WO2005106981A1

WO2005106981A1 - Group iii nitride semiconductor light-emitting device

Info

Publication number: WO2005106981A1
Application number: PCT/JP2005/008551
Authority: WO
Inventors: Hisayuki Miki; Akira Bando; Takashi Udagawa
Original assignee: Showa Denko K.K.
Priority date: 2004-04-28
Filing date: 2005-04-28
Publication date: 2005-11-10

Abstract

A Group III nitride semiconductor light-emitting device provided on a crystal substrate with an n-type and p-type Group III nitride semiconductor of AlXGaYInZN1-aMa, wherein 0 ≤ X ≤ 1, 0 ≤ Y ≤ 1, 0 ≤ Z ≤ 1, X + Y + Z = 1, M denotes a Group V element other than N and 0 ≤ a < 1, includes an n-type electrode contact layer of Group III nitride semiconductor including a region doped with Ge and a light-emitting layer including a region undoped or doped with at least one element selected from the group consisting of Si, C, Sn and Pb.

Description

DESCRΓPTION

GROUP IN NITRIDE SEMICONDUCTOR LIGHT-EMITTING DEVICE

Cross Reference to Related Applications: This application is an application filed under 35 U.S. C. § 111 (a) claiming the benefit pursuant to 35 U.S.C. § 119 (e) (1) of the filing dates of Provisional Application No. 60/570,500 filed May 13, 2004 and Japanese Patent Applications No. 2004-132784 filed April 28, 2004 and No. 2005-17991 filed January 26, 2005 pursuant to 35 U.S.C. § 111 (b).

Technical Field: This invention relates to a Group III nitride semiconductor light-emitting device provided in an n-type contact layer thereof with a region doped with Ge. Background Art: The Group JJI nitride semiconductors have been heretofore finding utility as a functional material for configuring Group IJJ nitride semiconductor light-emitting devices of the pn junction-type structure, such as light-emitting diodes (LEDs) and laser diodes (LDs), which emit a visible light of a short wavelength (refer, for example, to JP-A 2000-332364). In the configuration of an LED assuming emission in the near ultraviolet band, the blue color band or the green color band, for example, the n-type or p-type aluminum gallium nitride (AlχGa_γN: 0 < X, Y < 1, X + Y = 1) is utilized for configuring a clad layer of the light-emitting device (refer, for example, to JP-A 2003- 229645). Then, gallium indium nitride (Ga_YIn_zN: 0 < Y, Z < 1, Y + Z = 1) is utilized for forming an active layer of the light-emitting layer (refer, for example to JP-B SHO 55-3834). In the conventional Group in nitride semiconductor light-emitting devices, the light-emitting layers thereof generally have an n-type or a p-type Group HI nitride semiconductor layer joined thereto. These layers are intended to configure a light- emitting part in a hetero-junction structure with the object of acquiring emission of high intensity. For the purpose of configuring a light-emitting part in a double-hetero (DH) junction structure, it has heretofore been customary to form light-emitting layers of GaylnzN (0 < Y, Z < 1, Y + Z = 1), for example, and have an n-type or a p-type Group III nitride semiconductor layer joined thereto as a clad layer (refer, for example, to "Group JH-N Compound Semiconductors" written by Isamu Akasaki and issued by Baifukan (K.K.) on May 20, 1995, chapter 13}). The contact layer which is intended to form an n-type electrode, for example, has heretofore been formed solely of a Group III nitride semiconductor having silicon (Si) added thereto. An n-type Al_xGa_γΝ (0 < X, Y < 1, X + Y = 1) layer, for example, which has the resistance thereof controlled by adjusting the doping amount of silicon has been available (refer, for example, to Japanese Patent No. 3383242). Similarly, Si has been used also as a donor element for doping an active layer. The so-called codoped configuration which uses an InGaN layer of a comparatively large film thickness for a light-emitting layer is doped with zinc (Zn) forming the center of emission and additionally doped with Si (refer, for example, to JP-A HEI 8-316528). Further, in the case of using a quantum well structure, it has been proposed to have a well layer doped and a barrier layer also doped as well. In the devices reported in these documents, Si is mainly used as the dopant for their light-emitting layers. (Refer, for example, to JP-A HEI 8-264831 and JP-A HEI 9-36423). Si, however, has also been used as the dopant for the n-electrode contact layer (otherwise called "n-contact layer") and no case of using an Si-doped active layer on a Ge-coped n-contact layer has been known to date. Germanium (Ge) has been well known as an n-type impurity (refer, for example, to JP-A HEI 4-170397). The germanium, however, is deficient in doping efficiency as compared with Si (refer, for example, Jpn. J. Appl. Phys., 31 (9 A) (1992), 2883) and, therefore, is rated as disadvantageous for producing an n-type Group III nitride semiconductor layer of low resistance. Further, it is held that when an n-type Group in nitride semiconductor layer is doped with Ge to a high concentration, the doping is at a disadvantage in inducing the surface of the layer to sustain pits which impair flatness (refer, for example, to "Group in Nitride Semiconductor Compounds," Clarendon Press (Oxford), 1998, page 104). In the Group ffl nitride semiconductor, Group IV elements, such as Si and Ge, which exhibit n-type conductivity when they are doped, are considered to be enabled to exist by being substituted for a Group HI element in a crystal. In gallium nitride, for example, they are substituted for Ga. Since Ge has a large atomic radius as compared with Si which is generally used as an n-dopant and is closely equal to Ga, it is considered to induce no change in the lattice constant of a Group in nitride semiconductor crystal even when it is made to dope this crystal. This fact is believed to be favorable because the active layer (light-emitting layer) formed on the n-contact layer does not suffer the crystallinity thereof to decline. The present inventors have found that the output of emission of a Group ffl nitride semiconductor light-emitting device is enhanced when the n-type electrode contact layer contains Ge as a dopant and the light-emitting layer is doped with any of the elements Si, C, Sn and Pb. This invention has been perfected on the basis of this knowledge. This invention is aimed at providing a light-emitting device possessing a Group III nitride semiconductor containing Ge as a dopant, which light-emitting device excels in the output of emission without impairing the flatness and the crystallinity of the semiconductor. Disclosure of the Invention: The Group HI nitride semiconductor light-emitting device according to this invention is provided on a crystal substrate with an n-type and p-type Group HI nitride semiconductor of Al_xGaγInzNι-aM_a, wherein O ≤ X ≤ l, O ≤ Y ≤ l, 0 < Z ≤ l and X + Y + Z = 1, M denotes a Group V element other than nitrogen (N) and 0 < a < 1, and comprises an n-type electrode contact layer of Group III nitride semiconductor including a region doped with germanium (Ge) and a light-emitting layer including a region not doped or doped with at least one element selected from the group consisting of Si, C, Sn and Pb. In the Group HI nitride semiconductor light-emitting device, the n-type electrode contact layer of Group in nitride semiconductor is provided with a Ge atom low concentration layer and a Ge atom-doped Ge atom high concentration layer. In the Group ffl nitride semiconductor light-emitting device, the Ge atom low concentration layer is a Ge atom-doped Ge atom low concentration layer or an undoped layer. In the Group ffl nitride semiconductor light-emitting device, the n-type electrode contact layer is configured in a structure having Ge atom low concentration layer and high concentration layer alternately stacked periodically. In the Group ffl nitride semiconductor light-emitting device, the Ge atom high concentration layer has a smaller thickness than the Ge atom low concentration layer. In the Group ffl nitride semiconductor light-emitting device, the Ge atom high concentration layer has a surface having concave pits formed therein. In the Group ffl nitride semiconductor light-emitting device, the Ge atom high concentration layer, when formed in a thickness of 10 nm or more, has a surface having pits formed therein in a number in the range of 1 x 10⁵ /cm² to 1 x 10¹⁰ /cm². This invention comprises the fact that the surface of the Ge atom low concentration layer has flatness of not more than 10 A in Ra. This invention comprises the fact that the concentration of germanium atoms in the aforementioned Ge-doped Group ffl nitride semiconductor layer is not less than 1 x 10¹⁷ cm^"3 and not more than 1 x 10²⁰ cm^"3. In the Group ffl nitride semiconductor light-emitting device, the light-emitting layer has a multiple quantum well structure. In the Group ffl nitride semiconductor light-emitting device, the region of the light-emitting layer is doped with Si and is a barrier layer of a multiple quantum well structure. In the Group ffl nitride semiconductor light-emitting device, the region of the light-emitting layer is doped with Si and has an Si atom concentration of 5 x 10 cm^" or more and 1 x 10¹⁹ cm^"3 or less. According to this invention, by causing the n-type electrode contact layer of the Group ffl nitride semiconductor light-emitting device to contain Ge as a dopant and causing the light-emitting layer thereof to be doped with at least one of the elements Si, C, Sn and Pb, it is made possible to obtain a light-emitting device excelling in crystallinity and abounding in intensity of the output of emission while avoiding to impair the flatness of the semiconductor layers. The n-type Group ffl nitride semiconductor layer manufactured by using the technique disclosed by this invention enjoys low resistance and excels in the flatness of surface as compared with the film formed by being doped with Ge without differentiating concentration thereof and avoids impairing the crystallinity of a light- emitting layer formed thereon. By inducing the formation of fine pits on a dry etched surface, it is enabled to realize a good contact resistance with an electrode. Ge is less easily diffused in a crystal than Si. By using Ge as a dopant in the configuration of a device, therefore, it is made possible to give an abrupt interface between the doped layer and the undoped layer. The device realizes the characteristic property of keeping this abruptness of the interface intact in spite of aging. Brief Explanation of the Drawing: Fig. 1 is a schematic cross section illustrating the constitution of a stacked structure described in Example 1. Fig. 2 is a schematic plan view of the LED described in Example 1. Fig. 3 is a schematic cross section illustrating the constitution of a stacked structure described in Example 2. Fig. 4 is a schematic plan view of the LED described in Example 2. Fig. 5 is a schematic cross section illustrating the constitution of a stacked structure described in Comparative Example 1. Fig. 6 is a cross section depicting an artist's concept of the layer structure in which the pits occurring in a Ge high concentration layer are filled with a Ge low concentration layer.

