using a GaAs layer. Also, in order that a cathode electrode makes ohmic contact with the InGaP layer, AuGeNi is used for the material of the cathode electrode.

To measure the light output of the thyristor, a constant current source and resistors are connected to the thyristor as shown in FIG. . The gate electrode is connected to the anode electrode through the resistor , and the constant current source is connected between the anode electrode and the cathode electrode . The light output of the thyristor was measured under the constant cathode current (e.g., 10 mA) by means of a photodiode.

The resulting light output was increased by about 3% in average compared with the typical value of the conventional thyristor. This shows that the light absorption by the In0.5Ga0.5P layer is negligibly small.

In the case where In1−GaxAs1−yPy is used for the material of the uppermost layer, an absorption coefficient may be caused to be small by using the compositions x and y on a large absorption end energy side, respectively. For illustrating this, a composition diagram of In1−xGaxAs1−yPy is shown in FIG. . In the figure, a solid line designates a contour of energy gap Eg, and a dotted line a contour of lattice constant. According to this composition diagram, a line designating an absorption energy of 1.6 eV corresponds to a 780 nm wave length of emitted light, and a lattice constant of 5.65 Å corresponds to that of GaAs. Therefore, it is apparent from FIG. 18 that an absorption coefficient may be decreased by using a composition on a high energy side from the dot within compositions having a lattice constant equal to that of GaAs.

In the case where AlxGayIn1−x−y is used for the material of the uppermost layer, it is required to select each composition x or y so that AlxGayIn1−x−yP is lattice matched with GaAs. FIG. 19 is a graph for illustrating the relationship between a lattice constant and an energy gap of AlGaInP. In the figure, ordinate designates a lattice constant and obscissa an energy gap Eg. A shaded area shows a composition region in which AlxGayIn1−x−yP may be formed, and the composition indicated by a solid line within said composition region is lattice matched with GaAs. In this composition, the energy gap is sufficiently larger with respect to a wave length of 780 nm, then an absorption coefficient is considered to be sufficiently smaller compared with that of GaAs.

Sixth Embodiment

The embodiment of the present invention will now be described, in which the problems caused in the conventional thyristor shown in FIG. .

The light-emitting thyristor was fabricated, in which only each concentration of the p-type GaAs layer and the p-type AlGaAs layer in the structure of FIG. 4 was varied. A table 1 shows the kind of material, the film thickness, the type of impurity, and the concentration of impurity of each layer and the substrate.

The substrate is composed of GaAs, and the impurity therein is Zn. The buffer layer is composed of GaAs with 500 nm thickness, and the impurity therein is Zn. The anode layer is composed of Al0.3Ga0.7As with 500 nm thickness, and the impurity therein is zn. The n-type gate layer is composed of Al0.13Ga0.87As with 200 nm, and the impurity therein is Si. The p-type gate layer is composed of Al0.13Ga0.87As with 800 thickness, and the impurity therein is Zn. The cathode layer is composed of Al0.3Ga0.7As with 500 nm thickness, and the impurity therein is Si. The ohmic contact layer is composed of GaAs with 30 nm thickness, and the impurity therein is Si.

Four types of impurity concentration, i.e., Nos.- were prepared as shown in Table 1. Nos.- impurity concentration in each of the four layers , , , are the same. That is, Si impurity concentration of the layer is 1×1018/cm3, Zn impurity concentration of the layer is 1×1017/cm3, Si impurity concentration of the layer is 3×1018/cm3, and Si impurity concentration of the layer is 3×1018/cm3.

On the other hand, each Zn impurity concentration of the layers and in No. is 2×1017/cm3, each Zn impurity concentration of the layers and in No. is 2×1018/cm3, each Zn impurity concentration of the layers and in No. is 5×1017/cm3, and each Zn impurity concentration of the layers and in No. is 1×1018/cm3.

Apparent from the above, each impurity concentration of the layers and is not lower than that Si impurity concentration of the layer in Nos. and .

For the light-emitting thyristors each having impurity concentration of Nos.-, the current-light output characteristics were measured, respectively. The resulting current-light output characteristics are shown in FIG. . The light-emitting thyristor of No. wherein each Zn concentration of the layers and is sufficiently lower than Si concentration of the layer has the highest amount of emitted light. The thyristor of No. has the next higher amount of emitted light. On the other hand, the thyristors of No. and No. wherein each Zn concentration of the layers and is equal to or lower than Si concentration of the layer have the lower amount of emitted light. Therefore, it is noted that the luminous efficiency of the thyristor is not decreased, in which each Zn concentration of the layers and is set so as to be lower than Si concentration of the layer . This is because the impurity diffusion from the layers and to the layer is limited.

