200812111 九、發明說明 【發明所屬之技術領域】 本發明係關於以發光輸出高，而且驅動電壓低作爲特 徵之II族氮化物半導體發光元件的製造方法。 【先前技術】 ΙΠ族氮化物半導體發光兀件係以將發光層挟持於其間 的形式而配置η型半導體層和p型半導體層之方式構成， 以從各別接觸而設置之負極及正極注入電洞及電子而在半 導體層內的Ρ Ν接合而再結合之情事而得到發光。此發光 的強度係因爲與再結合之電洞和電子的載體數成比例，所 以如流過更多的電流則發光元件的發光輸出係就因爲那樣 而變高。但是，實際上，因爲有由半導體層的體電阻（ bulk resistor )及電極構成材料的阻抗之發熱，所以爲了 提高發光輸出而繼續增加施加電流之情事係不實用。 因此，在評估發光元件的特性上，比較供給了 一定電 流（稱爲額定電流）之情況之發光輸出與此時之順方向電 壓（設此爲驅動電壓V f )之情事爲佳。也就是，在額定電 流I f，以較低的驅動電壓V f而且發光輸出高係作爲發光 元件之良好的特性而成爲目標。 在構成發光元件之電極材料作爲一定之情況，驅動電 壓V f係相依於η型半導體層和ρ型半導體層之構成。特 別是因爲Ρ型半導體層係與電極材料的接觸區域廣大，爲 了下降驅動電壓V f，係將ρ型半導體層干預而產生的阻抗 -4- 200812111 份儘可能低地抑制成爲必要。 例如，可進行：以在p型半導體層與電極材料之間’ 更插入高濃度摻雜不純物之第2p型半導體層（P型接觸層 )’與電極材料之歐姆性變爲良好而變低驅動電壓Vf (例 如：參照日本特開平8-974 1號公報）。但是，驅動電壓 Vf之下降還不充分。 【發明內容】 本發明的目的係鑑於上述的狀況，提供：發光輸出高 、而且驅動電壓低之m族氮化物半導體發光元件。 於第1圖表示m族氮化物半導體發光元件的順方向之 電流-電壓特性圖。在此圖，在額定電流if之驅動電壓vf 係作爲合計了電流-電壓特性之理想特性（圖中的直線 a )與額定電流if之交點之電壓vG、與p型半導體層和電 極材料間的歐姆性質之電壓ve之物而表示。因而，爲了 下降在額定電流之驅動電壓Vf係變低電壓v〇和電壓vc 成爲必要。 電壓係ve係因爲相依於使用的電極材料種類，所以 由該選擇而實質上地決定。爲此，爲了使用廣範圍之電極 材料所以不變低ve、而變低電壓v〇之情事爲必要。如用 電流-電壓特性的圖，在一定電流下變低電壓v〇係繼續變 大爲理想線之直線A的傾斜。 在表示發光元件的理想特性上，經常利用以下的式（ 1 ) 〇 200812111 I = I〇(EXP(qV/n-kT)-l) (1) 此式係表示在第1圖的順方向之電流-電壓特性圖之 直線A之式，直線A的傾斜係成爲式中的q / n · k T。此 η係在此式被稱爲η値，而在變低電壓V 〇係情事係變爲將 變大直線Α的傾斜，也就是變小η値作爲目標。 本發明者係發現：在使m族氮化物半導體的ρ型半導 體層成長時’若以該成長速度成爲8〜20nm/分般的成長 條件進行，則可得結晶性優良的p型ffl族氮化物半導體層 ，上述η値爲2以下般地變得非常小之情事，而完成了本 發明。 也就是，本發明係提供下述之發明。 (1 )於基板上使由ΠΙ族氮化物半導體所構成之，將η 型半導體層、發光層及ρ型半導體層以此順序成長之後， 從於該η型半導體層及ρ型半導體層各別形成負極及正極 之情事所構成之瓜族氮化物半導體發光元件的製造方法； 以該Ρ型半導體層的成長速度爲8〜20nm/分作爲其特徵 之m族氮化物半導體發光元件的製造方法。 (2 )如上述1項所記載的製造方法，在使ρ型半導 體層成長時的成長裝置內壓力爲1〇〜50kPa。 (3 )藉由記載於上述1或2項的製造方法而製造之 ΠΙ族氮化物半導體發光元件。 (4 )於基板上由孤族氮化物半導體所構成之，η型半 導體層、發光層及Ρ型半導體層以此順序層積，正極及負 極各別接於Ρ型半導體層及η型半導體層而設置之發光元 -6 - 200812111 件；以下述式（1 )表示的發光元件的電流—電壓曲線的n 値爲2以下作爲其特徵之瓜族氮化物半導體發光元件。 I = I〇(EXP(qV/n-kT) -1) ( 1 ) (5 )由記載於上述第3或第4項之皿族氮化物半導 體發光元件所構成之燈。 (6 )安裝記載於上述第5項之燈的電子機器。 (7 )安裝記載於上述第6項的電子機器之機械裝置 〇 如藉由將p型半導體層的成長速度控制在8〜20nm/ 分作爲要點之本發明，則可得結晶性優良之p型瓜族氮化 物半導體層，其結果，可得驅動電壓低而且發光輸出高的, 瓜族氮化物半導體發光元件。 【實施方式】 本發明係提供：在p型m族氮化物半導體層，p型半 導體層的成長速度作爲8〜2 Onm/分作爲其特徵之瓜族氮 化物半導體發光元件之製造方法。 於以下敘述發明的詳細的內容。 第2圖爲表示有關本發明之，具備了 p型半導體層之 皿族氮化物半導體發光元件的剖面之模式圖。在本圖，7 爲正極，由光透過性或反射性正極7 a和銲墊層7 b構成。 5爲p型半導體層，此係由p型覆蓋層5 a和p型接觸層 5 b所構成。1爲基板、2爲緩衝層、3爲n型半導體層、4 爲發光層及6爲負極。 -7- 200812111 在本案發明，於基板〗係可無任何限制地使用由：藍 寶石單結晶（A1203 ; A面、C面、Μ面、R面）、尖晶石 單結晶（MgAl204 ) 、ΖηΟ單結晶、LiA102單結晶、 L i G a Ο2單結晶、M g Ο單結晶或G a2 Ο3單結晶等的氧化物 單結晶基板、以及Si單結晶、SiC單結晶、GaAs單結晶 、A1N單結晶、GaN單結晶或是ZrB2等的硼化物單結晶 等的非氧化物單結晶基板選擇之一般周知的基板材料。而 且，基板的面方位係不特別限定，其off angle亦可任意 地選擇。 作爲構成緩衝層、η型半導體層、發光層及p型半導 體層之ffl族氮化物半導體，係以一般式AMriyGaiiyN ( 0 SxSl、〇Sy$l、〇S χ + y ^ 1 )表示之各種組成的半導體 爲一般周知。在構成於本發明之緩衝層、η型半導體層、 發光層及Ρ型半導體層之瓜族氮化物半導體，亦可無任何 限制地使用以一般式 AlxInyGa卜x.yN ( 0 $ X S 1、0 g y $ 1 、〇‘ χ + y s 1 )表示之各種組成的半導體。 作爲成長這些m族氮化物半導體的方法，係有：有機 金屬氣相成長法（M0CVD法）、分子束磊晶成長法（ MBE )、氫化物氣相成長法（HVPE )等。理想爲組成控 制爲容易，具備量產性之Μ 0 CVD法爲適宜，但未必要限 定於同法。 在將M0CVD法作爲上述半導體層的成長方法而採用 的情況，係作爲ΙΠ族的Ga原料，以有機金屬材料之三甲 基鎵（TMGa )或三乙基鎵（TEGa )爲主，同樣地作爲ΙΠ 200812111 族的A1的原料，以三甲基鋁（TMAl )或三乙基鋁（TEAl )爲主而使用。另外關於爲發光層的構成材料之I η係作 爲其原料使用三甲基銦（ΤΜΙ )或三乙基銦（ΤΕΙ )。作爲 V族的Ν源，使用氨（Ν Η 3 )或聯胺（Ν 2 Η 4 )等。 作爲η型半導體層的摻雜物材料，使用s i或G e。作 爲Si原料使用單砂院（SiH4)或二砂院（Si2H6)，作爲 Ge原料使用鍺烷（Germane )或有機鍺化合物。在？型半 導體層，作爲摻雜物使用Mg。作爲此原料，例如使用雙 環戊一稀錶 B i s ( c y c 1 〇 p e η t a d i e n y 1 ) m ag n e s i u m ( C p 2 M g ) 或雙乙基環戊二燃鍾 Bis ( ethylcy clopentadienyl ) magnesium ( (EtCp) 2 M g ) 0 接著，敘述關於作爲成長法採用了 一般之MO CVD法 的各半導體層。 (緩衝層） Φ 作爲緩衝層，係已知開示於日本專利3 0260 8 7號公報 等之低溫緩衝層或開示於日本專利2003 -243 302號公報等 之高溫緩衝層，可無任何限制地使用這些緩衝層。 供給於成長的基板1係可由前期記載中選擇，但在此 ' 係敘述關於使用了藍寶石基板之情況。在將同基板配置於 已設置在溫度及壓力的控制爲可能之反應空間之附Sic膜 的石墨（graphite )製治具（承受器）上之狀態，於此場 所與氫載體氣體（carrier gas )、氮載體氣體、一起送入 NH3氣體和TMA1。附SiC膜石墨（graphite )製治具係藉 200812111 由來自RF線圈之誘導加熱而加熱至必要的溫度，在基板 上係形成A1N緩衝層。作爲溫度，於使A1N的低溫緩衝 成長係控制在從500°C至700 °C的溫度，爲了之後的結晶 化提昇至11〇〇 °C前後的溫度。在使高溫A1N緩衝層成長 的情況，係可不2段的加熱，而可一下子在1 〇 〇 〇 °c至 1 200 °C的溫度下。而且，在使用前述記載中的A1N單結晶 基板、G aN單結晶基板的情況，未必需要使緩衝層成長， 而於上述基板上使後述的η型半導體層直接成長。 (η型半導體層） 作爲η型半導體層係各種組成及構造之物係一般周知 ，在本案發明亦包含這些一般周知之物，亦可使用任何的 組成及構造之物。通常，η型半導體層係由：由未摻雜 GaN層所構成的基材層、含有Si或Ge等的11型摻雜物， 設置負極之η型接觸層以及具有比發光層大的能隙能量之 η型覆蓋（clad )層所構成。η型接觸層係可兼爲η型覆蓋 層及/或基材層。 接著緩衝層的形成之後，於緩衝層上使由未摻雜GaN 層所構成的基材層成長。溫度係作爲1 000〜1 200 °C，將壓 力控制下的原料、NH3氣體和TMGa與載體氣體一起送到 緩衝層上。TMGa的供給量係藉由與同時流動的NH3的比 例而被限制，但作爲成長速度而控制在1 // m /小時〜3 μ m /小時之間係對抑制轉位等的結晶缺陷之產生爲有效。關 於成長壓力係在確保上述的成長速度之上，20〜60kPa ( -10- 200812111 200〜600mbar )的區域爲最合適。 , 接著未摻雜GaN層的成長後，使n型接觸層成長。該 成長條件係與未摻雜GaN層的成長條件相同。摻雜物係與 載體氣體一起供給，但該供給濃度係以與T M G a供給量的 比例進行控制。本發明係藉由將後述的p型半導體層以特 定的成長速度形成，應該變低發光元件的驅動電壓，但驅 動電壓係因爲因η型接觸層的摻雜物濃度亦當然受到影響 ，所以適合ρ型半導體層的成長條件同時決定η型接觸層 的摻雜物濃度。作爲摻雜物的供給條件，係在令M / Ga 比（M = Si或Ge )作爲1 ·0χ1 0·3〜6·0χ1 0·3的範圍，成爲可 變低驅動電壓。 未摻雜GaN層及含有摻雜物的η型半導體層之膜厚， 係個別作爲1〜4 // m爲理想，但未必限定於此範圍。作爲 用以抑制來自基板及緩衝層的結晶缺陷之向上層的傳播之 手段，亦可使未摻雜GaN層及/或含有摻雜物的η型半導 體層之膜厚增加，但藉由厚膜化，因爲誘發晶圓本身的彎 曲所以不太是上策。在本發明，在前述的範圍內設定個別 層的膜厚爲理想。 (發光層） 作爲發光層係各種組成及構造之物係一般周知，在本 案發明亦包含這些一般周知之物，亦可使用任何的組成及 構造之物。 例如多重量子井構造的發光層係一邊使成爲阻障層之 -11 - 200812111 η型GaN層與成爲井層之GalnN層交互地層積、一 。載體氣體係選擇使用N2或H2。NH3和TEGa或 係與此載體氣體一起供給。[Technical Field] The present invention relates to a method of manufacturing a Group II nitride semiconductor light-emitting device characterized by high light-emitting output and low driving voltage. [Prior Art] A bismuth nitride semiconductor light-emitting device is configured by disposing an n-type semiconductor layer and a p-type semiconductor layer in a form in which a light-emitting layer is held therebetween, and injecting electricity from a negative electrode and a positive electrode provided separately from each other. The holes and the electrons are combined in the semiconductor layer to be combined with each other to obtain light. The intensity of this luminescence is proportional to the number of recombined holes and electron carriers, so that if more current flows, the illuminating output of the illuminating element becomes higher. However, in actuality, since heat is generated by the bulk resistor of the semiconductor layer and the impedance of the electrode constituent material, it is not practical to continue to increase the applied current in order to increase the light-emitting output. Therefore, in evaluating the characteristics of the light-emitting element, it is preferable to compare the light-emitting