Best Mode of Embodying the Invention: The light-emitting device of this invention is a Group III nitride semiconductor light-emitting device is provided on a crystal substrate with an n-type and p-type Group ffl nitride semiconductor of AlχGaγIn_zNι.aMa, wherein O ≤ X ≤ l, O ≤ Y ≤ l, O ≤ Z ≤ l, X + Y + Z = 1, M denotes a Group V element other than nitrogen (N) and 0 < a < 1, and comprises an n-type electrode contact layer of Group ffl nitride semiconductor including a region doped with germanium (Ge) and a light-emitting layer including a region not doped or doped with at least one element selected from the group consisting of Si, C, Sn and Pb. The n-type electrode contact layer is preferred to have the Ge concentration periodically varied or have the Ge-doped region and the undoped region stacked periodically. In the light-emitting device of this invention as described above, the Group III nitride semiconductor device containing germanium atom in the n-contact layer contemplated by this invention and at least one of the elements Si, C, Sn and Pb in the light-emitting layer is configured on a substrate formed of an oxide single crystal material, such as sapphire (α-Al₂O₃ single crystal), zinc oxide (ZnO) or gallium lithium oxide (GaLiO₂), or a Group IN semiconductor single crystal, such as silicon (Si) single crystal or a cubic or hexagonal silicon carbide (SiC). For the material of the substrate, Group TTT-N compound semiconductor single crystal materials, such as gallium phosphide (GaP) and gallium arsenide (GaAs) are also available. The single crystal plate formed of gallium nitride crystal is included therein. The optically transparent single crystal materials which are capable of passing the light emitted from the light- emitting layer can be effectively utilized as the substrate. For the purpose of stacking a gallium nitride-based compound semiconductor on the aforementioned substrate which in principle allows no lattice matching with a gallium nitride-based compound except the GaN substrate, the low-temperature buffer method disclosed in Japanese Patent No. 3026087 and JP-A HEI 4-297023 and the lattice unmatching crystal epitaxial growth technique called the seeding process (SP) disclosed in JP-A 2003-243302 can be used. Particularly, the SP method which produces an AIN crystal film at a temperature high enough for the production of a GaN- based crystal proves an excellent lattice unmatching crystal epitaxial growth technique from the viewpoint of the enhancement of productivity. When the lattice unmatching crystal epitaxial growth technique, such as the low temperature buffer method or the SP method, is used, the gallium nitride-based compound semiconductor to be stacked as the base is preferred to be GaN which is undoped or doped to a low level of about 5 x 10¹⁷ cm^"3. The film thickness of the base layer is preferred to be in the range of 1 to 20 μm and more preferably in the range of 5 to 15 μm. The light-emitting device which is provided with a Group ffl nitride semiconductor containing germanium atom in the n-contact layer contemplated by this invention and the aforementioned Si in the light-emitting layer of AlχGaγInzNι-aM_a, wherein 0 < X < 1, O ≤ Y ≤ l, O ≤ Z ≤ l, X + Y + Z = 1, M denotes a Group V element other than nitrogen (N) and 0 < a < 1, can be configures by vapor phase growth means, such as the metal organic chemical vapor deposition method (abbreviated as MOCND, MONPE or OMNPE), the molecular beam epitaxial (MBE) method, the halogen gas phase growth method and the hydride gas phase growth method. As the source for germanium to be added, organic germanium compounds, such as germane gas (GeELt), tetramethyl germanium ((CH₃)₄Ge) and tetraethyl germanium ((C₂H₅)₄Ge), can be used. For the MBE method, elemental germanium can be used as the doping source. By the MOCND method, for example, the n-type gallium nitride layer is formed on the sapphire substrate by the use of (CH₃)₄Ge. As regards the structure of the n-contact layer, the technique for producing the flat surface is directed toward producing the structure having the concentration of germanium atom varied periodically. This region is formed by periodically varying the amount of the doping source to be supplied to the vapor phase growth reaction system with time during the duration of the vapor phase growth of the Group ffl nitride semiconductor layer. For example, a thin layer containing Ge atom at a high concentration is formed by forming an undoped thin layer by suspending the supply of the Ge doping source to the vapor phase growth region and thereafter supplying instantaneously the Ge doping source in a large amount of the vapor phase growth region. The region having the germanium atom concentration varied periodically can be formed through fluctuation of the amount of the Ge doping source to be supplied to the vapor phase growth reaction system. It may be formed in conjunction with a thin layer containing Ge atom in a high concentration by growing a thin layer having a low Ge atom concentration and subsequently suspending this growth till the growth conditions, such as the N/III ratio, are adjusted so as to suit the addition of the Ge atom at a high concentration. In this invention, the layer having Ge atom in a low concentration means a layer doped with Ge atom to a low concentration and an undoped layer. These layers may coexist. When the Ge atom low concentration layer and high concentration layer are alternately varied periodically, the Ge concentration does not need to be fixed between the individual periods. In the case of forming the region having the Ge atom concentration periodically varied, the layer thickness of the entire region having the germanium atom concentration periodically varied is properly 0.1 μm or more and 10 μm or less, preferably 0.3 μm or more and 5 μm or less, and more preferably 0.5 μm or more and 3 μm or less. If the layer thickness falls short of 0.1 μm, the shortage will result in preventing the n-type Group ffl nitride semiconductor layer of low resistance from being produced. If the layer thickness exceeds 10 μm, the overage will give no proportionate addition to the effect to be derived. The total of the film thickness of the n-type Group ffl nitride semiconductor layer containing Ge at a high concentration and the film thickness of the n-type Group III nitride semiconductor layer containing Ge at a low concentration, namely the periodic film thickness, is properly 1 nm or more and 1000 nm or less, preferably 4 nm or more and 400 nm or less, and more preferably 6 nm or more and 100 nm or less. If the total of the film thicknesses falls short of 1 nm, the shortage will result in rendering it difficult to attain the effect of periodically stacking the Ge doped layer. Conversely, if this total exceeds 1000 nm, the overage will result in obstructing suppression of the occurrence of pits or inducing addition to the resistance. That is, when the Ge high concentration layer has a greater thickness than the Ge low concentration layer within one period, the occurrence of pits cannot be suppressed and the flatness of the surface is not easy to obtain. The flatness of the surface is good when the Ge low concentration layer has a thickness equal to or greater than the Ge high concentration layer. Thus, the Ge low concentration layer is preferred to have a greater thickness than the Ge high concentration layer. When the Ge low concentration layer is formed of an undoped n-type Group ffl nitride semiconductor thin film, the effect of filling the pits occurring in the surface of the n-type Group III nitride semiconductor thin layer containing Ge atom at a high concentration is further exalted and the effectiveness of producing a Ge-doped Group ffl nitride semiconductor thin film having flatness of surface is enhanced. An undue increase of the thickness of the low concentration layer, however, is at a disadvantage in suffering the resistance to rise and adding to the contact resistance of the n-electrode. When the low concentration layer is large, this fact puts the production of the Group ffl nitride semiconductor light-emitting device having a low forward voltage (so- called Nf) or threshold voltage (so-called Nth) at a disadvantage. The thickness of the n-type Group ffl nitride semiconductor thin layer having a low Ge atom concentration, therefore, is properly 500 nm or less. This layer thickness is preferably decreased in proportion as the Ge concentration of the low concentration layer decreases and the carrier concentration decreases. The number of stacking cycles is properly 1 or more and 10000 or less, preferably 10 or more and 1000 or less, and more preferably 20 or more and 200 or less. For example, by taking the composite of a high concentration Ge-doped GaΝ thin layer having a thickness of 10 nm and a low concentration Ge-doped GaΝ layer having a thickness of 10 nm as one period and performing stacking of such composites up to 100 periods, it is made possible to form a region which has a total thickness of 2 μm and has the Ge atom concentration periodically varied. The film thickness of the n-type Group ffl nitride semiconductor layer containing Ge in a high concentration is properly 0.5 nm or more and 500 nm or less, preferably 2 nm or more and 200 nm or less, and more preferably 3 nm or more and 50 or less. If the film thickness falls short of 0.5 nm, the shortage will result in preventing the Ge doping from proceeding sufficiently and suffering the resistance to rise unduly. Conversely, if the film thickness exceeds 500 nm, the overage will result in preventing the low concentration layer from filling the pits completely and, when the low concentration layer is given a thickness sufficient for completing the filling, likewise suffering the resistance to rise unduly. Then, the film thickness of the n-type Group ffl nitride semiconductor layer containing Ge in a low concentration is properly 0.5 nm or more and 500 nm or less, preferably 2 nm or more and 200 nm or less, and more preferably 3 nm or more and 50 nm or less. If this film thickness falls short