Seventh Embodiment

In the sixth embodiment described above, it is made apparent that the advantageous effect may be obtained by lowering each impurity concentration of the layers and than that of the layer . However, the lowering of each impurity concentration of the layers and increases each resistance of these layers, having an effect on the characteristic of the thyristor. In order to avoid this, the layer in the sixth embodiment is divided into two layers - and - as shown in FIG. . In this thyristor, the impurity concentration of each layer is set as shown in Table 2. The impurity concentration of the upper layer - of two divided layers is set to lower concentration, i.e., 2×1016/cm3, and that of the lower layer - is set to 2×1013/cm3. Each impurity concentration of another layers is the same as in the sixth embodiment.

An impurity concentration of each layer after growing was measured by a secondary ion mass spectroscopy, the result (measured concentration) of which is also shown in Table 2. According to the measuring result, it is noted that the impurity concentration of the upper layer - was 4×1017/cm3 larger than the set concentration (2×1016/cm3). This is because Zn was diffused from the lower layer - to the upper layer - during growing the upper layer.

The current-light output characteristic of the thyristor according to this embodiment was similar to that of No. case of the sixth embodiment. Therefore, it is noted that there is an effect when the impurity concentration of the part of the layer near the layer is low.

While a p-type substrate is used in the sixth and seventh embodiments, an n-type substrate may be used. In this case, the impurity concentration of the anode layer may be lower than that of an n-type gate layer. Also, while the embodiments are illustrated for an impurity Zn which may be easily diffused, the present invention may be applicable to another kind of impurities for the fifth and sixth layers.

Eighth Embodiment

Three fundamental structures of self-scanning light-emitting device to which the light-emitting thyristor of the present invention can be applied will now be described.

FIG. 22 shows an equivalent circuit diagram of a first fundamental structure of the self-scanning light-emitting device. According to the structure, light-emitting thyristors . . . T−2, T−1, T0, T+1, T+2 . . . are used as light-emitting elements, each of thyristors comprising gate electrodes . . . G−2, G−1, G0, G+1, G+2 . . . , respectively. Supply voltage VGK is applied to all of the gate electrodes via a load resistor RL, respectively. The neighboring gate electrodes are electrically connected to each other via a resistor RI to obtain interaction. Each of three transfer clock (φ1, φ2, φ3) lines is connected to the anode electrode of each light-emitting element at intervals of three elements (in a repeated manner).

The operation of this self-scanning light-emitting device will now be described. Assume that the transfer clock φ3 is at a high level, and the light-emitting thyristor T0 is turned on. At this time, the voltage of the gate electrode G0 is lowered to a level near zero volts due to the characteristic of the light-emitting thyristor. Assuming that the supply voltage VGK is 5 volts, the gate voltage of each light-emitting thyristor is determined by the resistor network consisting of the load resistors RL and the interactive resistors RI. The gate voltage of a thyristor near the light-emitting thyristor T0 is lowered most, and the gate voltage V(G) of each subsequent thyristor rises as it is remote from the thyristor T0. This can be expressed as follows:

(0)<(+1)=(−1)<(+2)=(−2)  (1)

The difference among these voltages can be set by properly selecting the values of the load resistor RL and the interactive resistor RI.

It is known that the turn-on voltage VON of the light-emitting thyristor is a voltage that is higher than the gate voltage V(G) by the diffusion potential Vdif of pn junction as shown in the following formula.

ON()+dif  (2)

Consequently, by setting the voltage applied to the anode to a level higher than this turn-on voltage VON, the light-emitting thyristor may be turned on.

In the state where the light-emitting thyristor T0 is turned on, the next transfer clock φ1 is raised to a high level. Although this transfer clock φ1 is applied to the light-emitting thyristors T+1, and T−2 simultaneously, only the light-emitting thyristor T+1 can be turned on by setting the high-level voltage VH of the transfer clock φ1 to the following range.

(−2)+difH(+1)+dif  (3)

By doing this, the light-emitting thyristors T0 and T+1 are turned on simultaneously. When the transfer clock φ3 is lowered to a low level, the light-emitting thyristors T0 is turned off, and this completes transferring ON state from the thyristor T0 to the thyristor T+1.

Based on the principle described above, the ON state of the light-emitting thyristor is sequentially transferred by setting the high-level voltage of the transfer clocks φ1, φ2 and φ3 in such a manner as to overlap sequentially and slightly with each other. In this way, the self-scanning light-emitting device according to the present invention is accomplished.

