JPH0360800B2 - - Google Patents

Info

Publication number
JPH0360800B2
JPH0360800B2 JP59042851A JP4285184A JPH0360800B2 JP H0360800 B2 JPH0360800 B2 JP H0360800B2 JP 59042851 A JP59042851 A JP 59042851A JP 4285184 A JP4285184 A JP 4285184A JP H0360800 B2 JPH0360800 B2 JP H0360800B2
Authority
JP
Japan
Prior art keywords
film thickness
growth
flow rate
gas
substrate
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
JP59042851A
Other languages
Japanese (ja)
Other versions
JPS60191096A (en
Inventor
Yukio Noda
Yukitoshi Kushiro
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
KDDI Corp
Original Assignee
Kokusai Denshin Denwa KK
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kokusai Denshin Denwa KK filed Critical Kokusai Denshin Denwa KK
Priority to JP4285184A priority Critical patent/JPS60191096A/en
Publication of JPS60191096A publication Critical patent/JPS60191096A/en
Publication of JPH0360800B2 publication Critical patent/JPH0360800B2/ja
Granted legal-status Critical Current

Links

Classifications

    • CCHEMISTRY; METALLURGY
    • C30CRYSTAL GROWTH
    • C30BSINGLE-CRYSTAL GROWTH; UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL; PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; APPARATUS THEREFOR
    • C30B25/00Single-crystal growth by chemical reaction of reactive gases, e.g. chemical vapour-deposition growth
    • C30B25/02Epitaxial-layer growth
    • C30B25/14Feed and outlet means for the gases; Modifying the flow of the reactive gases

Landscapes

  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • General Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Crystallography & Structural Chemistry (AREA)
  • Materials Engineering (AREA)
  • Metallurgy (AREA)
  • Organic Chemistry (AREA)
  • Crystals, And After-Treatments Of Crystals (AREA)

Description

【発明の詳細な説明】[Detailed description of the invention]