of 0.5 nm, the shortage will result in preventing the pits formed in the Ge-doped layer from being filled thoroughly and impairing the flatness of the surface. This film thickness, therefore, is preferred to be greater than the film thickness of the n-type Group III nitride semiconductor layer containing Ge in a high concentration. If it exceeds 500 nm, however, the overage will be at a disadvantage in suffering the resistance to rise unduly. The concentration of the Ge atom in the interior of the n-type Group ffl nitride semiconductor layer containing Ge in a high concentration is properly 1 x 10¹⁷ cm^"3 or more and 1 x 10²⁰ cm^"3 or less, preferably 1 x 10¹⁸ cm^"3 or more and 5 x 10¹⁹ cm^"3 or less, and more preferably 3 x 10 cm^" or more and 2 x 10 cm^" or less. The concentration of Ge atom in the interior of the n-type Group III nitride semiconductor layer containing Ge in a high concentration does not need to be fixed but may be continuously or discontinuously varied. The Ge atom concentration in the interior of an n-type Group III nitride semiconductor layer doped with Ge to a low concentration is lower than the Ge atom concentration in the interior of an n-type Group III nitride semiconductor layer containing Ge in a high concentration, is properly the minimum limit of the determination by the following analytical method or more and 2 x 10¹⁹ cm^"3 or less, preferably the minimum limit of the determination or more and 1 x 10¹⁹ cm^"3 or less, and more preferably the minimum limit of the determination or more and 5 x 10 cm^" or less. It is rather preferable to keep the layer not doped, namely undoped. Then, the Ge atom concentration in the interior of the n-type Group ffl nitride semiconductor layer containing Ge in a low concentration does not need to be uniform but may be varied continuously or discontinuously. If the Ge atom concentration exceeds 2 x 10¹⁹ cm^"3, the overage will be at a disadvantage in suddenly increasing the density of pits in the surface. The Ge atom concentration can be determined, for example, by the method of secondary ion mass spectroscopy (abbreviated as "SIMS"). This method relates to a technique that comprises irradiating the surface of a sample with a primary ion, thereby ionizing the relevant elements, expelling the ions and subjecting the ions to mass analysis. It permits observation and determination of concentration distributions of specific elements in the direction of depth of the sample. This technique is effective in the analysis of Ge element which is present in the Group ffl nitride semiconductor layer. The increase of the Ge concentration in the Ge high concentration layer to 1 x 10¹⁷ cm^"3 or more can contribute to the configuration of an LED having a low forward voltage. When the upper limit of the concentration of Ge in the Ge high concentration layer is preferably set at 1 x 10²⁰ cm^"3, the carrier concentration in the whole Ge- containing region having the germanium atom concentration periodically varied, for example, is roughly (3 to 4) x 10¹⁹ cm^"3. If the layer is doped with Ge in excess of this atom concentration, the overage will be at a disadvantage in suddenly increasing the density of pits in the surface. The carrier concentration in the Ge-doped layer can be determined as regarding the whole structure having a high concentration layer and a low concentration layer alternately stacked as one layer. The carrier concentration found in this case is approximately the average of the products of the amounts of Ge used for doping the high concentration layer and the low concentration layer multiplied by the ratio of film thicknesses of the layers. The determination of the carrier concentration can be carried out by the Hall effect determination according to the ordinary Nan de Paw method or by the C-N method. The region having the concentration of Ge atom periodically varied can be disposed anywhere inside the n-type Group ffl nitride semiconductor layer. For example, it may be disposed as directly joined to the surface of the crystal substrate. It may otherwise be disposed as joined to the upper surface of a buffer layer which is provided on the surface of the crystal substrate. When the region having the concentration of Ge atom periodically varied is disposed below the n-type Group ffl nitride semiconductor layer approximating closely to the crystal substrate or the buffer layer, the n-type Group ffl nitride semiconductor layer excelling in crystallinity is obtained. By thus disposing the region having the concentration of Ge atom periodically varied, it is made possible to prevent the misfit transfer due to the lattice mismatch with the crystal substrate from being propagated to the parts beyond the relevant layer. In this case, the periodic layer thickness may be increased from 0.5 μm to 5 μm. In the region having the concentration of Ge atom periodically varied, the threading dislocation from below can be prevented from being propagated to the upper layers. Thus, when the region having the concentration of Ge atom periodically varied is disposed as a base layer for the configuration of a light-emitting layer on the n-type Group III nitride semiconductor layer, it proves effective in forming a light-emitting layer excelling in crystallinity. This disposition consequently can contribute to the production of a Group ffl nitride semiconductor light-emitting device having high intensity of emission. The n-type Group III nitride semiconductor contact layer manufactured using the technique disclosed by this invention allows an n-type semiconductor crystal film of very low resistance because it is capable of realizing an even surface in spite of such a high dopant concentration as inherently induces the occurrence of pits in the surface. The high concentration layer to be used by the technique proposed by this invention has such a high concentration that the layer, when Ge is used as a dopant, inherently induces the occurrence of pits in the surface. By filling the pits with a low concentration layer, the layer is enabled to realize an even surface as compared with the layer doped with Ge to a high concentration by the conventional method. Specifically, by this invention, in the interface between the Ge high concentration layer and Ge low concentration layer, the surface on the high concentration layer side is made to contain concave pits and the surface on the low concentration layer side is made to acquire an even surface. A cross section depicting an artist's concept of the layer structure in which the pits occurring in the Ge high concentration layer 4aare filled with the Ge low concentration layer 4b is shown in Fig. 6. The pits which occur in the Ge high concentration layer 4a of this invention are thought to occur at the positions of the so-called threading dislocation originating in the interface between a substrate and a nitride-based compound semiconductor layer. More often than not, the density of pits occurring in the high concentration layer roughly concurs with the density of threading dislocations in the base. The threading dislocations in the base occur in the range of 1 x 10⁷/cm² to 1 x 10¹⁰/cm² generally in the GaN crystal on a sapphire substrate. The product having less than 1 x 10⁷/cm² of pits has not been fully materialized to date. The product having pits exceeding 1 x 10¹⁰/cm² cannot show fully satisfactorily the function thereof when used as a substrate for an electronic device. The density of pits, though depending on the density of threading dislocations in the base, is in the range of 1 x 10⁵/cm² to 1 x 10¹⁰/cm². Generally, it is in the range of 1 x 10⁶/cm² to 1 x 10⁹/cm². These pits occur on the surface when the high concentration layer alone is formed in a film thickness of about 10 nm or more. They can be visually observed by such means as an atomic force microscope (AFM). When the film thickness is further increased to about 500 nm, the pits can be observed with an optical microscope. The low concentration layer of this invention is preferred to have an even surface. The flatness of the surface is preferred to be 10 A or less in terms of Ra and more preferably 5 A or less. The technique proposed by this invention, when applied to a stacked layer which is configured by filling the pits occurring in a high concentration layer with a low concentration layer, is useful in a contact layer. In the case of a gallium nitride-based compound semiconductor, it is usual in the manufacture of an n-electrode to carry out the removal of a film by dry etching. This dry etching treatment inflicts concave pits on a stacked layer which is contemplated by this invention. The concaves are capable of suppressing the drive voltage of a given device to a low level because they add to the surface area of the electrode metal and lower the contact resistance by virtue of the anchor effect. The active layer may be configured with a single layer of InGaN, for example, or may be configured in a quantum well structure. The effect of this invention ^"is showed particularly when the quantum well structure is adopted. The quantum well structure may be a single quantum well structure which is formed of a sole layer. The multiple quantum well structure having a well layer that is an active layer and a barrier layer alternately stacked up to a plurality of repetitions proves favorable because it produces an enhanced output of emission. The number of repetitions is preferably from 3 through 10 approximately and more preferably from 3 through 6 approximately. In the case of the multiple quantum well structure, all the well layers (active layers) do not need to be provided with a thick film part and a thin film part. The thick film parts and the thin film parts may have their dimensions and the surface ratios varied individually. Incidentally, in the case of the multiple quantum well structure, the whole mass combining the well layer (active layer) and the barrier layer is referred to as the "light- emitting layer" in the present specification. The film thickness of the barrier layer is preferably 70 A and more, preferably