output in the case where a constant current (referred to as a rated current) is supplied and the forward-direction voltage at this time (this is the driving voltage V f ). That is, the rated current I f is targeted at a low driving voltage V f and a high luminous output as a good characteristic of the light-emitting element. In the case where the electrode material constituting the light-emitting element is constant, the driving voltage Vf depends on the configuration of the n-type semiconductor layer and the p-type semiconductor layer. In particular, since the contact area of the Ρ-type semiconductor layer and the electrode material is large, in order to lower the driving voltage V f , it is necessary to suppress the impedance -4- 200812111 which is generated by the p-type semiconductor layer interference as low as possible. For example, it is possible to perform a lowering of the ohmic property of the second p-type semiconductor layer (P-type contact layer) with a higher concentration of doped impurities between the p-type semiconductor layer and the electrode material. The voltage Vf (for example, refer to Japanese Laid-Open Patent Publication No. Hei 8-974 No. 1). However, the drop in the driving voltage Vf is not sufficient. SUMMARY OF THE INVENTION An object of the present invention is to provide a group-m nitride semiconductor light-emitting device having a high light-emitting output and a low driving voltage in view of the above circumstances. Fig. 1 is a graph showing the current-voltage characteristics in the forward direction of the group m nitride semiconductor light-emitting device. In this figure, the driving voltage vf at the rated current if is a voltage vG which sums the ideal characteristic of the current-voltage characteristic (the straight line a in the figure) and the rated current if, and the p-type semiconductor layer and the electrode material. It is expressed by the ohmic nature of the voltage ve. Therefore, it is necessary to lower the voltage v 〇 and the voltage vc in order to lower the driving voltage Vf at the rated current. Since the voltage system ve depends on the type of electrode material used, it is substantially determined by this selection. For this reason, in order to use a wide range of electrode materials, it is necessary to keep the low ve and the low voltage v〇. If the current-voltage characteristic is used, the voltage becomes lower at a constant current and the slope of the line A continues to become the ideal line. In expressing the ideal characteristics of the light-emitting element, the following formula (1) is often used. 〇200812111 I = I〇(EXP(qV/n-kT)-l) (1) This equation is shown in the forward direction of Fig. 1 In the equation of the line A of the current-voltage characteristic diagram, the inclination of the straight line A becomes q / n · k T in the equation. This η is called η値 in this equation, and becomes a target when the voltage becomes lower and the V 〇 system becomes the inclination of the straight line ,, that is, becomes smaller η値. The present inventors have found that when the p-type semiconductor layer of the m-type nitride semiconductor is grown, when the growth rate is 8 to 20 nm/min, the p-type ffl nitrogen having excellent crystallinity can be obtained. The present invention has been completed in the case where the above η 値 is extremely small as 2 or less. That is, the present invention provides the following invention. (1) A yttrium-based nitride semiconductor is formed on a substrate, and an n-type semiconductor layer, a light-emitting layer, and a p-type semiconductor layer are grown in this order, and then the n-type semiconductor layer and the p-type semiconductor layer are separately formed. A method for producing a cuban nitride semiconductor light-emitting device comprising the negative electrode and the positive electrode; and a method for producing a group m nitride semiconductor light-emitting device characterized by a growth rate of the bismuth semiconductor layer of 8 to 20 nm/min. (2) The production method according to the above item 1, wherein the pressure in the growth apparatus when the p-type semiconductor layer is grown is 1 Torr to 50 kPa. (3) A bismuth nitride semiconductor light-emitting device produced by the above-described one or two production methods. (4) The substrate is composed of an orphan nitride semiconductor, and the n-type semiconductor layer, the light-emitting layer, and the germanium-type semiconductor layer are laminated in this order, and the positive electrode and the negative electrode are respectively connected to the germanium-type semiconductor layer and the n-type semiconductor layer. Further, the illuminating element -6 - 200812111 is provided; and the quaternary nitride semiconductor light-emitting element having the n 値 of the current-voltage curve of the light-emitting element represented by the following formula (1) is 2 or less. I = I 〇 (EXP (qV / n - kT) - 1) (1) (5) A lamp comprising the dish nitride semiconductor light-emitting device described in the above item 3 or 4. (6) An electronic device in which the lamp of the above item 5 is mounted. (7) The mechanical device of the electronic device described in the above-mentioned item 6, for example, by controlling the growth rate of the p-type semiconductor layer to 8 to 20 nm/min, the p-type excellent in crystallinity is obtained. As a result of the melon nitride semiconductor layer, a quaternary nitride semiconductor light-emitting device having a low driving voltage and a high light-emitting output can be obtained. [Embodiment] The present invention provides a method for producing a cuban nitride semiconductor light-emitting device characterized by a growth rate of a p-type semiconductor layer and a p-type semiconductor layer of 8 to 2 Onm/min. The details of the invention are described below. Fig. 2 is a schematic view showing a cross section of a dish nitride semiconductor light-emitting device having a p-type semiconductor layer according to the present invention. In the