(産業上の利用分野) 本発明は、InP系の如き化合物半導体単層膜あ
るいは多層膜の位相成長方法に関するものであ
る。 (従来技術とその問題点) 化合物半導体層を結晶成長させる方法のひとつ
に気相成長法があり、これは結晶成長させるべき
原料を気化(ガス化)して成長させるものであ
る。 しかし、この気相成長方法においては、第1図
の模式図に示すごとく、ガスの噴射口、すなわ
ち、基板1上へ流入するガス流の上流端(以下基
板上流端という)からの距離xが離れるに従つ
て、結晶成長させた半導体層2の膜厚D(x)が
減少する傾向がある。一般に、基板上流端からの
距離xと半導体層2の膜厚D(x)との関係は膜
厚分布の均一性と呼ばれ、膜厚D(x)が距離x
によつて異なる場合には“膜厚分布の均一性が悪
い”というように表現されている。 さらに、気相成長方法にて成長させた半導体層
2の表面に、主として成長丘およびヒロツクと呼
ばれる表面欠陥の存在によつて表面モホロジーが
悪化するという問題がある。 このように、気相成長方法によつて成長された
半導体層2の膜厚の均一性あるいは表面モホロジ
ーが悪くなれば、デバイス作製時の歩留りが悪く
なり、また、距離x方向に長いデバイスを作製す
ることが困難となる。従つて、半導体層2の膜厚
の均一性あるいは表面モホロジーを向上させるこ
とは、半導体層2を作製する上で重要な問題であ
る。 従来は、膜厚の均一性と表面モホロジーとは全
く別の原因に起因するものと考えられていて、そ
れぞれに対し次に述べるように別々の対策を施し
ていた。 (1) 膜厚の均一性の向上に対する対策:Komeno
〔J.Electrochem・Soc.124(1977)1440〕によれ
ば、基板1の上流端から距離xにおける膜厚D
(x)はCを定数とすれば D(x)=C1/(25.0μ/ρV)1/2・(Ks/Ds)・
x1/2+1…(1) で与えられることが示されている。ここで、μは
気体(キヤリアガスと原料ガスを総称したもの
で、以下「気体」と略す)の粘性係数、ρは気体
密度、Vは気体の流速、Ksは基板の表面におけ
る化学反応の速度定数、Dsは拡散係数である。
この基板の表面における化学反応の速度定数Ks
は温度のみの関数であり、かつ表面における化学
反応が1次反応であると仮定している。 (1)式を更に変形すれば、膜厚に関する次式が得
られる。 (2)式の第1式は〔{D(0)/D(x)}−1〕2
xとが比例関係にありその比例定数がFhである
ことを示している。 ここでD(0)は、基板上流端からの距離xが
零である膜厚、Fhは膜厚の均一性を表わすパラ
メータとして用いられる膜厚パラメータである。 (2)式の第2式から明らかなように、膜厚パラメ
ータFhは気体の流速Vすなわち気体の流量に反
比例する。従つて、膜厚パラメータFhを小さく
して膜厚の均一性を良くするには、気体の流速V
すなわち気体の流量を大きくすればよい。この
Komeno氏等の提案は広く一般に支持されてい
る。 尚、実際に行われている膜厚の均一性を向上さ
せる対策として、次の方法がとられている。 a○ 気体の流速が大きくなると境界層の厚さが薄
くなり拡散律速が抑制され、膜厚分布の均一性
が向上するという予想に基づき、気体の流量を
できる限り増加し、気体の流速を大きくして成
長させる。これは上述したKomeno氏等の考え
に基づいた改善例である。 b○ 基板面を気体の流れに対しθなる角度で傾斜
させ、さらに基板を回転させる。角度θは実験
にて最適値を見出す。 しかしながら、a○については通常の約5倍程度
気体の流量を大きくするため、AsH3、PH3とい
つた極めて毒性の強いV族系ガスの流量も大きく
しなければならず、安全性の面において問題があ
る。また、気体の消費量も大きくなり、排ガスの
処理能力も大きくしなければならず、経済性の面
でも問題がある。b○については、1個の成長室を
持ち、主として単一層を作製するための気相成長
装置においては、その実施が比較的容易である
が、2個以上の成長室を持つ半導体レーザに代表
される多層構造を作製するための気相成長装置に
おいては、その実施は不可能ではないにしても極
めて困難である。 (2) 表面モホロジーを向上させるための対策: 例えばInP結晶表面には、直径が1mm以内、高
さ数μmで円錐に近い形状をした成長丘および長
軸数百μm、短軸数10μmの楕円形底部を呈し数
μmの高さを有するヒロツクが存在する。成長丘
およびヒロツクの成因には諸説もあり要約すると
以下のようになる。 基板に付着している異物が成長時結晶化の核
として作用し異常成長が起こりそこに表面欠陥
が形成される。あるいは、成長中基板上に反応
管の内壁析出物が落下し、これが前記基板付着
異物と同じ作用を及ぼす。 基板エツチング時にエツチヤントの種類、エ
ツチング条件及びエツチング処理方法が最適化
されていないと基板表面の一部が酸化等の劣化
を起こし、この劣化が原因となつて表面欠陥が
形成される。 化合物半導体成長時のあるいは族を供給
するガスの分圧があるいは族を供給するガ
スの分圧よりも大きい場合には、あるいは
族の液化した金属が基板上に形成されやすくな
り、そこにおいてVSL機構で説明されるよう
な原理で異常成長が起こる。 従来においては、これ等の成因に準拠してそれ
ぞれ表面モホロジーを改善してきた。例えば、
について、反応管内壁析出物の基板上への落下を
防ぐため基板成長面を下側にすること、につい
て基板エツチング処理を最適化し基板表面の劣化
を抑制すること、についてまたは族を供給
するガスの分圧をまたは族を供給するガスの
分圧より大きくするといつたことが行われてき
た。しかしながら、の方法は1個の成長室を持
ち主として単一層を作製するための気相成長装置
においてその実施は容易であるが、2個以上の成
長室を持ち半導体レーザに代表される多層構造を
作製するための気相成長装置においてその実施は
困難である。また、の方法に関しては、例えば
InP基板のエツチングの場合を考えると、通常2
種類のエツチヤントすなわち1H2SO4:8H2O2
1H2O溶液および0.6Br:99.4CH3OH溶液を使い、
各々0℃で2分および25℃で3分というエツチン
グ条件で処理しているが、エツチヤントを前記2
種類に限定してもそれぞれの溶液の混合比を変え
さらにエツチング条件として温度および処理時間
を変えてエツチングの処理を最適化するのは決し
て容易ではない。さらに、の方法はその実施が
容易であるという長所は持つているものの、例え
ば−族化合物半導体成長時はAsH3,PH3
いつた極めて毒性の強い族供給ガスを通常と比
べ多量に使用せねばならず安全上問題がある。 異常の説明から明らかなように、従来は膜厚の
均一性と表面モホロジーとが各々異なつた原因に
より悪化するので、その原因毎に別々の対策を施
しており、膜厚の均一性および表面モホロジーを
同時に満足するような良質な半導体層を得ること
が困難であつた。 (発明の目的) 本発明は、上述した従来技術の欠点に鑑み、気
体の流速すなわち気体の流量を変えることなく原
料ガスの分圧を変えることにより、膜厚の均一性
と表面モホロジーとを同時に改善し、優れた化合
物半導体層を成長させ得る気相成長方法を提供す
ることを目的とする。 (発明の構成および作用) この目的達成のために、本発明による気相成長
方法は、InP基板上にInP単層膜あるいはIn1-a
GaxAs1-yPy系(0x1,0y1)の多
層膜を常圧下で成長させる化合物半導体層のハイ
ドライド気相成長法において、気体の流量一定の
条件の下で結晶成長させるべき該化合物半導体層
の各原料ガスの分圧は7/1500から7/2800までの範
囲で同等になるようにして成長させることを特徴
とする構成を有している。 以下本発明を詳細に説明する。 前掲の(2)式から、膜厚パラメータFhは気体の
流速Vすなわち気体の流量に反比例することは明
らかであり、前述のように膜厚パラメータFh
気体の流速Vのみによつて決定され、原料ガスの
分圧比には依存しないという説が広く一般に支持
されてきた。 しかし、本願発明者は気相成長時において基板
上に生起する現象は近似的に(1)式で表わされる成
長機構に従うものの、基板の表面反応を示す速度
定数Ksは単純な1次反応ではなく、原料ガスの
流量にも影響されるものと考えた。すなわち、成
長温度や気体の流速Vを一定にしておいても、原
料ガスと気体(=原料ガスの流量+キヤリアガス
の流量)との比、すなわち原料ガスの分圧を変え
れば膜厚パラメータFhが変わると考えた。 この論理を実証するために、本願発明者は特に
結晶成長が難しいとされているInPホモエピタキ
シヤル層について、成長温度700(℃)、気体の流
量1500(c.c.・min-1)を一定にし、原料ガスの分
(Industrial Application Field) The present invention relates to a method for phase growth of a single layer or multilayer film of a compound semiconductor such as InP. (Prior Art and Its Problems) One of the methods for growing crystals of compound semiconductor layers is vapor phase growth, in which the raw material for crystal growth is vaporized (gasified). However, in this vapor phase growth method, the distance x from the gas injection port, that is, the upstream end of the gas flow flowing onto the substrate 1 (hereinafter referred to as the substrate upstream end), is As the distance increases, the thickness D(x) of the crystal-grown semiconductor layer 2 tends to decrease. Generally, the relationship between the distance x from the upstream end of the substrate and the film thickness D(x) of the semiconductor layer 2 is called the uniformity of film thickness distribution, and the film thickness D(x) is