140 A or more. If the film thickness of the barrier layer is unduly small, the shortage will result in obstructing the impartation of flatness to the upper surface of the barrier layer and inducing degradation of the emission efficiency and the aging property. If the film thickness is unduly large, the overage will induce an increase of the drive voltage and a decrease of the emission. Thus, the film thickness of the barrier layer is preferred to be 500 A or less. The active layer is preferred to be a gallium nitride-based compound semiconductor which contains In. The In-containing gallium nitride-based compound semiconductor is capable of emitting light in the region of wavelength of blue color with high intensity. In the case of the multiple quantum structure, the barrier layer may be formed of InGaN having a smaller In ratio than the InGaN forming the well layer (active layer), besides GaN and AlGaN. Among other materials mentioned above, GaN proves particularly favorable. When the active layer is configured in the multiple quantum well structure in an undoped form, the well layer may be configured in a structure containing a region of a large film thickness and a region of a small film thickness. By configuring the well layer in this structure, the decrease of the drive voltage can be materialized. This structure can be formed by growing a well layer at a comparatively low temperature ranging from 600°C to 900°C and subsequently performing a step of increasing the temperature while the growth is kept in a suspended state. When the active layer is doped with Si, the organic silicon materials can be used besides silane (SiE ) and disilane (Si₂H₆) which are well known as the sources for dopants. While silane (S1H₄) and disilane (Si₂H₆) may be supplied each in the form of 100% gas, they are preferably supplied from cylinders containing them in a diluted form from the viewpoint of safety. When the active layer is doped with Si, the entire region may be doped or only a part of the region may be doped. Particularly, when the barrier layer in the configuration adopting the quantum well structure is doped with an n-dopant, the doping of the barrier with Si proves favorable because the n-dopant brings the effect of lowering the drive voltage of the device. In this case, the doping may be performed on the whole barrier layer and also on part of the region. Particularly, by selectively doping the region which lies directly below the well layer, it is made possible to reconcile high output and low drive voltage. The concentration of doping with Si is properly 5 x 10¹⁶ cm^"3 or more and 1 x 10¹⁹ cm^"3 or less. If this concentration falls short of the lower limit of this range, the shortage will prevent the reduction of the drive voltage from being realized. If it exceeds the upper limit of the range, the overage will result in degrading the crystallinity and the surface flatness. The concentration is more preferably 1 x 10 cm^" or more and 5 x 10 cm^" or less and most preferably 1 x 10 cm^"3 or more and 1 x 10¹⁸ cm^"3 or less. The active layer prefers Si doping and may be also doped with C, Sn or Pb. For these dopants, raw materials, such as methane (CF- ), tetramethyl tin (TMSn) and tetramethyl lead (TMPb) are used. Preferably an n-clad layer is interposed between the contact layer and the light- emitting layer. The interposed n-clad layer is capable of compensating the deterioration of the flatness of the outermost surface of the n-contact layer. The n-clad layer may be formed of AlGaN, GaN, InGaN, etc. When it is made of InGaN, it is only natural that it be given a composition larger than the band gap of InGaN of the active layer. The carrier concentration in the n-clad layer may be equal to or larger or smaller than that of the n-contact layer. For the purpose of enabling the active layer formed thereon to acquire enhanced crystallinity, the n-clad layer is preferred to acquire a surface of high flatness by suitably adjusting the growth conditions, such as the speed of growth, the temperature of growth, the pressure of growth and the amount of a dopant to be used. The n-clad layer may be configured by having two layers differing in composition and lattice constant alternately stacked up to several repetitions. In this case, the amount of a dopant and the film thickness may be varied besides the composition, depending on the layers to be stacked. The p-type layer generally has a thickness in the range of 0.01 to 1 μm and is composed of a p-clad layer contiguous with the active layer and a p-contact layer intended to form a positive electrode. The p-clad layer and the p-contact layer can concurrently function. The p-clad layer is formed by using GaN, AlGaN, etc. and is doped with Mg as a p-dopant. The outermost surface of the layer is preferred to be formed as a layer having a high carrier concentration with a view to allowing easy contact with the electrode. Most layers, however, tolerate high resistance. The amount of a dopant may be safely decreased and the hydrogen which is held to obstruct the activation of a dopant may be safely contained. These actions are rather preferable because they enhance the reverse voltage when the relevant layer is incorporated in a completed device. The p-clad layer may be configured by alternately stacking two layers differing in composition and lattice constant up to a plurality of repetitions. The component layers used in this case may be varied in the amount of a dopant and the film thickness besides the composition. For the p-contact layer, GaN, AlGaN, InGaN, etc. may be used. This layer is doped with Mg as an impurity. The gallium nitride-based compound semiconductor doped with Mg in a state fresh from the reaction furnace generally shows high resistance. It is held to show p-conductivity when it is subjected to a treatment for activation, such as the annealing treatment, the treatment of electron beam irradiation, and the treatment of microwave irradiation. This semiconductor, however, may be put to use without undergoing the treatment of activation as already pointed out. For the p-contact layer, boron phosphide doped with a p-type impurity may be used. The boron phosphide which is doped with a p-type impurity shows p- conductivity without undergoing the aforementioned treatment for conversion to a p- type. The method for growing the gallium nitride-based compound semiconductor which is used for forming the n-type layer, the active layer and the p-type layer does not need to be particularly restricted. The universally known methods, such as MBE, MOCVD and HNPE, may be used for the growth under universally known conditions. Among other methods cited above, the MOCND method proves particularly favorable. As the raw materials, ammonia, hydrazine, azides, etc. may be used for the sources of nitrogen. As the Group ffl organic metals, trimethyl gallium (TMGa), triethyl gallium (TEGa), trimethyl indium (TMIn), trimethyl aluminum (TMA1), etc. may be used. Then, as the dopant sources, silane, disilane, germane, organic germanium raw materials, biscycopentadienyl magnesium (Cp₂Mg), etc. may be used. As the carrier gases, nitrogen and hydrogen may be used. The growth of the active layer containing In is preferably performed at a substrate temperature in the range of 650 to 900°C. If the substrate temperature falls short of the lower limit of this range, the shortage will result in preventing the active layer to be produced from acquiring good crystallinity. If this temperature exceeds the upper limit of the range, the overage will possibly result in decreasing the amount of In to be incorporated in the active layer and possibly preventing a device to be manufactured from emitting at a wavelength aimed at. When the active layer is in the multiple quantum well structure, the growth of part of the region of the barrier layer is preferred to be performed at a substrate temperature higher than is used in the growth of the well layer (active layer). This temperature is appropriately in the range of 700 to 1000°C. The negative electrode is well known in various compositions and various structures. Any of these known negative electrodes can be used without any particular restriction. As the contact material for use in the negative electrode destined to contact the n-contact layer, Cr, W, N, etc. can be used besides Al, Ti, Νi and Au. Naturally, the whole negative electrode may be configured in a multilayer structure so as to be endowed with a bonding property. Particularly, the coating of the outermost surface of the negative electrode with Au proves favorable for the purpose of facilitating the bonding. The positive electrode is well known in various compositions and various structures. Any of these known positive electrodes can be used without any particular restriction. The raw materials for the transparent positive electrode may contain Pt, Pd, Au,

