JP2004111819A - Heat treatment method and heat treatment apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately apply heat treatment without being affected by an error in measurement of a temperature measuring means. <P>SOLUTION: The heat treatment method is used to apply heat treatment by dividing a wafer into a plurality of zones and controlling each of zones individually. One or a plurality of pieces of data among process data such as electric power, temperature or the like during heat treatment are sampled in realtime and stored, and the data is statistically processed for each of steps, and then a present statistical value during the heat treatment is compared with a past statistical value of the same process data or another related statistical value to grasp the present process state so that the heat treatment is controlled based on the obtained result. In order to implement the method, the heat treatment apparatus is provided with a means for measuring electric power or the like, a storage means of data, and a control means wherein a statistically processed value of the data stored in the storage means is compared with the past statistical value or related another statistical value to grasp the present process state and the apparatus is controlled on the basis of the result. <P>COPYRIGHT: (C)2004,JPO

Description

ここで、
Ｐ_{Ｃ０ＡＶＥ}：中心部加熱手段への供給電力平均
Ｐ_{Ｆ０ＡＶＥ}：前部加熱手段への供給電力平均
Ｐ_{Ｓ０ＡＶＥ}：側部加熱手段への供給電力平均
Ｐ_{Ｂ０ＡＶＥ}：後部加熱手段への供給電力平均
Ｐ_{Ｃ０ＳＴＤ}：中心部加熱手段への供給電力標準偏差
Ｐ_{Ｆ０ＳＴＤ}：前部加熱手段への供給電力標準偏差
Ｐ_{Ｓ０ＳＴＤ}：側部加熱手段への供給電力標準偏差
Ｐ_{Ｂ０ＳＴＤ}：後部加熱手段への供給電力標準偏差
Ｎ　　　：サンプリング点数
（ｊ）　：各サンプリング点
電力については、各ゾーン間の比率が重要であるため、中央部に対する平均値の比率もそれぞれη_Ｆ０、η_Ｂ０、η_Ｓ０として算出し、保存する。
【００６１】
η_Ｆ０＝Ｐ_{Ｆ０ＡＶＥ}／Ｐ_{Ｃ０ＡＶＥ}（９）
η_Ｓ０＝Ｐ_{Ｓ０ＡＶＥ}／Ｐ_{Ｃ０ＡＶＥ}（１０）
η_Ｂ０＝Ｐ_{Ｂ０ＡＶＥ}／Ｐ_{Ｃ０ＡＶＥ}（１１）
一方、レシピ決定時には、これらの値に対して、いくつかの許容範囲を事前に設定する。例えば、電力修正手段２０Ｃ、２０Ｆ、２０Ｓ、２０Ｂに対する電力修正指示を自動的に設定するかどうかを判定する上下限値△_Ｃや、作業者に対してアラームを発生させるかどうかを判定する上下限値△_Ａを事前に設定する。なおここでは、説明を簡単にするために、上下限値△_Ｃ、△_Ａとして、すべてのゾーンに対して同じ値を設定して使用するが、この設定値は、諸条件に応じて、各ゾーンで全て同じ値にしたり、個別の値にしたりする。大きさも諸条件に応じて異なる。
【００６２】
調整終了後、通常操業に切り替えて操業を開始する。通常操業時の本熱処理装置の処理フローを図８に示す。
【００６３】