If the thickness differs depending on the conditions, it is expressed as "poor uniformity of film thickness distribution." Furthermore, there is a problem in that the surface morphology of the semiconductor layer 2 grown by the vapor phase growth method is deteriorated mainly due to the presence of surface defects called growth hills and hills. In this way, if the uniformity of the film thickness or the surface morphology of the semiconductor layer 2 grown by the vapor phase growth method deteriorates, the yield during device fabrication will decrease, and it will be difficult to fabricate a device with a long distance in the x direction. It becomes difficult to do so. Therefore, improving the uniformity of the film thickness or surface morphology of the semiconductor layer 2 is an important issue in manufacturing the semiconductor layer 2. Conventionally, film thickness uniformity and surface morphology were considered to be caused by completely different causes, and separate countermeasures were taken for each as described below. (1) Measures to improve film thickness uniformity: Komeno
According to [J. Electrochem Soc. 124 (1977) 1440], the film thickness D at a distance x from the upstream end of the substrate 1
If (x) is a constant, then D(x)=C1/(25.0μ/ρV) 1/2・(K s /D s )・
It is shown that it is given by x 1/2 + 1...(1). Here, μ is the viscosity coefficient of the gas (carrier gas and source gas are collectively referred to as "gas" hereafter), ρ is the gas density, V is the gas flow rate, and K s is the rate of chemical reaction on the surface of the substrate. The constant, Ds, is the diffusion coefficient.
The rate constant K s of the chemical reaction on the surface of this substrate
is a function of temperature only and assumes that the chemical reaction at the surface is a first-order reaction. By further transforming equation (1), the following equation regarding film thickness can be obtained. The first equation of equation (2) shows that [{D(0)/D(x)}-1] 2 and x are in a proportional relationship, and the proportionality constant is F h . Here, D(0) is the film thickness at which the distance x from the upstream end of the substrate is zero, and F h is a film thickness parameter used as a parameter representing the uniformity of the film thickness. As is clear from the second equation (2), the film thickness parameter F h is inversely proportional to the gas flow rate V, that is, the gas flow rate. Therefore, in order to reduce the film thickness parameter F h and improve the uniformity of the film thickness, the gas flow velocity V
In other words, the gas flow rate may be increased. this
Komeno et al.'s proposal has wide public support. The following method is actually used to improve the uniformity of film thickness. a○ Based on the prediction that as the gas flow rate increases, the thickness of the boundary layer becomes thinner, diffusion control is suppressed, and the uniformity of the film thickness distribution improves. and grow it. This is an example of an improvement based on the idea of Mr. Komeno et al. mentioned above. b○ The substrate surface is tilted at an angle θ with respect to the gas flow, and the substrate is further rotated. The optimum value for the angle θ is found through experiments. However, for a○, the flow rate of gas is about five times higher than usual, so the flow rate of highly toxic Group V gases such as AsH 3 and PH 3 must also be increased, which poses safety concerns. There is a problem with this. In addition, the amount of gas consumed increases, and the processing capacity for exhaust gas must also be increased, which poses problems in terms of economic efficiency. Regarding b○, it is relatively easy to implement in a vapor phase growth apparatus that has one growth chamber and is mainly used for producing a single layer, but it is typical for semiconductor lasers that have two or more growth chambers. This is extremely difficult, if not impossible, to implement in a vapor phase growth apparatus for producing a multilayer structure. (2) Measures to improve surface morphology: For example, the InP crystal surface has growth hills with a diameter of 1 mm or less and a height of several μm, close to a cone, and an ellipse with a major axis of several hundred μm and a minor axis of several 10 μm. There are hills with a shaped bottom and a height of several μm. There are various theories as to the origin of growth mounds and hirotsuku, and they can be summarized as follows. Foreign matter adhering to the substrate acts as a nucleus for crystallization during growth, causing abnormal growth and forming surface defects there. Alternatively, precipitates on the inner wall of the reaction tube fall onto the substrate during growth, which exerts the same effect as the foreign matter adhering to the substrate. If the type of etchant, etching conditions, and etching treatment method are not optimized during substrate etching, a portion of the substrate surface will undergo deterioration such as oxidation, and this deterioration will cause surface defects to be formed. If the partial pressure of the gas supplying the group during compound semiconductor growth is greater than the partial pressure of the gas supplying the group, liquefied metal of the group tends to be formed on the substrate, where the VSL mechanism Abnormal growth occurs based on the principle explained in . Conventionally, surface morphology has been improved based on these factors. for example,