Cr, Ni, Cu, Co, etc. It is known that the positive electrode formed in a partly oxidized structure acquires enhanced perviousness to light. As the raw materials for the reflective positive electrode, Rh, Ag, Al, etc. can be used besides the raw materials enumerated above. These positive electrodes can be formed by methods, such as sputtering and vacuum evaporation. Particularly, the use of the sputtering proves favorable because it permits the ohmic contact to be obtained by properly controlling the sputtering conditions without requiring the annealing treatment after the formation of the electrode film. The light-emitting device may be configured in the flip chip form which is furnished with a reflective positive electrode, in the lattice form which is furnished with a light-pervious positive electrode, or in the faceup form which is furnished with a comb-shaped positive electrode. Now, this invention will be described below with reference to examples thereof. It should be noted, however, that this invention is not limited to these examples. Example 1 : First, this invention will be specifically described below by citing as an example the case of configuring a Group III nitride semiconductor light-emitting diode by stacking an undoped light-emitting layer on a Ge-doped GaN layer having the concentration of Ge periodically varied. In the description, the dopant concentrations reported were invariably determined by the SIMS method described above. The film thicknesses reported therein were determined by the method using the reflectance spectrum of white color and the observation through the section tunneling electron microscope (TEM). The same will apply to Example 2 and subsequent examples. Fig. 1 schematically illustrates the sectional structure of an epitaxially laminated structure 11 for the configuration of LED described in the example. The schematic cross section of an LED chip configured in this example is shown in Fig. 2. The epitaxially stacked structure was configured by utilizing the ordinary reduced pressure MOCND means in accordance with the following procedure. First, a (OOOl)-sapphire substrate 101 was mounted on a susceptor made of high-purity graphite designed specially for a semiconductor destined to be heated to a film forming temperature with a radio frequency (RF) induction heater. After this mounting, the interior of a vapor phase growth reaction furnace made of stainless steel was swept with a stream of nitrogen gas till it was thoroughly purged. The stream of nitrogen gas through the interior of the vapor phase growth reaction furnace was continued for 8 minutes. Then, the induction heater was set operating to elevate the temperature of the substrate 101 from room temperature to 600°C over a period of 10 minutes. While the temperature of the substrate 101 was kept at 600°C, hydrogen gas and nitrogen gas were fed to the vapor phase growth reaction furnace till the pressure therein reached 1.5 x 10⁴ pascals (Pa). The resultant mixture was left standing at the temperature under the pressure mentioned above for 2 minutes to effect thermal cleaning of the surface of the substrate 101. After the thermal cleaning was terminated, the supply of the nitrogen gas into the vapor phase growth reaction furnace was stopped. The supply of the hydrogen gas was continued. Thereafter, the temperature of the substrate 101 in the atmosphere of hydrogen was elevated to 1120°C. After confirming that the temperature was stabilized at 1120°C, a hydrogen gas entraining the vapor of trimethyl aluminum (TMAl) was supplied to the interior of the vapor phase growth reaction furnace for 8 minutes and 30 seconds. Consequently, the vapor-entraining hydrogen gas was allowed to react with the nitrogen (N) atom produced in consequence of the decomposition of the nitrogen (N)-containing deposit adhering formerly to the inner wall of the vapor phase growth reaction furnace to induce adhesion of an aluminum nitride (AIN) thin film (not shown) several nm in thickness on the sapphire substrate 101. After the supply of the hydrogen gas entraining the vapor of TMAl to the interior of the vapor phase growth reaction furnace was stopped to terminate the growth of AIN, the furnace was left standing for 4 minutes to effect thorough expulsion of the residual TMAl from the interior of the vapor phase growth furnace. Subsequently, supply of ammonia (NH₃) gas to the interior of the vapor phase growth reaction furnace was started. After 4 minutes from the start of the supply, the temperature of the susceptor was lowered to 1040°C while the supply of ammonia gas was continued. After the arrival of the temperature of the susceptor at 1040°C was confirmed, the stabilization of the temperature was awaited for a while and the supply of trimethyl gallium (TMGa) to the interior of the vapor phase growth reaction furnace was then started and the consequent growth of an undoped GaN layer 102 was allowed to continue for 4 hours. The undoped GaN layer 104 acquired a film thickness of 8 μm. Then, the cycle comprising the steps of elevating the wafer temperature to