figure, 7 is a positive electrode, and is composed of a light transmissive or reflective positive electrode 7a and a pad layer 7b. 5 is a p-type semiconductor layer composed of a p-type cladding layer 5 a and a p-type contact layer 5 b. 1 is a substrate, 2 is a buffer layer, 3 is an n-type semiconductor layer, 4 is a light-emitting layer, and 6 is a negative electrode. -7- 200812111 In the present invention, the substrate can be used without any restrictions: sapphire single crystal (A1203; A surface, C surface, Μ surface, R surface), spinel single crystal (MgAl204), ΖηΟ An oxide single crystal substrate such as crystal, LiA102 single crystal, L i G a Ο 2 single crystal, Mg Ο single crystal or Ga 2 Ο 3 single crystal, and Si single crystal, SiC single crystal, GaAs single crystal, A1N single crystal, A generally known substrate material selected from a non-oxide single crystal substrate such as a GaN single crystal or a boride single crystal such as ZrB2. Further, the plane orientation of the substrate is not particularly limited, and the off angle thereof can be arbitrarily selected. The ffl-type nitride semiconductor constituting the buffer layer, the n-type semiconductor layer, the light-emitting layer, and the p-type semiconductor layer is represented by a general formula of AMriyGaiiyN (0 SxSl, 〇Sy$l, 〇S χ + y ^ 1 ). Semiconductors are generally known. The cassant nitride semiconductor constituting the buffer layer, the n-type semiconductor layer, the luminescent layer, and the bismuth-type semiconductor layer of the present invention can also be used without any limitation in the general formula AlxInyGabx.yN (0$XS1, 0) Gy $ 1 , 〇 ' χ + ys 1 ) represent semiconductors of various compositions. Examples of the method of growing these group-m nitride semiconductors include an organometallic vapor phase growth method (M0CVD method), a molecular beam epitaxy growth method (MBE), and a hydride vapor phase growth method (HVPE). Ideally, it is easy to control the composition and has mass production. The CVD method is suitable, but it is not necessarily limited to the same method. In the case where the M0CVD method is used as the growth method of the semiconductor layer, it is mainly a Ga raw material of the lanthanum, and an organometallic material such as trimethylgallium (TMGa) or triethylgallium (TEGa) is used as the main material. ΙΠ The raw material of A1 of the 200812111 family is mainly composed of trimethylaluminum (TMAl) or triethylaluminum (TEAl). Further, as the raw material of the constituent material of the light-emitting layer, trimethylindium (ruthenium) or triethylindium (ruthenium) is used as a raw material. As the source of the V group, ammonia (Ν Η 3 ) or hydrazine (Ν 2 Η 4 ) or the like is used. As the dopant material of the n-type semiconductor layer, s i or G e is used. As a Si raw material, a single sand yard (SiH4) or a two sand yard (Si2H6) is used, and as a Ge raw material, a germane (Germane) or an organic germanium compound is used. in? A type of semiconductor layer, Mg is used as a dopant. As such a raw material, for example, a dicyclopentanthene table B is ( cyc 1 〇pe η tadieny 1 ) m ag nesium ( C p 2 M g ) or a diethylcyclopentanthene quinone Bis (ethylcy clopentadienyl ) magnesium ( (EtCp) 2 M g ) 0 Next, each semiconductor layer which is a general growth method using MO CVD method will be described. (buffer layer) Φ is a low-temperature buffer layer which is disclosed in Japanese Patent No. 3,260,807, or a high-temperature buffer layer which is disclosed in Japanese Patent Laid-Open Publication No. 2003-243302, and can be used without any limitation. These buffer layers. The substrate 1 to be grown can be selected from the previous description, but the case where the sapphire substrate is used is described here. In the state where the same substrate is placed on a graphite fixture (sustainer) provided with a Sic film which is set in a reaction space where temperature and pressure are controlled, the site and the hydrogen carrier gas (carrier gas) Nitrogen carrier gas, together with NH3 gas and TMA1. A SiC-coated graphite fixture was used to heat the necessary temperature from the RF coil to the necessary temperature to form an A1N buffer layer on the substrate. As the temperature, the low-temperature buffer growth of A1N was controlled at a temperature of from 500 ° C to 700 ° C, and the temperature was raised to a temperature before and after 11 ° C for the subsequent crystallization. In the case where the high-temperature A1N buffer layer is grown, it may be heated in two stages, and may be at a temperature of 1 〇 〇 〇 °c to 1 200 °C at a time. Further, when the A1N single crystal substrate or the GaN single crystal substrate described above is used, it is not necessary to grow the buffer layer, and the n-type semiconductor layer to be described later is directly grown on the substrate. (n-type semiconductor layer) As the n-type semiconductor layer, various compositions and structures are generally known, and the present invention also includes these generally known materials, and any of the compositions and structures may be used. Generally, the n-type semiconductor layer is composed of a base material layer composed of an undoped GaN layer, a type 11 dopant containing Si or Ge, etc., an n-type contact layer provided with a negative electrode, and a larger energy gap than the light-emitting layer. The n-type cladding (clad) layer of energy is formed. The n-type contact layer may also serve as an n-type cladding layer and/or a substrate layer. After the formation of the buffer layer, the substrate layer composed of the undoped GaN layer is grown on the buffer layer. The temperature is taken as 1 000 to 1 200 ° C, and the raw material under pressure control, NH 3 gas and TMGa are supplied to the buffer layer together with the carrier gas. The supply amount of TMGa is limited by the ratio of NH3 flowing at the same time, but it is controlled at a growth rate of 1 // m / hr to 3 μ m / hr to suppress the occurrence of