In order to prevent precipitates on the inner wall of the reaction tube from falling onto the substrate, the growth surface of the substrate should be on the lower side; in order to optimize the substrate etching process and suppress deterioration of the substrate surface; It has been done to increase the partial pressure to or greater than the partial pressure of the gas supplying the group. However, although this method is easy to implement in a vapor phase growth apparatus that has one growth chamber and is used to fabricate a single layer, it has two or more growth chambers and is used to fabricate a multilayer structure such as a semiconductor laser. It is difficult to implement this method using a vapor phase growth apparatus for manufacturing. Also, regarding the method, for example,
Considering the case of etching an InP substrate, it is usually 2
Types of etchingants i.e. 1H 2 SO 4 : 8H 2 O 2 :
Using 1H2O solution and 0.6Br: 99.4CH3OH solution,
The etching conditions were 0°C for 2 minutes and 25°C for 3 minutes.
Even if the types are limited, it is by no means easy to optimize the etching process by changing the mixing ratio of each solution and further changing the temperature and processing time as etching conditions. Furthermore, although the method has the advantage of being easy to implement, for example, when growing - group compound semiconductors, extremely toxic group supply gases such as As H 3 and PH 3 are used in larger quantities than usual. There is a safety issue when using it. As is clear from the explanation of the abnormality, in the past, film thickness uniformity and surface morphology deteriorated due to different causes, so separate countermeasures were taken for each cause. It has been difficult to obtain a high-quality semiconductor layer that satisfies both of the above requirements. (Object of the Invention) In view of the above-mentioned shortcomings of the prior art, the present invention improves film thickness uniformity and surface morphology at the same time by changing the partial pressure of the raw material gas without changing the gas flow rate, that is, the gas flow rate. It is an object of the present invention to provide an improved vapor phase growth method capable of growing an excellent compound semiconductor layer. (Structure and operation of the invention) In order to achieve this object, the vapor phase growth method according to the present invention provides an InP single layer film or an In 1-a film on an InP substrate.
In the hydride vapor phase growth method for compound semiconductor layers in which a multilayer film of Ga x As 1-y P y system (0x1, 0y1) is grown under normal pressure, the compound semiconductor to be crystal grown under conditions of a constant gas flow rate. The structure is characterized in that growth is performed so that the partial pressures of the raw material gases in each layer are equal in the range from 7/1500 to 7/2800. The present invention will be explained in detail below. From equation (2) above, it is clear that the film thickness parameter F h is inversely proportional to the gas flow rate V, that is, the gas flow rate, and as mentioned above, the film thickness parameter F h is determined only by the gas flow rate V. The theory that it does not depend on the partial pressure ratio of the raw material gas has been widely supported. However, the inventor of the present application found that although the phenomenon that occurs on the substrate during vapor phase growth approximately follows the growth mechanism expressed by equation (1), the rate constant K s indicating the surface reaction of the substrate is not a simple first-order reaction. It was thought that this would be influenced by the flow rate of the raw material gas. In other words, even if the growth temperature and gas flow rate V are kept constant, the film thickness parameter F h can be changed by changing the ratio of the source gas to the gas (= flow rate of source gas + flow rate of carrier gas), that is, the partial pressure of the source gas. I thought that would change. In order to prove this logic, the inventor of the present application fixed the growth temperature of 700 (℃) and the gas flow rate of 1500 (cc min -1 ) for the InP homoepitaxial layer, which is said to be particularly difficult to crystallize. Minutes of raw material gas