1140°C, allowing the temperature to stabilize, then feeding tetramethyl germanium (hereinafter abbreviated as "(CH₃)₄Ge") to the furnace and thereafter stopping the feeding was carried out up to 100 repetitions to effect configuration of a Ge-doped GaN layer 103 measuring 2.0 μm in thickness and having Ge concentration periodically varied. In this contact layer, the Ge concentration in the Ge high concentration layer was about 1.2 x 10¹⁹ cm^"3 and the Ge concentration in the Ge low concentration layer was about 1 x 10¹⁸ cm^"3. The thickness of the Ge high concentration layer was about 8 nm and the thickness of the Ge low concentration layer was about 10 nm. Incidentally, the high concentration layer used in the present example, in a separate film forming experiment, was found to form 2 x 10⁷ pits per cm² of the surface thereof when the layer was formed in a thickness of 10 nm. After the stacking of the Ge-doped GaN layer, an undoped n-type In₀.o6Gao.₉ N clad layer 104 was stacked thereon. The film thickness of this clad layer 104 was 12.5 nm. Then, the temperature of the substrate 101 was set at 730°C, and a light-emitting layer 105 of a five-period type multiple quantum well structure including a barrier layer of GaN and a well layer of In₀.₂₅Ga₀.₇5N was disposed on the undoped n-type