crystal defects such as indexing. To be effective. The growth pressure is the most suitable for ensuring the above growth rate, and the area of 20 to 60 kPa (-10-200812111 200 to 600 mbar) is most suitable. Then, after the growth of the undoped GaN layer, the n-type contact layer is grown. The growth conditions are the same as those for the undoped GaN layer. The dopant system is supplied together with the carrier gas, but the supply concentration is controlled in proportion to the supply amount of T M G a . In the present invention, the p-type semiconductor layer to be described later is formed at a specific growth rate, and the driving voltage of the light-emitting element should be lowered. However, since the driving voltage is naturally affected by the dopant concentration of the n-type contact layer, it is suitable. The growth conditions of the p-type semiconductor layer simultaneously determine the dopant concentration of the n-type contact layer. The supply condition of the dopant is such that the M / Ga ratio (M = Si or Ge) is in the range of 1·0χ1 0·3 to 6·0χ1 0·3, and becomes a variable driving voltage. The film thickness of the undoped GaN layer and the n-type semiconductor layer containing the dopant is preferably 1 to 4 // m, but is not necessarily limited to this range. As a means for suppressing the propagation of the upper layer from the crystal defects of the substrate and the buffer layer, the film thickness of the undoped GaN layer and/or the n-type semiconductor layer containing the dopant may be increased, but by thick film Because it induces the bending of the wafer itself, it is not the best policy. In the present invention, it is preferred to set the film thickness of the individual layers within the above range. (Light Emitting Layer) As the light emitting layer, various compositions and structures are generally known, and the present invention also includes such generally known materials, and any composition and structure can be used. For example, the light-emitting layer of the multiple quantum well structure is formed by alternately layering the -11 - 200812111 η-type GaN layer which becomes the barrier layer with the GalnN layer which becomes the well layer. The carrier gas system is selected to use N2 or H2. NH3 and TEGa are supplied together with this carrier gas.
GalnN層的成長係再加上供給TMI。總之取一 成長時間，一邊斷續地供給In之程序。在GalnN 長係因爲於載體氣體中存在H2而In濃度的控制變 ，在此層’作爲載體氣體使用H2不是上策。阻障層 GaN層）和井層（GalnN層）之膜厚係選擇發光輸 得最高之條件。在決定了最適膜厚之上，適宜選擇 原料供給量和成長時間。向阻障層之摻雜物量亦成 發光元件的驅動電壓之高低的條件，而該濃度係對 型半導體層的成長條件而選擇。作爲摻雜物爲Si或 哪一個都可以。 成長溫度係在7 0 0 °C至1 0 0 0 °C之間爲理想，但 定於此範圍。但是，在井層的成長係在高的溫度爲 以將ϊη帶進成長膜中，實質上形成井層爲困難。 成長溫度係在不太變高的範圍內選擇。在本發明係 層的成長溫度作爲70(TC至1 000°C的範圍，但即使 障層和井層的成長溫度亦不是障礙。成長壓力係一 成長速度的平衡、一邊設定。在本發明係成長β 10kPa ( 200 mbar )至 60kPa ( 600 mba〇 之間爲理 未必限定於此範圍。 爲井層和阻障層的數，哪一方都是3層至7層 ，但未必限定於此範圍。發光層係最後使阻障層成 邊形成 TMGa 邊控制 層的成 得困難 (η型 出爲變 m族的 爲左右 應於P 丨Ge的 未必限 變得難 於是， 就發光 改變阻 邊取與 壓力係 想，但 爲合適 長成爲 -12- 200812111 結束（最終阻障層）。此阻障層係防止由井層的載體之溢 流、同時在繼續的P型半導體層的成長，完成防止來自最 終阻障層之In的脫離的任務。 (P型半導體層） P型半導體層亦各種組成及構造之物係一般周知，在 本案發明亦包含這些一般周知之物，亦可使用任何的組成 及構造之物。p型半導體層，通常由：於其上形成正極之 P型接觸層和能隙能量比發光層大的p型覆蓋層形成。P 型接觸層亦可兼爲p型覆蓋層。 在本發明，成長速度被控制在8〜20nm/分之p型半 導體層，係於其上形成正極之p型接觸層。p型覆蓋層的 成長速度未必需要在此範圍。 控制了成長速度的P型接觸層之Mg摻雜物的濃度係 不特別限制，但爲了確保良好的結晶性，係Mg摻雜物的 濃度爲0.9x1 02G〜2x1 02G原子/ cm3爲理想。於p型接觸層 中，係與Mg摻雜物一起，氫原子爲ΙχΙΟ18〜1x1 〇21原子/ cm3左右的濃度存在著亦佳。 在P型半導體層的成長，首先直接接於發光層的最終 阻障層上而將p型覆蓋層，於其上使成長速度被控制在 8〜2 0nm/分之p型接觸層層積。於p型覆蓋層係使用GaN 或GaAIN爲理想。另外’ p型覆蓋層係使組成或晶格常數 相異之層交互地層積亦佳，使層的厚度和爲摻雜物之M g 的濃度變化亦佳。 -13- 200812111 於有關本發明之成長速度控制在8〜20nm/分之p型 接觸層係使用GaAIN爲理想。其成長係如以下般地進行。 將TMGa、TMA1及爲摻雜物之Cp2Mg、與載體氣體（氫 或氮、或是兩者的混合氣體）及NH3氣體一起送到上述的 P型覆蓋層上。 此時的成長溫度係980〜11 00°C的範圍爲最佳。若爲比 98 0 °C低的溫度，則形成結晶性低的磊晶（epitaxial )層， 因結晶缺陷起因而招致驅動電壓的上昇。另外在比1 10 01 高的溫度，係在位於下層之發光層之中，井層在p型半導 體層成長過程被置於高溫度的環境下，有受到熱損傷的可 能性。此情況係在作爲發光元件的時點之強度下降，或是 有帶來在耐性試驗下之強度劣化的危險。 關於成長壓力，無特別限制，但理想爲 50kPa ( 5 0 0mbar )以下爲佳。作爲此理由，係因爲若在此壓力下 進行成長，則可將p型半導體層中的面內方向之A1濃度 變得均勻，在讓按照必要而使GaAIN的A1組成變化之p 型半導體層成長的情況，控制爲容易。在比此壓力高的條 件，係已供給之TMA1與NH3的反應變得顯著，在到達於 成長途中之基板前，TMA1被消耗，變得難以得到作爲目 的之A1組成。關於作爲摻雜物而送入之Mg亦可謂同樣之 情事。也就是，若爲5 0 k P a ( 5 0 0 m b a r )以下的成長條件 ’則P型半導體層中的2次元方向（成長基板的面內方向 )之Mg濃度分布變爲均勻（成長基板的面內均勻性）。 亦已知：藉由使用之載體氣體流量而GaAIN接觸層中 -14- 200812111 的面內方向之A1組成、Mg濃度之分布係進行變化。但是 5發現：比起載體氣體條件，由成長壓力的條件而接觸層 中的A1組成、Mg的面內均勻性被大幅地左右。因而作爲 在 50kPa ( 500 mbar)以下 10kpa ( 1〇〇 mbar)以上的成長 壓力爲合適。 p型半導體層的成長速度係主要相依於爲原料之 TMGa之供給量。如每單位時間的TMGa供給量變多，則 可於那些短時間內得到目的之膜厚。如摻雜物原料亦同樣 地變多供給量則爲佳。但是，變快了成長速度之p型半導 體層係容易導入結晶缺陷，即使以必要濃度含有摻雜物， 起因於結晶缺陷而施體能階（donor level )增加，結果上 驅動電壓V f變得不下降。反之，若過度抑制成長速度， 則至達到目的膜厚之成長時間變長，有變大在成長期間中 之朝向發光層之熱損傷的疑慮。本發明者係進行：硏討每 單位時間之TMGa的供給量，將p型半導體的成長速度限 定於某範圍，而且改變符合於此之摻雜物的供給量條件。 然後，了解：作爲滿足驅動電壓Vf變低及無向發光層的 熱損傷之 P型接觸層的成長條件，係其成長速度爲 8〜2 0 n m /分爲理想。 而且，成長速度的決定，係藉由晶圓剖面之電子顯微 鏡（TEM )觀察而計測p型接觸層的膜厚，以成長時間反 除而求出。因而，作成改變了幾個條件之觀察用樣本，以 事前先調查成長速度與每單位時間的TMGa的供給量之關 係而可事先決定成爲必要之成長條件。 -15- 200812111 P型接觸層的成長之後，與停止基板加熱一起送入n2 氣體，將反應空間內進行潔淨、同時冷卻晶個，冷卻至可 取出至成長裝置外之前。而且，在本方法係確認在此時點 ，p型半導體層變成作爲目的之P型。因而，在此之後用 ^ 以活性化的熱處理係不必要。 、. 接著，說明關於設置於η型接觸層及p型接觸層上之 負極及正極。 (負極） 作爲負極係各種組成及構造之物係一般周知，在本案 發明亦包含這些一般周知之物，亦可使用任何的組成及構 造之物。該製造方法亦各種製法爲一般周知，可使用那些 一般周知的方法。 於向η型接觸層上的負極形成面的製作係可利用一般 周知的光蝕刻技術及一般的鈾刻技術。藉由這些技術，從 ® 從晶圓的最上層挖至η型接觸層的位置，可使預定形成負 極的區域之η型接觸層露出。作爲負極材料，係除了作爲 接於η型接觸層的接觸金屬之Al、Ti、Ni、Au之外，可 利用Cr、W、V等的金屬材料。爲了使向^型接觸層的密 ^ 接性提高’將接觸金屬作爲由上述金屬複數選擇之多層構 造亦佳。而且，最表面若爲Au則接合性變爲良好。 (正極） 被設置於p型接觸層上的正極，亦可知各種組成及構 -16- 200812111 造之光透過性或反射性正極，在本案發明亦包含這 周知之物，亦可使用任何的組成及構造之物。該製 亦各種製法爲一般周知，可使用那些一般周知的方、) 在本發明係，在將光透過性的ΙΤΟ使用於正極 情況，因爲驅動電壓的下降效果大所以爲理想。作 的組成係作爲50% S Ιη<1 00%及〇%<Sn $ 50%爲理 此範圍內可滿足低的膜阻抗和高的光透過率。I η爲 Sn爲1 0%係特別理想。於ΙΤΟ係作爲不純物而含 、皿族、IV族或V族的元素亦佳。另外，代替 ITO 使用SnO、ZnO或InO等。 ΙΤΟ膜的膜厚係50〜500nm爲最佳。如爲50nm 則ΙΤΟ膜本身的膜阻抗變高，驅動電壓變高。另外 比5 0 0nm厚則向上面的發光之取出效率變低，發光 變高。 關於ITO膜的成膜方法，係可使用一般周知的 鍍法或濺鍍法。於真空蒸鍍係於加熱方法有阻抗加 或電子束加熱方式等，但於金屬以外的材料之蒸鍍 子束加熱方式爲合適。另外，亦可使用：將成爲原 合物作爲液狀，將此塗佈於表面之後藉由適當處理 氧化物膜的方法。 在蒸鍍法係由條件而ITO膜的結晶性受到影響 適切地選擇條件則無該限制。而且在室溫作成ITO 況，係用以透明化的熱處理成爲必要。 在藉由濺鍍法的製膜，係因爲被置於電漿的高 些一般 造方法 去。 材料的 Μ ITO 想。在 90%、 有Π族 ，亦可 以下， 反之若 輸出不 真空蒸 熱方式 ，係電 料之化 而作爲 ，但如 膜的情 會 b i 51 -17- 200812111 境下，所以P型接觸層表面容易受到因電漿之損傷，因而 在接觸阻抗變高的傾向，但在設法於製膜條件而可變小向 P型接觸層表面之影響。 通常，在形成了光透過性或反射性正極之後，於該一 部分表面製作構成銲墊部之銲墊層。與銲墊層合倂而構成 正極。