【表】【table】

【表】 圧を変えて実験を行つた。表1はハイドライド気
相成長による実験結果である。 表1から明らかなように、成長温度と気体の流
量1500(c.c.・min-1)を一定にしても、10%In−
HCl流量や10%PH3流量を変えること、すなわち
原料ガス流量(原料ガスの分圧)を変えることに
よつて、膜厚パラメータFhも変わり、特に原料
ガス流量が各々70(c.c.・min-1)で膜厚パラメー
タFhが最小値となつている。 尚、この時の10%In−HCl原料ガスの分圧は 原料ガス流量/気体の流量=70×10/100/1500=7
/1500 である。 また、成長温度を一定し、気体の流速V(気体
の流量)を1500(c.c.・min-1)から2800(c.c.・
min-1)に変えると、従来の改善例で述べたよう
に膜厚パラメータFhは0.11(mm-1)から0.02(mm-1
と改善されるものの、気体の流量が2800(c.c.・
min-1)と一定にしても原料ガスの流量を変える
ことによつて膜厚パラメータFhも変える。この
ように、成長温度と気体の流速V(気体の流量)
を一定にしても、原料ガスの分圧を変えることに
より膜厚パラメータFhを変えることができ、か
つ膜厚パラメータFhが最少となる原料ガスの分
圧の最適値がある。この原料ガスの分圧(原料ガ
ス流量)の最適値は結晶成長する原料により固有
の値を持つものと考えられる。例えば、InP基板
と格子整合する三元混晶InGaAsでは気体の流量
が1000c.c.・min-1で、かつ10%In−HCl,10%Ga
−Cl、10%AsH3の流量がそれぞれ58c.c.・min-1
8c.c.,min-1,65c.c.・min-1のとき最適膜厚パラ
メータとしてFh=0.05が得られた。また、InP基
板と格子整合し、波長換算の禁制帯幅が1.5μmで
ある四元混晶InGaAsPでは、気体の流量が1100
c.c.・min-1でかつ10%In−HCl,10%−GaCl,10
%PH3,10%AsH3の流量がそれぞれ60c.c.・
min-1、5c.c.・min-1、20c.c.・min-1、46c.c.・
min-1のとき最適膜厚パラメータとしてFh=0.04
が得られた。 また、ヘテロエピタキシヤル成長においても上
述した理論を適用できるかどうかを確認するため
に、次の実験を追試した。 InP基板上に、In1-xGaxAs1-yPy系(0x
1,0y1)の多層膜であるInP/
InGaAsP/InGaAsの三層をヘテロエピタキシヤ
ル成長させる際、キヤツプ層であるInP層を表1
中のウエハーF62と同一成長条件で成長させた。
この時のInPキヤツプ層の膜厚パラメータFh
0.10〜0.20(mm-1)となり、ホモエピタキシヤル成
長時の値(0.11mm-1)とほぼ等しくなつた。すな
わち、ヘテロエピタキシヤル成長においても、膜
厚パラメータFhが最少となる原料ガスの分圧の
最適値はほぼ一定である。 なお、第2図はInP基板上に表1に掲げた各サ
ンプルF62,F89,F95,F97,F99,F105につい
て原料ガスの分圧条件でハイドライド気相成長を
用いて成長させたInPホモエピタキシヤル層の膜
厚分布を図示したものである。横軸は基板上流端
からの距離x/mm、縦軸は膜厚D(x)を基板上
流端x=0における膜厚D(0)で規格化した膜
厚D(x)/D(0)である。 第3図は、表1の実験結果を前述した(2)式に基
づいて、縦軸に〔D(0)/D(x)−1〕2を、横
軸に距離x/mmをとつた平面上にプロツトしたも
のであり、各サンプルF62,F89,F95,F97,
F99,F105の測定結果はほぼ直線上にのる。 (2)式の第1式から明らかなようにこの直線の勾
配よりFhを決定することができる。 第4図は、縦軸に第3図の直線の勾配から膜厚
パラメータFhを求め、横軸に族の原料ガスで
あるInClガスの分圧PIoClの関数として図示したも
ので、同図から原料ガスの分圧と膜厚パラメータ
Fhとの関係が明瞭にわかるようにしたものであ
る。ここで白丸は流量1500c.c.・min-1の場合であ
り、黒丸は2800c.c.・min-1の場合である。この図
からわかるように、流量一定の条件の下で分圧の
変化をすることにより、最適の膜厚パラメータ
Fhが得られることがわかる。 以上の説明では膜厚の均一性と原料ガスの分圧
との関係についてのみ述べたが、次に原料ガスの
分圧あるいは膜厚の均一性と表面モホロジーとの
関連について説明する。 表2は成長温度700(℃)、気体の流量1500(c.c.・
min-1)を一定にして成長したウエハーF62,F89
およびF95の成長丘の面密度、成長丘の直径、ヒ
ロツクの面密度を測定した結果を示している。
[Table] Experiments were conducted by changing the pressure. Table 1 shows the experimental results using hydride vapor phase growth. As is clear from Table 1, even if the growth temperature and gas flow rate of 1500 (cc min -1 ) are constant, 10% In-
By changing the HCl flow rate and 10% PH 3 flow rate, that is, by changing the raw material gas flow rate (partial pressure of raw material gas), the film thickness parameter F h also changes . 1 ), the film thickness parameter F h has the minimum value. The partial pressure of the 10% In-HCl raw material gas at this time is: Raw material gas flow rate/Gas flow rate = 70 x 10/100/1500 = 7
/1500. In addition, the growth temperature was kept constant, and the gas flow rate V (gas flow rate) was changed from 1500 (cc min -1 ) to 2800 (cc min -1).
min -1 ), the film thickness parameter F h changes from 0.11 (mm -1 ) to 0.02 (mm -1 ) as described in the conventional improvement example.
However, the gas flow rate is 2800 (cc・
min -1 ), the film thickness parameter F h is also changed by changing the flow rate of the source gas. In this way, the growth temperature and gas flow rate V (gas flow rate)
Even if F h is kept constant, the film thickness parameter F h can be changed by changing the partial pressure of the raw material gas, and there is an optimum value of the partial pressure of the raw material gas at which the film thickness parameter F h is minimized. It is considered that the optimum value of the partial pressure of the raw material gas (raw material gas flow rate) has a value specific to the raw material used for crystal growth. For example, in the case of ternary mixed crystal InGaAs that is lattice-matched to the InP substrate, the gas flow rate is 1000 c.c.・min -1 , and 10% In-HCl and 10% Ga