Ｓ５までの処理内容は、上述した図７とほぼ同様であり、ステップ毎に各電力の統計値Ｐ_ＣＡＶＥ，Ｐ_ＣＳＴＤ，Ｐ_ＦＡＶＥ，Ｐ_ＦＳＴＤ，Ｐ_ＳＡＶＥ，Ｐ_ＳＳＴＤ，Ｐ_ＢＡＶＥ，Ｐ_ＢＳＴＤを上記式（１）〜（８）と同様に計算する。
【００６４】
ここで、
Ｐ_ＣＡＶＥ：中心部加熱手段への供給電力平均
Ｐ_ＦＡＶＥ：前部加熱手段への供給電力平均
Ｐ_ＳＡＶＥ：側部加熟手段への供給電力平均
Ｐ_ＢＡＶＥ：後部加熱手段への供給電力平均
Ｐ_ＣＳＴＤ：中心部加熱手段への供給電力標準偏差
Ｐ_ＦＳＴＤ：前部加熱手段への供給電力標準偏差
Ｐ_ＳＳＴＤ：側部加熱手段への供給電力標準偏差
Ｐ_ＢＳＴＤ：後部加熱手段への供給電力標準偏差
次に前述と同様に中央部に対する平均値の比率をそれぞれ、η_Ｆ、η_Ｂ、η_Ｓとして（１２）〜（１４）で算出する。
【００６５】
η_Ｆ＝Ｐ_ＦＡＶＥ／Ｐ_ＣＡＶＥ　　　　　　（１２）
η_Ｓ＝Ｐ_ＳＡＶＥ／Ｐ_ＣＡＶＥ　　　　　　（１３）
η_Ｂ＝Ｐ_ＢＡＶＥ／Ｐ_ＣＡＶＥ　　　　　　（１４）
次に、統計量比較検定処理Ｓ６において、同一製造指示、同一レシピ、同一品種のゾーン間電力比率η_Ｆ０とη_Ｆ、η_Ｓ０とη_Ｓ、η_Ｂ０とη_Ｂを比較する。なおここでは、説明を簡素化するために、η_Ｆのみについて説明する。
【００６６】
η_Ｆ０−△_Ｃ＜η_Ｆ＜η_Ｆ０＋△_Ｃであれば、調整時と同じ状態で処理されていると判断し、特に何もせず、Ｓ３の処理に戻る（Ｓ７）。
【００６７】
η_Ｆ０−△_Ｃ＞η_Ｆ、かつη_Ｆ０−△_Ａ＜η_Ｆ
または、
η_Ｆ０＋△_Ｃ＜η_Ｆ、かつη_Ｆ０＋△_Ａ＞η_Ｆ
であれば修正する。電力修正ゲインを決定し（Ｓ８）、自動修正をする場合（Ｓ１０）は、電力修正手段２０Ｆに対して、η_Ｆ０に近づくように、修正指令を出す（Ｓ１１）。なお、この場合、各温度制御手段１４Ｃ、１４Ｆ、１４Ｓ、１４Ｂに対して、電力がη_Ｆ０になるように目標温度Ｔ_Ｆを修正するようにしてもよい。
【００６８】
Ｓ１０において自動修正をしない場合は、作業者に対してガイダンスをアラーム表示機１９に表示させ（Ｓ１２）、変更自身は作業者に行わせる。
【００６９】
また、η_Ｆ０−△_Ａ＞η_Ｆまたは、η_Ｆ０＋△_Ａ＜η_Ｆであれば、アラーム表示機１９でアラームを発生させると共に（Ｓ９）、作業者に対処を依頼する旨を表示させる（Ｓ１２）。
【００７０】
なお、発生手段１７で電力修正ゲインを決定し（Ｓ８）、自動修正を行わなかった場合、又はアラームを発生させた（Ｓ９）場合は、作業者にガイダンスを出力し、作業者が対処完了するまで、処理を停止する。
【００７１】
一方、分散の場合は次のように処理する。分散については、特に、過去データとの関連が重要であり、各ゾーン毎に個別に比較を行う。通常、一定温度でエピタキシャル成長している場合には、分散は非常に小さいが、ガス流量の変動（ガス流量測定手段がない場合）や、予期せぬ装置の異常が発生した場合には、分散が大きくなる。
【００７２】
分散値の算出は、石英で構成されたチャンバ外から放射温度計で温度を測定する加熱装置においては、特に重要となる。例えば、ウェーハの水平性が失われた場合、放射温度計の視野の関係からウェーハ回転に応じた温度の変動が検出される。温度制御手段では、上記変動を抑制するために、加熱手段に対しての周期的な供給電力の変更指示を出す。この結果、供給電力の分散が大きくなる。この分散の管理を行っていれば、ウェーハの水平性の監視が容易にでき、作業者に対して装置再組み立ての指示などができる。
【００７３】
また、統計量として、ＦＦＴなどの周波数解析結果を使用すれば、サセプタの回転周期と、電力変動の周期が同じであることから、上記同様にウェーハの水平性の監視が可能となる。
【００７４】
分散についての管理方法は、上述した電力比率と同様である。
【００７５】
各加熱手段に供給する電力の分散を
Ｐ_ＣＳＴＤ
Ｐ_ＦＳＴＤ
Ｐ_ＳＳＴＤ
Ｐ_ＢＳＴＤ
とし、それに対して、許容範囲を設定する。例えば、作業者に対してアラームを発生させるときの上下限値を△_ＡＳＴＤとする。なおここでは、説明を簡単にするために、△_ＡＳＴＤとしてすべてのゾーンで同じ値を使用するが、この△_ＡＳＴＤは諸条件に応じた値となる。即ち、△_ＡＳＴＤは、各ゾーンで同じ値であったり、個々に異なる値であったりする。大きさも諸条件に応じて異なる。
【００７６】
ここで、Ｐ_{Ｆ０ＳＴＤ}とＰ_ＦＳＴＤとを比較し、Ｐ_{Ｆ０ＳＴＤ}−△_ＡＳＴＤ＜Ｐ_ＦＳＴＤ＜Ｐ_{Ｆ０ＳＴＤ}＋△_Ａであれば、調整時と同じ状態で処理されていると判断し、特に何もしない。
【００７７】
Ｐ_Ｆ０−△_Ａ＞Ｐ_Ｆまたは、Ｐ_Ｆ０＋△_Ａ＜Ｐ_Ｆであれば、アラームを出し、作業者にガス流量の変動チェックもしくは、サセプタの設置状況のチェックなど、装置上の異常チェックを依頼する。
【００７８】
このとき、ガス流量についても同様の統計値の処理をしている場合には、上記電力の分散値と、ガス流量の分散値を用いれば、電力の分散値変化がガス流量の変動の結果であることが明らかとなるため、作業者に対しては、ガス流量の変動確認を依頼する。