作爲銲墊層的材料，已知各種構造之物，在本發明 ，亦可不特別限制這些一般周知之物而使用。在使用於負 極材料之Al、Ti、Ni、Au之外，Cr、W、V亦可無任何限 制而使用。然而，運用與已使用之光透過性或反射性正極 之密接性良好的材料爲最佳。厚度係有：如對於接合時的 應力而不向光透過性或反射性正極帶來損傷般之充分地變 厚之必要另外，最表層係與接合球（bonding ball)之密 接性優良的材料，例如作爲Αιι爲最佳。 本發明的瓜族氮化物半導體發光元件，係例如可藉由 該業界周知的手段而設置透明蓋體而作爲燈。另外，亦可 組合具有本發明的氮化鎵系化合物半導體發光元件和螢光 體之蓋體而製作白色的燈。 另外，由本發明的m族氮化物半導體發光元件而製作 的燈係因爲驅動電壓低、發光輸出高，所以安裝藉由此技 術而製作的燈之攜帶式電話、顯示器、面板類等的電子機 器、或安裝了該電子機器之汽車、電腦、遊戲機等的機械 裝置類，係成爲可以低電力驅動，可實現高的特性。特別 是，在攜帶式電話、遊戲機、玩具、汽車零件等之由電池 驅動之機器類，發揮省電力的效果。 -18- 200812111 實施例 以下藉由實施例及比較例而詳細地說明本發明，但本 發明並非僅被限定於這些實施例。 (實施例1 ) 在反應爐的承受器上設置藍寶石基板，將爐內壓力控 制在20kPa ( 2 00 mbar )、基板溫度控制在1 1 〇〇 °c，將 TMA1和NH3與H2載體氣體一起送到基板上，形成A1N 緩衝層。此成長時間作爲1 〇分。 之後，將壓力作爲 40kPa ( 400 mbar )、溫度作爲 1 03 0 °C而供給TMGa和NH3而於A1N緩衝層上，3小時成 長由未摻雜GaN所構成之基材層。接著，一邊維持壓力和 溫度、一邊作爲η型摻雜物而供給四甲基鍺，1小時成長 η型GaN層。四甲基鍺的供給量係以η型GaN層的載體濃 度成爲4.0x1 018cnT3的方式調整。由此而形成n型接觸層 〇 之後，壓力係作爲 40kpa ( 400 mbar )、溫度作爲 7 5 0°C而將載體氣體從H2轉換至N2，一邊供給TEGa和 NH3、作爲摻雜物供給SiH4、一邊將阻障層成長7分鐘， 之後更供給TMIn而使井層5分鐘成長。交互地5次反覆 此阻障層和井層的成長，於最後成長最終的阻障層而作爲 發光層。摻雜物的SiH4的量係以阻障層及井層中的Si濃 度成爲2.0x1 017原子/ cm3的方式調整。 •19- 200812111 之後，將壓力作爲 20kPa ( 200 mbar )、溫度作爲 l〇〇〇°C而再次將載體氣體轉換爲H2，供給TMGa和TMA1 ’作爲摻雜物而送入Cp2Mg而3分鐘成長由Gao^AlusN 所構成之P型覆蓋層。 之後，一邊維持壓力和溫度、一邊進行由 GaonAUnN所構成的p型接觸層之成長。作爲此條件以 成長速度成爲8nm/分的方式固定TMGa供給量。成長時 間作爲15分。p型接觸層中的Mg濃度係以成爲1x1 Ο”原 子/ cm3的方式，調整Cp2Mg的供給量。 之後，停止向誘導線圈的電力投入，停止加熱而將載 體氣體轉換至N2，與潔淨爐內一起，將得到的m族氮化 物半導體層積物冷卻至取出到爐外的溫度。 使取出至爐外的m族氮化物半導體層積物之η型接觸 層的一部分藉由光蝕刻和乾蝕刻而露出，於其上製作了由 c r和T i的金屬層所構成的負極。另外，於Ρ型接觸層上 係藉由蒸鍍法而製作厚度3 5 0nm的ITO 1(In:Sn = 9:l )，於其上製作以Ti、An、A1及Au之順序層積的銲墊層 ，作爲正極。之後，在進行硏磨基板裏面和切割之後，分 割爲各發光元件。 於得到的發光元件流過20mA的電流而使其發光、同 時測定驅動電壓Vf和發光輸出，結果爲3.2V和9mW。另 外，將電流-電壓特性曲線的測定結果表示於第3圖。由 此圖使用上述（1 )式而求出的η値爲1 .8。而且，將這些 的結果與其他的實施例及比較例匯集而表示於下述的表1 -20· 200812111 (實施例2 ) 將P型接觸層的成長條件，除了將成長速度作爲 2 Onm /分、成長時間作爲6分以外，與實施例1同樣地製 作瓜族氮化物半導體發光元件。 將得到的發光元件與實施例〗進行相同的評估，結果 驅動電壓爲3.2V、發光輸出爲9mW。另外，將電流-電壓 特性曲線的測定結果表示於第4圖。由此圖使用上述（1 )式而求出的η値爲2.0。 (比較例1 ) 將Ρ型接觸層的成長條件，除了將成長速度作爲7nm /分、成長時間作爲1 7.2分以外，與實施例1同樣地製 作瓜族氮化物半導體發光元件。 • 將得到的發光元件與實施例1進行相同的評估，結果 驅動電壓爲3.3V、發光輸出爲8mW。另外，將電流-電壓 特性曲線的測定結果表示於第5圖。由此圖使用上述（1 )式而求出的η値爲2.6。 (比較例2 ) 將ρ型接觸層的成長條件，除了將成長速度作爲 2 1 nm /分、成長時間作爲5.7分以外，與實施例5同樣地 製作m族氮化物半導體發光元件。 -21 - 200812111 將得到的發光元件與實施例1進行相同的評估，結果 驅動電壓爲3.4V、發光輸出爲8m W。另外，將電流-電壓 特性曲線的測定結果表示於第6圖。由此圖使用上述（1 )式而求出的η値爲3.1。 表1 接觸層的成長速度V (單位nm/分） 成長時間 分 驅動電壓Vf (施加20mA) V 發光輸出 mW(20mA) η値 實施例1 8 15 3.2 9 1.8 實施例2 20 6 3.2 9 2 比較例1 7 17.2 3.3 8 2.6 比較例2 21 5.7 3.4 8 3.1 產業上的可利用性 本發明的m族氮化物半導體發光元件係因爲具有良好的發 光輸出、驅動電壓下降，所以其產業上的利用價値非常大 【圖式簡單說明】 第1圖爲說明發光元件的順方向的電流-電壓特性之 圖。 ^ 第2圖爲表示本發明的瓜族氮化物半導體發光元件之 剖面之模式圖。 第3圖爲表示在實施例1可得到之發光元件的電流一 電壓特性曲線的測定結果之圖。 第4圖爲表示在實施例2可得到之發光元件的電流一 -22- 200812111 電壓特性曲線的測定結果之圖。 第5圖爲表示在比較例1可得到之發光元件的電流一 電壓特性曲線的測定結果之圖。 第6圖爲表示在比較例2可得到之發光元件的電流-•電壓特性曲線的測定結果之圖。 【主要元件符號說明】 _ 1 :基板 2 :緩衝層 3 : η型半導體層 4 :發光層 5b : ρ型接觸層 5a : p型覆蓋層 5 : ρ型半導體層 6 :負極 • h :光透過性或反射性正極 7b :銲墊層 7 :正極 -23-The growth of the GalnN layer is coupled to the supply of TMI. In short, take a long time and intermittently supply the program of In. In the GalnN long system, the control of the In concentration is caused by the presence of H2 in the carrier gas, and it is not preferable to use H2 as the carrier gas in this layer. The film thickness of the barrier layer GaN layer and the well layer (GalnN layer) selects the highest luminescence output condition. Above the optimum film thickness, it is appropriate to select the raw material supply amount and growth time. The amount of the dopant to the barrier layer also becomes a condition for the driving voltage of the light-emitting element, and the concentration is selected for the growth condition of the semiconductor layer. As the dopant, Si or whichever is possible. The growth temperature is ideal between 700 ° C and 1 0 0 ° C, but is within this range. However, it is difficult to form a well layer in the growth of the well layer at a high temperature in order to bring ϊη into the growth film. The growth temperature is selected within a range that is not too high. The growth temperature of the layer of the present invention is in the range of 70 (TC to 1 000 ° C, but the growth temperature of the barrier layer and the well layer is not an obstacle. The growth pressure is set at the balance of the growth rate. The growth of β 10 kPa (200 mbar ) to 60 kPa (600 mba 为 is not necessarily limited to this range. The number of the well layer and the barrier layer is 3 to 7 layers, but is not necessarily limited to this range. The light-emitting layer finally makes it difficult to form the barrier layer to form the TMGa edge control layer (the n-type is changed to the m-group, and the right and left is not necessarily limited to the P 丨 Ge, and the luminescence changes the barrier edge. The pressure system is thought, but it is suitable for the end of -12-200812111 (final barrier layer). This barrier layer prevents the overflow of the carrier by the well layer, while continuing the growth of the P-type semiconductor layer, completes the prevention from the final The task of detachment of In of the barrier layer (P-type semiconductor layer) The P-type semiconductor layer is also generally known in various compositions and structures, and the present invention also includes these generally known objects, and any composition and structure may be used. Object.p The semiconductor layer is generally formed of a P-type contact layer on which a positive electrode is formed and a p-type cap layer having a larger energy gap than the light-emitting layer. The P-type contact layer may also serve as a p-type cap layer. In the present invention, the growth rate The p-type semiconductor layer controlled at 8 to 20 nm/min is a p-type contact layer on which a positive electrode is formed. The growth rate of the p-type cladding layer is not necessarily required to be in this range. Mg of the P-type contact layer which controls the growth rate The concentration of the dopant is not particularly limited, but in order to ensure good crystallinity, the concentration of the Mg dopant is 0.9 x 1 02 G 2 2 x 1 02 G atoms / cm 3 . In the p type contact layer, the doping with Mg is preferable. The concentration of the hydrogen atom is preferably