Ino.06Gao.094N clad layer 104. In the light-emitting layer 105, first the GaN barrier layer was disposed as joined to the undoped n-type Ino.₀6Ga₀.₉₄N clad layer 104. The GaN barrier layer was grown using triethyl gallium (TEGa) as a gallium source. This layer had a thickness of 16 nm and was made undoped. The Ino.₂₅Gao.₇5N well layer was grown using triethyl gallium (TEGa) as a gallium source and trimethyl indium (TMIn) as an indium source. The layer had a thickness of 2.5 nm and was made undoped. On the light-emitting layer 105 of the multiple quantum well structure, a p-type Alo.o7Gao. ₃N clad layer 106 doped with magnesium (Mg) was formed. The thickness of this layer was 10 nm. On the p-type Alo.₀₇Gao.₉₃N clad layer 106, a p-type GaN contact layer 107 doped with Mg was further formed. As the source for Mg doping, bis- cyclopentadienyl Mg (Bis-Cp₂Mg) was used. The Mg was so added that the p-type GaN contact layer 107 might acquire a hole concentration of 8 x 10¹⁷ cm^"3. The thickness of the p-type GaN contact layer 107 was 100 nm. After the growth of the p-type GaN contact layer 107 was completed, the conduction of electricity to the induction heater was stopped and the temperature of the substrate 101 was lowered to room temperature over a period of about 20 minutes. During the fall of the temperature, the atmosphere in the vapor phase growth reaction furnace was formed solely of nitrogen and the flow rate of NH₃ was decreased. Thereafter, the supply of NH₃ was stopped. After the fall of the temperature of the substrate 101 to room temperature was confirmed, the stacked structure 11 was taken out of the vapor phase growth reaction furnace. At this point of time, the aforementioned p-type GaN contact layer 107 showed p-type conductivity without undergoing the annealing treatment necessary for electrical activation of the p-type carrier (Mg). Subsequently, the surface of the highly Ge-doped GaN layer 103 was exposed exclusively in the region planned to seat an n-type ohmic electrode 108 by utilizing the known photolithographic technique and the ordinary dry etching technique. On the exposed surface of the Ge-doped n-type GaN layer 103, an n-type ohmic electrode 108 having chromium (Cr) and gold (Au) stacked therein on the surface side was formed. On the entire surface of the p-type GaN contact layer. 107 constituting the residual surface of the stacked structure 11, a reflection type ohmic electrode 109 having platinum (Pt), silver (Ag) and gold (Au) stacked sequentially from the surface side was formed by utilizing the ordinary sputtering technique and the known photolithographic technique. Thereafter, an LED chip 10 measuring the square of 350 μm in the plan view was cut from the stacked structure and welded to a wire connecting auxiliary member called a "submount." The chip and the mount thus joined were mounted on a lead frame (not shown) so that a gold wire (not shown) connected to the lead frame might be enabled to advance the device drive current from the lead frame to the LED chip 10. The device drive current was passed in the forward direction between the n-type and p-type ohmic electrodes 108 and 109 via the lead frame. The forward voltage was 3.5 V when the forward current was set at 20 mA. The central wavelength of the blue band light emitted during the passage of the forward current at 20 mA was 460 nm. The intensity of the emission determined by using the ordinary integrating sphere reached 12 mW. This fact indicates that a Group ffl nitride semiconductor LED capable of producing emission of high intensity was obtained. Example 2: First, this invention will be specifically described by citing as an example the case of configuring a Group ffl nitride semiconductor light-emitting diode by stacking a light-emitting layer 111 of a multiple quantum well structure having a barrier layer exclusively doped with Si via an n-type clad layer 104 on a Ge-doped n-type GaN contact layer 103 having the Ge concentration periodically varied. Fig. 2 schematically illustrates a sectional construction of an epitaxially stacked structure 12 for the configuration of a LED described in the present example. The manufacture till an undoped base layer 102 was carried out by following the procedure of Example 1. Subsequently, the cycle comprising the steps of elevating the wafer temperature to 1150°C, awaiting stabilization of this temperature, supplying tetraethyl germanium (hereinafter referred to as "(C₂H₆)₄Ge"), and stopping the supply was performed up to 100 repetitions to form a Ge-doped GaN layer 103 measuring 2.0 μm in thickness and having the Ge concentration varied periodically. The Ge concentration and the thickness of the Ge-doped GaN layer 103 were roughly the same as in Example 1. After the stacking of the Ge-doped GaN layer 103, an Si-doped n-type Ino.o6Ge₀.₉₄N clad layer 104 was deposited thereon. The amount of Si doping was 1 x 10¹⁸ cm^"3 and the thickness of the clad layer 104 was 25 nm. Then, the temperature of the substrate 101 was set at 730°C and a light-emitting layer 111 of a 5-period multiple quantum well structure containing a barrier layer of Si- doped GaN and a well layer of undoped In₀.₂₅Gao.₇₅N was disposed on the clad layer 104 of Si-doped n-type Ino.o₆Gao. ₄N. In the light-emitting layer 111 of the multiple quantum well structure, first the GaN barrier layer was disposed as joined to the Si- doped n-type Ino.₀6Ga₀.₉₄N clad layer 104. The GaN barrier layer was grown using triethyl gallium (TEGa) as a gallium source. This layer was formed in a thickness of 16 nm and was doped with Si. The amount of Si doping was set at 5 x 10¹⁷ cm^'3. The In₀.₂5Ga₀.75N well layer was grown using triethyl gallium (TEGa) as a gallium source and trimethyl indium (TMIn) as an indium source. This layer was formed in a thickness of 2.5 nm and was made undoped. Thereafter, a p-type contact layer 107 was stacked by following the procedure of Example 1. The wafer 12 consequently formed was taken out of the reactor. Then, the surface of the highly Ge-doped GaN layer 103 was exposed exclusively in the region planned to seat an n-type ohmic electrode 108 by utilizing the known photolithographic technique and the ordinary dry etching technique. On the exposed surface of the Ge-doped n-type GaN layer 103, an n-type ohmic electrode 108 having titanium (Ti) and gold (Au) stacked on the surface side was formed. On the entire surface of the p-type GaN contact layer 107 constituting the residual surface of the stacked structure 12, a transparent type p-ohmic electrode 109 having platinum (Pt) and gold (Au) stacked sequentially from the surface side and an electrode 110 intended for bonding were formed. Thereafter, an LED chip 20 measuring the square of 350 μm in the plan view was cut from the stacked structure and mounted on a lead frame (not shown) so that a gold wire (not shown) connected to the lead frame might be enabled to advance the device drive current from the lead frame to the LED chip 20. The device drive current was passed in the forward direction between the n-type and p-type ohmic electrodes 108 and 109 via the lead frame. The forward voltage was 2.9 N when the forward current was set at 20 mA. The central wavelength of the blue band light emitted during the passage of the forward current at 20 mA was 460 nm. The intensity of the emission determined by using the ordinary integrating sphere reached 5.5 mW. This fact indicates that a Group III nitride semiconductor LED capable of producing emission of high intensity was obtained in spite of a low drive voltage. Comparative Example 1 : As an n-contact layer, a GaΝ layer 113 doped uniformly with Si at a concentration of 7 x 10¹⁸ cm^"3 was stacked in the place of a Ge-doped GaΝ layer 103 having the Ge concentration periodically varied. Subsequently, an LED was manufactured by forming electrodes under the same conditions as in Example 1 on a stacked structure 13 composed of an undoped InGaΝ clad layer 104, a light-emitting layer 105 of an undoped multiple quantum well structure, a p-type Alo.₀ Gao.93Ν clad layer 106 and a p-type GaN contact layer 107 as illustrated in Fig. 5 under the same conditions as in Example 1, mounting the resultant coated stacked structure on a submount, disposed on a lead frame and bound thereto by wire connection. As a result, the forward voltage was 3.5 V when the forward current was set at 20 mA. The central wavelength of the blue band light emitted during the passage of the forward current at 20 mA was 460 nm As the characteristics during the conduction of the forward current of 20 mA, the intensity of the emission determined by using the ordinary integrating sphere reached 7 mW, an output of emission lower than when a Ge-doped GaN layer 103 was used. Industrial Applicability: The light-emitting device obtained by using a gallium nitride-based compound semiconductor stacked structure according to this invention enjoys a very high utility value from the economic point of view because it shows an ideal output of emission.