【００７９】
この結果、測定誤差が生じた場合でも、その誤差の影響を受けずに、正確な熱処理をすることができるようになる。
【００８０】
［実施例］
図９に本加熱装置をエピタキシャル成長装置として使用した場合の結果を示す。本例では、統計量として電力比率η_Ｆを管理した結果、効果があった例を示す。
【００８１】
ここで、８バッチ目に▲１▼熱電対を交換し、１２バッチ目に▲２▼制御（管理）をスタートさせた。ここで、熱電対を交換した際に、設置不良もしくは、熱電対の誤差によってＦｒｏｎｔ温度が低く測定されたため、前部温度制御手段が、前部加熱手段により多くの電力供給指令をだし、その結果、前部がより加熱され、外周部膜厚が増加している。ここで、▲２▼で制御ＯＮにすると、η_Ｆがアラーム発生の管理範囲ＵＣＬ２を超えているため、即、オペレータに対して“加熱装置確認”アラームおよび、η_Ｆのチャートを表示した。
【００８２】
その結果、オペレータが▲１▼で交換した熱電対に問題ありと判断し、新しい熱電対をセットした。その結果、η_Ｆは制御開始管理範囲ＵＣＬ１、ＬＣＬ１間に入り、膜厚分布も減少した。
【００８３】
次に、２２バッチ目にη_ＦがＵＣＬ１を超えたため、発生手段１７は、前部電力修正手段２０Ｆに対して約５％の補正を加えることで、次バッチから即、膜厚分布が低下している。
【００８４】
通常、同一品種を連続して操業する場合、複数のバッチ終了まで膜厚などの品質を測定しない場合がある。本実施例の場合１８バッチから２５バッチまでは、同一品種であり通常膜厚測定を実施しない。そのため、本発明の加熱装置でない場合は、２５バッチ終了まで、２２バッチで悪化した膜厚分布の変化に気づかず、多くの不良を出していた可能性がある。しかしながら、本発明によれば、η_Ｆを管理することで、このチャンバの変化を捉え、前部への供給電力を下げることで、膜厚分布の悪化は、２２バッチの１バッチのみと最低限で抑えることができた。
【００８５】
［変形例］
（１）　上記実施形態では、電力の統計値を中心に説明したが、統計処理するデータとしては、これに限らず、他のデータを統計処理して使用してもよい。この場合も、程度には差があるが、上記実施形態とほぼ同様の作用、効果を得ることができる。
【００８６】
（２）　上記実施形態では、プロセスデータとしては、熱処理中の加熱手段１３Ｃ、１３Ｆ、１３Ｓ、１３Ｂの供給される電力、ウェーハ温度、ガス流量、ガス濃度、ガス温度を例に説明したが、これに限らず、統計処理できる全てのプロセスデータを使用することができる。この場合も、程度には差があるが、上記実施形態とほぼ同様の作用、効果を得ることができる。
【００８７】
（３）　関連する他の統計値としては、データ間の関係を表す、共分散値や相関計数等がある。
【００８８】
【発明の効果】
以上詳述したように、本発明に係る熱処理方法および熱処理装置によれば、次のような効果を奏することができる。
【００８９】
測温手段に測定誤差が生じる等の、実操業中に起こりうるプロセス状態の変化を確実に捉えられることができ、その誤差等の影響を受けずに、正確な熱処理をすることができる。例えば、一定のプロセス状態が維持でき、薄板状加熱対象物の面内温度分布を一定に維持することができる。
【００９０】
その結果、例えば、エピタキシャル成長炉においては、温度分布の不具合から生じる膜厚分布や比抵抗分布の悪化のない、また、熱応力によって発生するスリップが発生しない熱処理が可能となる。
【図面の簡単な説明】
【図１】本発明の実施形態に係る熱処理装置を示すブロック図である。
【図２】エピタキシャル成長装置における測温点の例を模式的に示した説明図である。
【図３】測温手段、ランプ、温度制御装置の配設例を示すブロック図である。
【図４】熱電対の劣化による測温誤差の例を示表である。
【図５】加熱手段への供給電力の変化と各部の実績温度との関係を示すグラフである。
【図６】膜厚分布の径時変化を示すグラフである。
【図７】レシピ決定時の処理を示すフローチャートである。
【図８】通常操業時の処理を示すフローチャートである。
【図９】具体的実施例での膜厚や温度差等の変化を示すグラフである。
【符号の説明】