about 18 to 1 x 1 〇 21 atoms / cm 3 . In the growth of the P - type semiconductor layer, firstly, the p-type cladding layer is directly attached to the final barrier layer of the light-emitting layer. The p-type contact layer is grown at a growth rate of 8 to 20 nm/min. It is preferable to use GaN or GaAIN for the p-type cap layer, and the p-type cap layer makes the composition or lattice constant different. The layer is also alternately layered, so that the thickness of the layer is a dopant The change in the concentration of M g is also good. -13- 200812111 It is preferable to use GaAIN for the p-type contact layer in which the growth rate of the present invention is controlled at 8 to 20 nm/min. The growth is carried out as follows: TMGa, TMA1 And the Cp2Mg of the dopant, and the carrier gas (hydrogen or nitrogen, or a mixed gas of the two) and the NH3 gas are sent to the P-type coating layer. The growth temperature at this time is 980 to 11 00 ° C. When the temperature is lower than 98 ° C, an epitaxial layer having low crystallinity is formed, which causes an increase in driving voltage due to crystal defects. In addition, at a temperature higher than 1 10 01, in the lower layer of the light-emitting layer, the well layer is exposed to a high temperature environment during the growth of the p-type semiconductor layer, and there is a possibility of thermal damage. In this case, the strength at the time of the light-emitting element is lowered, or there is a risk that the strength under the resistance test is deteriorated. There is no particular limitation on the growth pressure, but it is preferably 50 kPa (500 mbar) or less. For this reason, when the growth is performed under this pressure, the A1 concentration in the in-plane direction in the p-type semiconductor layer can be made uniform, and the p-type semiconductor layer in which the A1 composition of GaAIN is changed as necessary can be grown. In the case of control, it is easy. When the pressure is higher than this, the reaction between the supplied TMA1 and NH3 becomes remarkable, and before reaching the substrate in the middle of growth, TMA1 is consumed, and it becomes difficult to obtain the desired A1 composition. The same can be said about Mg fed as a dopant. In other words, in the case of a growth condition of 5 0 k P a (500 mbar) or less, the Mg concentration distribution in the second-order direction (in-plane direction of the growth substrate) in the P-type semiconductor layer becomes uniform (growth of the substrate) In-plane uniformity). It is also known that the distribution of the A1 composition and the Mg concentration in the in-plane direction of the -14-200812111 in the GaAIN contact layer is changed by the flow rate of the carrier gas used. However, it was found that the in-plane uniformity of the A1 composition and the Mg in the contact layer were greatly affected by the conditions of the growth pressure as compared with the carrier gas conditions. Therefore, it is suitable as a growth pressure of 10 kPa (1 mbar) or more at 50 kPa (500 mbar) or less. The growth rate of the p-type semiconductor layer mainly depends on the supply amount of TMGa as a raw material. If the amount of TMGa supplied per unit time is increased, the film thickness of the object can be obtained in a short time. It is preferable that the dopant raw material is also supplied in the same amount. However, the p-type semiconductor layer which has a faster growth rate is likely to introduce crystal defects, and even if the dopant is contained at a necessary concentration, the donor level is increased due to the crystal defect, and as a result, the upper driving voltage V f becomes not decline. On the other hand, if the growth rate is excessively suppressed, the growth time to the target film thickness becomes long, and there is a fear that the thermal damage to the light-emitting layer during the growth period is increased. The inventors of the present invention conducted a review of the supply amount of TMGa per unit time, limited the growth rate of the p-type semiconductor to a certain range, and changed the supply amount condition of the dopant in accordance with this. Then, it is understood that the growth condition of the P-type contact layer which satisfies the thermal breakdown of the drive voltage Vf and the thermal damage of the non-directional light-emitting layer is preferably 8 to 2 0 n m /min. Further, the determination of the growth rate is obtained by measuring the film thickness of the p-type contact layer by electron microscopic (TEM) observation of the wafer cross section, and determining by the growth time. Therefore, the observation sample in which several conditions have been changed is prepared, and the relationship between the growth rate and the supply amount of TMGa per unit time is investigated in advance, and the necessary growth conditions can be determined in advance. -15- 200812111 After the growth of the P-type contact layer, n2 gas is supplied together with the heating of the stop substrate, and the inside of the reaction space is cleaned and cooled, and cooled until it can be taken out of the growth apparatus. Further, in the present method, it was confirmed that the p-type semiconductor layer became the intended P-type at this point. Therefore, heat treatment with activation after this is unnecessary. Next, the negative electrode and the positive electrode provided on the n-type contact layer and the p-type contact layer will be described. (Negative Electrode) As a negative electrode, various compositions and structures are generally known, and the present invention also includes such generally known materials, and any composition and structure can be used. The production method is also generally known in various processes, and generally known methods can be used. The fabrication of the negative electrode forming surface on the n-type contact layer can be performed by a generally well-known photolithography technique and a general uranium engraving technique. By