Claims

CLAΓMS

1. A Group ffl nitride semiconductor light-emitting device provided on a crystal substrate with an n-type and p-type Group III nitride semiconductor of AlχGaγInzNι- aM_a, wherein O ≤ X ≤ l, O ≤ Y ≤ l, O ≤ Z ≤ l, X + Y + Z = 1, M denotes a Group V element other than N and 0 < a < 1, comprising an n-type electrode contact layer of Group ffl nitride semiconductor including a region doped with Ge and a light-emitting layer including a region undoped or doped with at least one element selected from the group consisting of Si, C, Sn and Pb.

2. A device according to claim 1, wherein the n-type electrode contact layer of Group ffl nitride semiconductor is provided with a Ge atom low concentration layer and a Ge atom-doped Ge atom high concentration layer.

3. A device according to claim 2, wherein the Ge atom low concentration layer is a Ge atom-doped Ge atom low concentration layer or an undoped layer.

4. A device according to any one of claims 1 to 3, wherem the n-type electrode contact layer is configured in a structure having Ge atom low concentration layer and Ge atom high concentration layer alternately stacked periodically.

5. A device according to any one of claims 2 to 4, wherein the Ge atom high concentration layer has a smaller thickness than the Ge atom low concentration layer.

6. A device according to any one of claims 2 to 5, wherein the Ge atom high concentration layer has a surface having concave pits formed therein.

7. A device according to any one of claims 2 to 6, wherein the Ge atom high concentration layer, when formed in a thickness of 10 nm or more, has a surface having pits formed therein in a number in a range of 1 x lOVcm² to 1 x 10¹⁰/cm².

8. A device according to any one of claims 2 to 7, wherein the Ge atom low concentration layer has a surface having flatness of 10 A or less in Ra.

9. A device according to any one of claims 1 to 8, wherein the region doped with Ge has a Ge atom concentration of 1 x 10¹⁷ cm^"3 or more and 1 x 10²⁰ cm^"3 or less.

10. A device according to any one of claims 1 to 9, wherein the light-emitting layer has a multiple quantum well structure.

11. A device according to any one of claims 1 to 10, wherein the region of the light- emitting layer is doped with Si and is a barrier layer of a multiple quantum well structure.

12. A device according to any one of claims 1 to 11, wherein the region of the light- emitting layer is doped with Si and has an Si atom concentration of 5 x 10 cm^" or more and 1 x 10¹⁹ cm^"3 or less.