-The flow rates of Cl and 10% AsH 3 were 58 c.c.・min -1 , respectively.
When the film thickness was 8 c.c., min -1 and 65 c.c.min -1 , F h =0.05 was obtained as the optimum film thickness parameter. In addition, in the case of quaternary mixed crystal InGaAsP, which is lattice-matched to the InP substrate and has a wavelength-converted forbidden band width of 1.5 μm, the gas flow rate is 1100 μm.
cc・min -1 and 10% In-HCl, 10%-GaCl, 10
The flow rate of %PH 3 and 10% AsH 3 is 60c.c.・
min -1 , 5c.c.・min -1 , 20c.c.・min -1 , 46c.c.・
When min -1, the optimal film thickness parameter is F h = 0.04
was gotten. In addition, in order to confirm whether the above-mentioned theory can be applied to heteroepitaxial growth, the following experiment was repeated. In 1-x Ga x As 1-y P y system (0x
InP/1,0y1) multilayer film
When growing three layers of InGaAsP/InGaAs heteroepitaxially, the InP layer as the cap layer is grown as shown in Table 1.
It was grown under the same growth conditions as wafer F62 inside.
At this time, the film thickness parameter F h of the InP cap layer is
The value was 0.10 to 0.20 (mm -1 ), which was almost equal to the value during homoepitaxial growth (0.11 mm -1 ). That is, even in heteroepitaxial growth, the optimal value of the partial pressure of the source gas at which the film thickness parameter F h is minimized is approximately constant. Figure 2 shows InP homoepitaxial growth on an InP substrate using hydride vapor phase epitaxy under the partial pressure conditions of the source gas for each of the samples F62, F89, F95, F97, F99, and F105 listed in Table 1. This is a diagram illustrating the thickness distribution of the layers. The horizontal axis is the distance x/mm from the upstream end of the substrate, and the vertical axis is the film thickness D(x)/D(0), which is the film thickness D(x) normalized by the film thickness D(0) at the upstream end of the substrate x=0. ). Figure 3 shows the experimental results in Table 1 based on equation (2) described above, with [D(0)/D(x)-1] 2 plotted on the vertical axis and distance x/mm on the horizontal axis. It is plotted on a plane, and each sample F62, F89, F95, F97,
The measurement results for F99 and F105 are almost on a straight line. As is clear from the first equation (2), F h can be determined from the slope of this straight line. In Figure 4, the film thickness parameter F h is determined from the gradient of the straight line in Figure 3 on the vertical axis, and is plotted on the horizontal axis as a function of the partial pressure P IoCl of InCl gas, which is the raw material gas of the group. Partial pressure of raw material gas and film thickness parameters from
This allows the relationship with F h to be clearly understood. Here, the white circles are for a flow rate of 1500 c.c.・min -1 , and the black circles are for a flow rate of 2800 c.c.・min -1 . As you can see from this figure, by changing the partial pressure under the condition of constant flow rate, the optimal film thickness parameter can be determined.
It can be seen that F h can be obtained. In the above explanation, only the relationship between the uniformity of the film thickness and the partial pressure of the raw material gas has been described. Next, the relationship between the partial pressure of the raw material gas or the uniformity of the film thickness and the surface morphology will be explained. Table 2 shows the growth temperature 700 (℃) and the gas flow rate 1500 (cc・
Wafers F62 and F89 grown with constant min -1 )
It shows the results of measuring the areal density of growth mounds, the diameter of growth mounds, and the areal density of hirotsuku in F95 and F95.