１１：熱処理装置、１２Ｃ，１２Ｆ，１２Ｓ，１２Ｂ：測温手段、１３Ｃ，１３Ｆ，１３Ｓ，１３Ｂ：加熱手段、１４Ｃ，１４Ｆ，１４Ｓ，１４Ｂ：温度制御手段、１５：データ保存手段、１６：統計解析手段、１７：ゲイン決定およびアラーム発生手段、１８Ｃ，１８Ｆ，１８Ｓ，１８Ｂ：電力測定手段、１９：アラーム表示機、２０Ｃ，２０Ｆ，２０Ｓ，２０Ｂ：電力修正手段。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a heat treatment method and a heat treatment apparatus for heat-treating a thin plate-shaped heating object, and in particular, to a heat treatment method and a heat treatment method suitable for a semiconductor treatment apparatus for performing heat treatment so as to always keep the in-plane temperature distribution of a semiconductor wafer constant. The present invention relates to a heat treatment apparatus.
[0002]
[Prior art]
2. Description of the Related Art In a process of processing a semiconductor wafer (hereinafter, simply referred to as a wafer) and the like, there are many processes that require heat treatment (heating) on the wafer. A heat treatment apparatus (semiconductor processing apparatus) for such a process has heating means. For example, an epitaxial growth apparatus (growth furnace) has a wafer heating apparatus (heat treatment apparatus) using an infrared lamp or the like. In epitaxial growth, it is necessary to control the in-plane temperature distribution of the wafer in order to optimize the wafer film thickness distribution and resistivity distribution and to control crystal defects such as slip due to the temperature distribution of the wafer. Therefore, the in-plane temperature distribution of the wafer is controlled by the heat treatment apparatus.
[0003]
[Patent Document 1]
JP-A-5-291169
As an example of controlling the in-plane temperature distribution of a wafer, there is a “semiconductor manufacturing apparatus” described in Patent Document 1. In order to control the in-plane temperature distribution of the wafer, the above-described semiconductor manufacturing apparatus having a plurality of temperature measuring means (hereinafter, appropriately referred to as measuring means) and a plurality of heating means was developed, and the temperature was measured by each temperature measuring means. Based on the temperature, a method of controlling by a dedicated temperature control means has been used.
[0004]
FIG. 2 is an explanatory diagram schematically showing an example of a temperature measuring point in a typical apparatus having a heat treatment apparatus (for example, an epitaxial growth apparatus).