FIG. 23 shows an equivalent circuit diagram of a second fundamental structure of the self-scanning light-emitting device. This self-scanning light-emitting device uses a diode as means for electrically connecting the gate electrodes of light-emitting thyristors to each other. That is, the diodes . . . D−2, D−1, D0, D+1 . . . are used in place of the interactive resistors RI in FIG. . The number of transfer clock lines may be only two due to the unidirectional of diode characteristics, then each of two clock (φ1, φ2) lines is connected to the anode electrode of each light-emitting element at intervals of two elements.

The operation of this self-scanning light-emitting device will now be described. Assuming that as the transfer clock φ2 is raised to a high level, the light-emitting thyristor T0 is turned on. At this time, the voltage of the gate electrode G0 is reduced to a level near zero volts due to the characteristic of the thyristor. Assuming that the supply voltage VGK is 5 volts, the gate voltage of each light-emitting thyristor is determined by the network consisting of the load resistors RL and the diodes D. The gate voltage of an thyristor nearest to the light-emitting thyristor T0 drops most, and the gate voltages of those thyristors rise as they are further away from the light-emitting thyristor T0.

The voltage reducing effect works only in the rightward direction from the light-emitting thyristor T0 due to the unidirectionality and asymmetry of diode characteristics. That is, the gate electrode G+1 is set at a higher voltage with respect to the gate electrode G0 by a forward rise voltage Vdif of the diode, while the gate electrode G+2 is set at a higher voltage with respect to the gate electrode G+1 by a forward rise voltage Vdif of the diode. On the other hand, current does not flow in the diode D−1 on the left side of the light-emitting thyristor T0 because the diode D−1 is reverse-viased. As a result, the gate electrode G−1 is at the same potential as the supply voltage VGK.

Although the next transfer clock φ1 is applied to the nearest light-emitting thyristor T+1, T−1; T+3, T−3; and so on, the thyristor having the lowest turn-on voltage among them is T+1, whose turn-on voltage is approximately the gate voltage of G+1+Vdif, about twice as high as Vdif. The thyristor having the second lowest turn-on voltage is T+3, about four times as high as Vdif. The turn-on voltage of the thyristors T−1 and T−3 is about VGK+Vdif.

It follows from the above discussion that by setting the high-level voltage of the transfer clock φ1 to a level about twice to four times as high as Vdif, only the light-emitting thyristor T+1 can be turned-on to perform a transfer operation.

FIG. 24 shows an equivalent circuit diagram of a third fundamental structure of the self-scanning light-emitting device. According to the structure, a transfer portion and a light-emitting portion are separated. The circuit structure of the transfer portion is the same as that shown in FIG. 23, and the light-emitting thyristors . . . T−1, T0, T+1, T+2 . . . are used as transfer elements in this embodiment.

The light-emitting portion comprises writable light-emitting elements L−1, L0, L+1, L+2 . . . , each gate thereof is connected to the gate . . . G−1, G0, G+1 . . . , of the transfer elements . . . T−1, T0, T+1,T+2, respectively. A write signal Sin is applied to all of the anode of the writable light-emitting elements.

In the following, the operation of this self-scanning light-emitting device will be described. Assuming that the transfer element T0 is in the ON state, the voltage of the gate electrode G0 lowers below the supply voltage VGK and to almost zero volts. Consequently, if the voltage of the write signal Sin is higher than the diffusion potential (about 1 volt) of the pn junction, the light-emitting element L0 can be turned into a light-emission state.

On the other hand, the voltage of the gate electrode G−1 is about 5 volts, and the voltage of the gate electrode G+1 is about 1 volt. Consequently, the write voltage of the light-emitting element L−1 is about 6 volts, and the write voltage of the light-emitting element L+1 is about 2 volts. It follows from this that the voltage of the write signal Sin which can write only in the light-emitting element L0 is a range of about 1-2 volts. When the light-emitting element L0 is turned on, that is, in the light-emitting state, the voltage of the write signal Sin is fixed to about 1 volt. Thus, an error of selecting other light-emitting elements can be prevented.

Light emission intensity is determined by the amount of current fed to the write signal Sin, an image can be written at any desired intensity. In order to transfer the light-emitting state to the next element, it is necessary to first turn off the element that is emitting light by temporarily reducing the voltage of the write signal Sin down to zero volts.

INDUSTRIAL APPLICABILITY

According to the present invention, a light-emitting thyristor having an improved luminous efficiency may be provided, and furthermore a self-scanning light-emitting device composed of an array of light-emitting thyristors of the present invention and having a self-scanning function may be provided.