【表】 膜厚パラメータFhが最少のウエハーF62は成長
丘、ヒロツクともに抑制され良好な表面モホロジ
ーが得られるのに対し、膜厚パラメータFhが最
も大きいウエハーF89は成長丘の直径、面密度が
非常に抑制されるものの、ウエハーの中央から距
離xが大きくなるにつれて無数の小さなヒロツク
が存在し、通常のウエハーとは異なる表面モホロ
ジーが観察された。また、膜厚パラメータFh
比較的大きいウエハーF95は成長丘およびヒロツ
クがともに多くその表面モホロジーは悪化してい
る。 上述のように、ウエハーF89及びF95はともに
表面モホロジーが悪いため使用できない。従つ
て、表1に記載したウエハーで使用できるのはウ
エハーF62,F97,F99及びF105となり、その時
の膜厚パラメータFhの最大値が0.18(mm-1)とな
る。 ここでF105はInCl原料ガスの分圧が小さいた
め、成長速度が遅くなり成長に時間がかかる点を
考慮してこれを除外すれば、F99,F97,F62が
好ましい実施例となる。これら実施例では、InCl
原料ガスの分圧とPH3原料ガスの分圧とは実施例
毎に同等であり、分圧は表1からそれぞれ10/280
0,7/2800,7/1500である。7/1500はほぼ13/2800
に相当するので、分圧比は7/1500から7/2800まで
が好ましい範囲となる。 第5図は、第4図に示した膜厚パラメータFh
と原料ガスの分圧(族のInClガスの分圧)との
関係と、さらにこのような表面モホロジーの関係
もわかるように、同一図面上に前記の成長丘の直
径、成長丘の面密度及びヒロツクの面密度の大き
さをそれぞれ右端の単位0.5mm、10個cm-2,10個
cm-2で斜線の長さとして図示したものである。同
図からも明らかなように、膜厚パラメータFh
最少になれば、図示された各斜線の長さも短くな
つて表面モホロジーも良好になり、優れた半導体
層を提供することができる。 また、膜厚パラメータFhの実験結果と同様に、
InP/InGaAsP/InGaAsといつた多層膜をヘテ
ロエピタキシヤル成長により成長させても、表2
とほぼ同一の実験結果が得られた。 (発明の効果) 以上のように、本発明は従来別々の観点から対
策を施していた膜厚の均一性と表面モホロジーの
改善を、原料ガスの分圧を変化させるというひと
つの手法で行い得たものであり、高品質のウエハ
ーが容易に得られることから作製の効率はもちろ
んのこと、デバイス作製時の歩留りの改善がで
き、さらに素子長の長いデバイスをも作製でき、
その効果は極めて大である。
[Table] Wafer F62, which has the smallest film thickness parameter F h , suppresses both growth hills and hills and obtains good surface morphology, whereas wafer F89, which has the largest film thickness parameter F h , has a large growth hill diameter and areal density. However, as the distance x increased from the center of the wafer, countless small hills were present, and a surface morphology different from that of a normal wafer was observed. Furthermore, wafer F95, which has a relatively large film thickness parameter F h , has many growth hills and hills, and its surface morphology is deteriorated. As mentioned above, both wafers F89 and F95 cannot be used due to poor surface morphology. Therefore, among the wafers listed in Table 1, wafers F62, F97, F99, and F105 can be used, and the maximum value of the film thickness parameter F h at that time is 0.18 (mm -1 ). Here, if F105 is excluded because the partial pressure of the InCl raw material gas is small, the growth rate is slow and the growth takes time, and if this is excluded, F99, F97, and F62 are preferred examples. In these examples, InCl
The partial pressure of the raw material gas and the partial pressure of the PH 3 raw material gas are the same for each example, and the partial pressures are each 10/280 from Table 1.
0.7/2800, 7/1500. 7/1500 is almost 13/2800
Therefore, the preferable range of the partial pressure ratio is from 7/1500 to 7/2800. Figure 5 shows the film thickness parameter F h shown in Figure 4.
The diameter of the growth hill, the areal density of the growth hill, and the surface density of the growth hill are plotted on the same drawing so that you can see the relationship between The size of the areal density of Hirotsuku is 0.5 mm, 10 pieces cm -2 , 10 pieces, respectively on the right side.
It is shown as the length of the diagonal line in cm -2 . As is clear from the figure, when the film thickness parameter F h is minimized, the length of each diagonal line shown in the figure becomes short, the surface morphology becomes good, and an excellent semiconductor layer can be provided. Also, similar to the experimental results for the film thickness parameter F h ,
Even if a multilayer film such as InP/InGaAsP/InGaAs is grown by heteroepitaxial growth, Table 2
Almost the same experimental results were obtained. (Effects of the Invention) As described above, the present invention makes it possible to improve film thickness uniformity and surface morphology, which have conventionally been taken from separate viewpoints, with a single method of changing the partial pressure of the raw material gas. Since high-quality wafers can be easily obtained, it is possible to not only improve the manufacturing efficiency but also improve the yield during device manufacturing, and it is also possible to manufacture devices with long element lengths.
The effect is extremely large.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は気相成長の一般的特性を説明するため
の略図、第2図,第3図,第4図及び第5図は本
発明の原理を説明するための特性図である。
FIG. 1 is a schematic diagram for explaining the general characteristics of vapor phase growth, and FIGS. 2, 3, 4, and 5 are characteristic diagrams for explaining the principle of the present invention.