[0005]
In FIG. 2, a wafer 1 is held by a disk-shaped susceptor 2, and this susceptor 2 is held by a disk-shaped susceptor ring 3. As a result, the wafer 1 to be heated is located at a predetermined position in the chamber 4. For example, in the case of an epitaxial growth apparatus, the reaction gas 8 is introduced into and extracted from the chamber 4, and at this time, the reaction gas 8 flows along a substantially linear flow path from the inlet to the outlet.
[0006]
In the example of FIG. 2, the temperature measuring means is provided at several places. The position where the normal direction passing through the center of the wafer 1 intersects the back surface of the susceptor 2 is the temperature measuring point by the central temperature measuring means 5C. The position of the susceptor ring 3 near the inlet of the reaction gas is a temperature measuring point by the front temperature measuring means 5F. The position of the susceptor ring 3 near the outlet of the reaction gas is a temperature measuring point by the rear temperature measuring means 5B. Positions 5F and 5B are angularly separated by 180 degrees from the center of wafer 1. The position of the susceptor ring 3 which is angularly separated by ± 90 degrees from the positions 5F and 5B is a temperature measuring point by the side temperature measuring means 5S.
[0007]
Outputs from the respective temperature measuring means 5C, 5F, 5S and 5B are provided to corresponding temperature control devices 6C, 6F, 6S and 6B as shown in FIG. Each of the temperature control devices 6C, 6F, 6S, 6B controls a corresponding one of the lamps 7C, 7F, 7S, 7B as a corresponding heating means based on the temperature measurement output from each of the temperature measuring means 5C, 5F, 5S, 5B.
[0008]
That is, in such a heat treatment apparatus, a desired temperature profile (command temperature change) for each part of the wafer is previously instructed to each of the temperature control apparatuses 6C, 6F, 6S, and 6B. Each of the temperature control devices 6C, 6F, 6S, 6B controls the lamps 7C, 7F, 7S, 7B as heating means so as to have the designated temperature profile.
[0009]
Although FIGS. 2 and 3 intentionally describe the case where each of the temperature measuring means 5C, 5F, 5S and 5B is a thermocouple, the same applies to the case where the temperature measuring means is a radiation thermometer.
[0010]
Also, the control method does not necessarily require four zones as described above, and the same applies to, for example, the case of two or more zones at the center and side.
[0011]
[Problems to be solved by the invention]
In a heat treatment apparatus having a plurality of zones, how to use each heating means becomes a major problem. For example, a method of controlling and heating a corresponding heating means according to the output of each temperature measuring means for each zone (Japanese Patent Laid-Open No. 5-291169), and a method of controlling the heating means to keep the temperature difference between zones constant In JP-A-6-260426, stable heating cannot be performed unless the state of each temperature measuring means is always healthy. That is, the temperature measurement by the temperature measuring means must always be accurate.
[0012]
Since the temperature control device controls the measured value to be the command value, if the temperature measuring means in a certain zone measures the temperature low (or high), the temperature control device raises the measured value to the command value (or Therefore, the temperature of a certain zone is actually raised (or lowered) to a value equal to or higher than (or lower than) the command value. This heating affects not only the zone to be heated but also other zones. As a result, the in-plane temperature distribution of the wafer is different from the desired distribution, the specific resistance distribution and the film thickness distribution deteriorate, and further, This was one of the causes of quality defects such as slip.
[0013]
One of the causes that the thermocouple as the temperature measuring means does not measure an accurate temperature is a contact-type temperature measurement in principle, and is caused by a change in the contact state. Further, some heat treatment apparatuses require a high temperature of 1000 ° C. or more, and the thermocouple itself deteriorates to introduce a reaction gas or the like. For example, as shown in FIG. 4, the temperature is measured as low as 10 ° C. or more. There are cases.
[0014]
Further, the measurement accuracy (JIS standard) of the thermocouple itself is within about ± 0.3%. For example, at 1473K, even within the JIS standard, the temperature error is about ± 4.5 ° C. (1473 ( K) × 0.3 (%)), as shown in FIG. 4, there is a measurement error even in a completely new thermocouple, and this effect due to the replacement of the thermocouple in the heat treatment furnace cannot be ignored.
[0015]
On the other hand, even in temperature measurement using a radiation thermometer, since the temperature is measured from outside the quartz chamber through quartz, depending on the reaction gas introduced into the chamber 4 during the heat treatment, molecules adhere to the quartz glass surface, and as a result, As a result, the transmittance for the radiation light of the wavelength for temperature measurement decreases, and the measurement temperature changes.