these techniques, from the uppermost layer of the wafer to the position of the n-type contact layer, the n-type contact layer of the region where the negative electrode is formed is exposed. As the negative electrode material, in addition to Al, Ti, Ni, and Au which are contact metals to the n-type contact layer, metal materials such as Cr, W, and V can be used. In order to improve the adhesion to the contact layer, it is also preferable to use the contact metal as a multilayer structure selected from the above plurality of metals. Further, if the outermost surface is Au, the bonding property becomes good. (Positive Electrode) The positive electrode provided on the p-type contact layer can also be known as a light-transmitting or reflective positive electrode made of various compositions and structures, and the present invention also includes such a known thing, and any composition can be used. And the structure of the object. In the case of the present invention, it is preferable to use a method in which the light-transmitting enthalpy is used for the positive electrode, because the effect of lowering the driving voltage is large. The composition is made as 50% S Ιη <1 00% and 〇% <Sn $ 50%. This range can satisfy low film impedance and high light transmittance. It is particularly preferable that I η is such that Sn is 10%. It is also preferred that the lanthanide is an impurity, a dish, a group IV or a group V. Further, SnO, ZnO, InO, or the like is used instead of ITO. The film thickness of the ruthenium film is preferably 50 to 500 nm. If it is 50 nm, the film resistance of the ruthenium film itself becomes high, and the driving voltage becomes high. Further, when it is thicker than 50,000 nm, the efficiency of extraction into the upper surface is lowered, and the light emission is increased. As for the film formation method of the ITO film, a generally known plating method or sputtering method can be used. In the vacuum deposition, the heating method may be an impedance addition or an electron beam heating method, but a vapor deposition beam heating method of a material other than metal is suitable. Further, a method in which the compound is applied as a liquid and the oxide film is appropriately treated after application to the surface can be used. In the vapor deposition method, the crystallinity of the ITO film is affected by the conditions, and the conditions are appropriately selected without limitation. Further, when ITO is formed at room temperature, heat treatment for transparency is necessary. The film formation by sputtering is due to the higher general method of being placed in the plasma. Μ ITO of material thinks. In 90%, there are Yi people, but also the following, on the other hand, if the output is not vacuum steaming, the electric material is used as the material, but if the film is in the environment, the surface of the P-type contact layer is easy. The contact resistance tends to be high due to damage to the plasma, but it is affected by the film formation conditions and the effect of the small-direction P-type contact layer surface. Usually, after a light transmissive or reflective positive electrode is formed, a pad layer constituting a pad portion is formed on the surface of the portion. It is combined with the pad to form a positive electrode. As the material of the pad layer, various structures are known, and in the present invention, these generally known materials can be used without particular limitation. In addition to Al, Ti, Ni, and Au used for the negative electrode material, Cr, W, and V may be used without any limitation. However, it is preferable to use a material having good adhesion to the light-transmitting or reflective positive electrode which has been used. The thickness is such that, for the stress at the time of bonding, it is not required to be sufficiently thickened by damage to the light transmissive or reflective positive electrode, and the outermost layer is excellent in adhesion to the bonding ball. For example, Αιι is the best. The cuban nitride semiconductor light-emitting device of the present invention can be provided, for example, as a lamp by a transparent cover body by means well known in the art. Further, a white lamp may be produced by combining a lid body having the gallium nitride-based compound semiconductor light-emitting device of the present invention and a phosphor. In addition, since the lamp system produced by the m-type nitride semiconductor light-emitting device of the present invention has a low driving voltage and a high light-emitting output, an electronic device such as a portable telephone, a display, or a panel manufactured by the above-described technology is mounted. Or a mechanical device such as a car, a computer, or a game machine on which the electronic device is mounted, which can be driven with low electric power and can realize high characteristics. In particular, battery-driven devices such as portable phones, game consoles, toys, and auto parts have the effect of saving power. -18- 200812111 EXAMPLES Hereinafter, the present invention will be specifically described by way of Examples and Comparative Examples, but the present invention is not limited to these Examples. (Example 1) A sapphire substrate was placed on a receiver of a reaction furnace, the pressure in the furnace was controlled at 20 kPa (200 mbar), the substrate temperature was controlled at 1 1 〇〇 ° C, and TMA1 and NH3 were supplied together with the H2 carrier gas. On the substrate, an A1N buffer layer is formed. This growth time is taken as 1 point. Thereafter, a pressure of 40 kPa (400 mbar) was applied as a temperature of 1300 °C to supply TMGa and NH3 to the A1N buffer layer, and a base layer composed of undoped GaN was grown for 3 hours. Next, while maintaining the pressure and temperature, tetramethylphosphonium was supplied as an n-type dopant, and the n-type GaN layer was grown for 1 hour. The supply amount of tetramethylphosphonium was adjusted so that the carrier concentration of the