Claims (1)

【特許請求の範囲】[Claims] 1 InP基板上にInP単層膜あるいはIn1-aGax
As1-yPy系(0x1,0y1)の多層膜
を常圧下で成長させる化合物半導体層のバイドラ
イド気相成長法において、気体の流量一定の条件
の下で結晶成長させるべき該化合物半導体層の各
原料ガスの分圧は7/1500から7/2800までの範囲で
同等になるようにして成長させることを特徴とす
る気相成長方法。
1 InP single layer film or In 1-a Ga x on InP substrate
In the bydride vapor phase epitaxy of a compound semiconductor layer in which a multilayer film of the As 1-y P y system (0x1,0y1) is grown under normal pressure, the crystal growth of the compound semiconductor layer is performed under conditions of a constant gas flow rate. A vapor phase growth method characterized by growing the partial pressure of each source gas at the same level in the range of 7/1500 to 7/2800.
JP4285184A 1984-03-08 1984-03-08 Gaseous phase growth method Granted JPS60191096A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP4285184A JPS60191096A (en) 1984-03-08 1984-03-08 Gaseous phase growth method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP4285184A JPS60191096A (en) 1984-03-08 1984-03-08 Gaseous phase growth method

Publications (2)

Publication Number Publication Date
JPS60191096A JPS60191096A (en) 1985-09-28
JPH0360800B2 true JPH0360800B2 (en) 1991-09-17

Family

ID=12647513

Family Applications (1)

Application Number Title Priority Date Filing Date
JP4285184A Granted JPS60191096A (en) 1984-03-08 1984-03-08 Gaseous phase growth method

Country Status (1)

Country Link
JP (1) JPS60191096A (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4818357A (en) * 1987-05-06 1989-04-04 Brown University Research Foundation Method and apparatus for sputter deposition of a semiconductor homojunction and semiconductor homojunction products created by same
JPH05291140A (en) * 1992-04-09 1993-11-05 Fujitsu Ltd Growth method of compound semiconductor thin film

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58208200A (en) * 1982-05-31 1983-12-03 Nippon Telegr & Teleph Corp <Ntt> Selective vapor phase epitaxial growth method of inp

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS58208200A (en) * 1982-05-31 1983-12-03 Nippon Telegr & Teleph Corp <Ntt> Selective vapor phase epitaxial growth method of inp

Also Published As

Publication number Publication date
JPS60191096A (en) 1985-09-28

Similar Documents

Publication Publication Date Title
US6617235B2 (en) Method of manufacturing Group III-V compound semiconductor
JPH0360800B2 (en)
Yamaguchi et al. Growth of Highly Crystalline GaN at High Growth Rate by Trihalide Vapor‐Phase Epitaxy
JPH10182282A (en) Vapor phase growth system and crystal growth method
JPH0529218A (en) Organic metal molecular beam epitaxial growth method
JP3242571B2 (en) Vapor growth method
JP3424507B2 (en) Method of manufacturing gallium nitride based compound semiconductor thin film
JP3052269B2 (en) Vapor phase growth apparatus and growth method thereof
JPS5922319A (en) Vapor growth of 3-5 group semiconductor
JPH11126754A (en) Gaseous-phase growing method of organic metal
JPS5984417A (en) Iii-v family mixed crystalline semiconductor device
JPH0330339A (en) Method for growing crystal of ii-vi compound semiconductor
JPH01173708A (en) Semiconductor device
JPH02241030A (en) Zinc diffusion method
JPH0684803A (en) Vapor phase epitaxial growth device
JPH08130187A (en) Semiconductor vapor phase growth apparatus and semiconductor vapor phase growth method
JPH06314658A (en) Vapor growing apparatus
JPH0594949A (en) Semiconductor vapor growth device
JPH11329976A (en) Iii group nitride crystal growing device
JPH0222814A (en) Manufacture of compound semiconductor device
JPS63227009A (en) Vapor growth method
JPH05304098A (en) Quantum dot crystal growth device and manufacture of semiconductor device
JPS6119120A (en) Manufacture of compound semiconductor substrate
JPS5945216B2 (en) Group 3-5 compound semiconductor crystal growth equipment
JPS59161015A (en) Vapor growth of gallium arsenide