CLAIMS

1. A light-emitting thyristor, comprising: a GaAs substrate; and a GaAs buffer layer provided on the GaAs substrate; and four layers consisting of a first conductivity type of AlGaAs layer and a second conductivity type of AlGaAs layer stacked alternately on the buffer layer wherein the four layers form the light-emitting thyristor; wherein the AlGaAs layer just above the buffer layer is composed of a plurality of AlGaAs layers, Al compositions thereof being increased upward in steps.

2. The light-emitting thyristor of claim 1, wherein a quantum well layer or a strained superlattice structure is inserted into the uppermost layer of the plurality of AlGaAs layers.

3. A light-emitting thyristor, comprising: a GaAs substrate; and a GaAs buffer layer provided on the GaAs substrate; and four layers consisting of a first conductivity type of AlGaAs layer and a second conductivity type of AlGaAs layer stacked alternately on the buffer layer wherein the four layers form the light-emitting thyristor; wherein the Al composition of the AlGaAs layer just above the buffer layer is increased upward continuously.

4. The light-emitting thyristor of claim 3, wherein a quantum well layer or a strained superlattice structure is inserted into the AlGaAs layer just above the buffer layer.

5. A light-emitting thyristor, comprising: a GaAs substrate; a GaAs buffer layer provided on the GaAs substrate; and four layers consisting of a first conductivity type of AlGaAs layer and a second conductivity type of AlGaAs layer stacked alternately on the buffer layer; wherein a quantum well layer on a strained superlattice structure is inserted between the buffer layer and the AlGaAs layer just above the buffer layer, or into the AlGaAs layer just above the buffer layer.

6. A self-scanning light-emitting device, comprising: a structure in which a plurality of light-emitting elements each having a control electrode for controlling threshold voltage or current for light-emitting operation are arranged, the control electrodes of the light-emitting elements are connected to the control electrodes of the light-emitting elements are connected to the control electrode of at least one light-emitting element located in the vicinity thereof via an interactive resistor, and a plurality of wirings to which voltage or current is applied are connected to electrodes for controlling the light emission of the light-emitting elements, wherein the light-emitting element is a light-emitting thyristor as set forth in any one of claims 1-5.

7. A self-scanning light-emitting device, comprising: a structure in which a plurality of light-emitting elements each having a control electrode for controlling threshold voltage or current for light-emitting operation are arranged, the control electrodes for the light-emitting elements are connected to the control electrode of at least one light-emitting element located in the vicinity thereof via an electrically unidirectional element, and a plurality of wiring to which voltage or current is applied are connected to electrodes for controlling the light-emission of light-emitting elements, wherein the light-emitting element is a light-emitting thyristor as set forth in any one of claims 1-5.

8. The self-scanning light-emitting device of claim 7, wherein the electrically unidirectional element is a diode.

9. A self-scanning light-emitting device, comprising: a self-scanning transfer element array having such a structure that a plurality of transfer elements each having a control electrode for controlling threshold voltage or current for transfer operation are arranged, the control electrodes of the transfer elements are connected to the control electrode of at least one transfer element located in the vicinity thereof via an interactive resistor, power-supply lines are connected to the transfer elements by electrical means, and clock lines are connected to the transfer elements, and a light-emitting element array having such a structure that a plurality of light-emitting elements each having a control electrode for controlling threshold voltage or current are arranged, the control electrodes of the light-emitting element array are connected to the control electrodes of said transfer elements by electrical means, and lines for applying current for light emission of the light-emitting element are provided, wherein the light-emitting element is a light-emitting thyristor as set forth in any one of claims 1-5.

10. A self-scanning light-emitting device, comprising: a self-scanning transfer element array having such a structure that a plurality of transfer elements each having a control electrode for controlling threshold voltage or current for transfer operation are arranged, the control electrodes of the transfer elements are connected to the control electrode of at least one transfer element located in the vicinity thereof via an electrically unidirectional element, power-supply lines are connected to the transfer elements by electrical means, and clock lines are connected to the transfer elements, and a light-emitting element array having such a structure that a plurality of light-emitting elements each having a control electrode for controlling threshold voltage or current are arranged, the control electrodes of the light-emitting element array are connected to the control electrodes of said transfer elements by electrical means, and line for applying current for light emission of the light-emitting element are provided, wherein the light-emitting element is a light-emitting thyristor as set forth in any one of claims 1-5.

11. The self-scanning light-emitting device of claim 10, wherein the electrically unidirectional element is a diode.

COPYRIGHT

User acknowledges that Fairview Research and its third party providers retain all right, title and interest in and to this xml under applicable copyright laws. User acquires no ownership rights to this xml including but not limited to its format. User hereby accepts the terms and conditions of the License Agreement.