n-type GaN layer became 4.0 x 1 018 cT3. After the n-type contact layer is formed by this, the pressure system is converted to H2 from H2 to N2 at a temperature of 750 ° C as a temperature of 750 ° C, and TEH and NH 3 are supplied as dopants, and SiH 4 is supplied as a dopant. After the barrier layer was grown for 7 minutes, TMIn was further supplied and the well layer was grown for 5 minutes. The interaction of the barrier layer and the well layer is repeated five times in an interactive manner, and finally the final barrier layer is grown as a light-emitting layer. The amount of SiH4 of the dopant was adjusted so that the Si concentration in the barrier layer and the well layer became 2.0 x 1 017 atoms/cm3. • After 19-200812111, the pressure is taken as 20 kPa (200 mbar), the temperature is converted to H2 again as l〇〇〇°C, and TMGa and TMA1' are supplied as dopants to Cp2Mg for 3 minutes. A P-type cover layer composed of Gao^AlusN. Thereafter, the growth of the p-type contact layer composed of GaonAUnN was carried out while maintaining the pressure and temperature. As a result, the TMGa supply amount was fixed so that the growth rate became 8 nm/min. The growth time is 15 points. The Mg concentration in the p-type contact layer adjusts the supply amount of Cp2Mg so as to be 1x1 Ο" atoms/cm3. Then, the power supply to the induction coil is stopped, the heating is stopped, and the carrier gas is converted to N2, and the inside of the clean furnace is stopped. Together, the obtained group m nitride semiconductor layer is cooled to a temperature taken out of the furnace. A part of the n-type contact layer of the group m nitride semiconductor layer taken out of the furnace is subjected to photo etching and dry etching. On the other hand, a negative electrode made of a metal layer of cr and Ti was formed thereon. Further, ITO 1 having a thickness of 350 nm was formed on the tantalum contact layer by vapor deposition (In: Sn = 9). :l), a pad layer laminated in the order of Ti, An, A1, and Au is formed thereon as a positive electrode. Thereafter, after honing the substrate and cutting, the substrate is divided into light-emitting elements. The element was allowed to emit light by a current of 20 mA, and the driving voltage Vf and the light-emitting output were measured, and as a result, it was 3.2 V and 9 mW. The measurement result of the current-voltage characteristic curve is shown in Fig. 3. The η 求出 obtained by the formula 1 is 1.8. Further, the results of these and other examples and comparative examples are shown in Table 1 -20. 200812111 (Example 2). The growth conditions of the P-type contact layer are set to 2 Onm / min. A quaternary nitride semiconductor light-emitting device was produced in the same manner as in Example 1 except that the growth time was 6 minutes. The obtained light-emitting device was evaluated in the same manner as in Example 1, and the driving voltage was 3.2 V and the light-emitting output was 9 mW. The measurement result of the current-voltage characteristic curve is shown in Fig. 4. The η 求出 obtained by the above formula (1) is 2.0. (Comparative Example 1) The growth conditions of the ruthenium-type contact layer are A quaternary nitride semiconductor light-emitting device was produced in the same manner as in Example 1 except that the growth rate was 7 nm/min and the growth time was 1 7.2. The obtained light-emitting device was evaluated in the same manner as in Example 1, and the driving voltage was 3.3. V. The light-emission output was 8 mW. The measurement result of the current-voltage characteristic curve is shown in Fig. 5. The η 求出 obtained by the above formula (1) is 2.6. (Comparative Example 2) Connect In the growth condition of the layer, a group m nitride semiconductor light-emitting device was produced in the same manner as in Example 5 except that the growth rate was 2 1 nm /min and the growth time was 5.7. -21 - 200812111 The obtained light-emitting device and examples (1) The same evaluation was performed, and the driving voltage was 3.4 V and the light-emitting output was 8 mW. The measurement result of the current-voltage characteristic curve is shown in Fig. 6. The graph obtained by the above formula (1) is used.値 is 3.1. Table 1 Growth rate of contact layer V (in nm/min) Time-divided drive voltage Vf (applied 20 mA) V illuminating output mW (20 mA) η 値 Example 1 8 15 3.2 9 1.8 Example 2 20 6 3.2 9 2 Comparative Example 1 7 17.2 3.3 8 2.6 Comparative Example 2 21 5.7 3.4 8 3.1 Industrial Applicability The m-nitride semiconductor light-emitting device of the present invention has an excellent light-emitting output and a drop in driving voltage, so its industry The utilization price 値 is very large [Simplified description of the drawing] Fig. 1 is a diagram for explaining the current-voltage characteristics of the light-emitting element in the forward direction. ^ Fig. 2 is a schematic view showing a cross section of a cuban nitride semiconductor light-emitting device of the present invention. Fig. 3 is a graph showing the measurement results of the current-voltage characteristic curve of the light-emitting element which can be obtained in the first embodiment. Fig. 4 is a graph showing the measurement results of the current characteristic curve of the current -22-200812111 of the light-emitting element which can be obtained in the second embodiment. Fig. 5 is a graph showing the measurement results of the current-voltage characteristic curve of the light-emitting element which can be obtained in Comparative Example 1. Fig. 6 is a graph showing the measurement results of the current-voltage characteristic curve of the light-emitting element which can be obtained in Comparative Example 2. [Description of main component symbols] _ 1 : Substrate 2 : Buffer layer 3 : η-type semiconductor layer 4 : Light-emitting layer 5b : p-type contact layer 5a : p-type cladding layer 5 : p-type semiconductor layer 6 : negative electrode · h : light transmission Or reflective positive electrode 7b: pad layer 7: positive -23-