[0016]
In addition, a thermocouple is generally used for calibration of the radiation thermometer, but since this thermocouple has the above-mentioned problem, if this management is neglected, the measurement temperature changes due to the error of the above-mentioned thermocouple. .
[0017]
The present invention has been made in view of the above points, and an object of the present invention is to provide a heat treatment method and a heat treatment apparatus capable of performing accurate heat treatment without being affected by a measurement error or the like of a temperature measuring unit.
[0018]
[Means for Solving the Problems]
In order to solve the above problem, the heat treatment method according to the first invention is a heat treatment method in which a thin plate-like heating object is divided into a plurality of zones, and each zone is individually controlled and heat-treated. One or more pieces of process data are collected and stored in real time, and statistical processing is performed for each step. The current statistical value during the heat treatment and the past statistical value of the same process data or the related statistical value are processed. The present invention is characterized in that the current process state is grasped by comparing with other statistical values, and control is performed based on the result.
[0019]
With the above-described configuration, by comparing the statistical value collected in real time and statistically processed with the past statistical value or other related statistical value and grasping the current process state, it is possible to determine whether the temperature measuring means is defective or measurement error. Accurate heat treatment can be performed without being affected by such factors. The process data during the heat treatment includes power, temperature, gas flow rate, gas concentration, gas temperature, and the like.
[0020]
A heat treatment method according to a second invention is the heat treatment method according to the first invention, wherein one or both of an average value and a variance value are used as the statistical value, and a power average value ratio between the zones is used for each step. Is maintained constant, and the variance value is managed to a fixed value or less.
[0021]
With the above configuration, if the average power ratio or the variance between the zones fluctuates greatly beyond the limit, it can be determined that an abnormality such as a measurement error of the temperature measuring means has occurred, and appropriate measures must be taken. Can be.
[0022]
A heat treatment apparatus according to a third aspect of the present invention is a heat treatment apparatus that includes a heating unit that is divided into a plurality of zones and is individually controlled for each zone to heat a thin plate-like heating target. Power measuring means for measuring electric power, temperature measuring means provided in each of the above zones or a part of the zones and measuring the temperature of the zone, and a flow rate of gas introduced to the surface of the thin plate-shaped heating object. Gas flow rate measuring means for measuring, gas concentration measuring means for measuring the concentration of gas introduced to the surface of the thin plate-shaped heating object, or gas for measuring the temperature of gas introduced to the surface of the thin plate-shaped heating object One or more of the temperature measurement means, storage means for storing the measurement results of the above means, and statistical processing of the data stored in the storage means; You To grasp the current process state by comparing the other statistics, characterized in that a control means for controlling based on the result.
[0023]
With the above-described configuration, the measurement value obtained by the power measurement unit or the like is collected in real time, and the measurement result is stored in the storage unit. The control unit statistically processes the data stored in the storage unit, compares the data with past statistical values and the like, grasps the current process state, and performs control based on the result. Thus, accurate heat treatment can be performed without being affected by a failure of the temperature measuring means, a measurement error, or the like.
[0024]
The heat treatment apparatus according to a fourth aspect of the present invention is the heat treatment apparatus according to the third aspect, wherein one or both of the average value and the variance value are used as the statistic, and the power average value ratio between the zones is used for each step. Is maintained constant, and the variance value is managed to a fixed value or less.
[0025]
With the above configuration, if the average power ratio or the variance between the zones fluctuates greatly beyond the limit, it can be determined that an abnormality such as a measurement error of the temperature measuring means has occurred, and appropriate measures must be taken. Can be.
[0026]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, a heat treatment method and a heat treatment apparatus according to an embodiment of the present invention will be described. FIG. 1 is a block diagram showing a heat treatment apparatus of the present embodiment, FIG. 5 is a graph showing a relationship between a change in power supplied to each heating means and an actual temperature of each part, and FIG. FIG. 7 is a flowchart showing a process at the time of recipe determination, FIG. 8 is a flowchart showing a process at the time of normal operation, and FIG. 9 is a graph showing a film thickness and a temperature difference in a specific embodiment. It is a graph which shows a change.
[0027]
[Heat treatment method]
First, a heat treatment method according to the present embodiment will be described.
[0028]
The heat treatment method according to the present embodiment enables stable temperature control by always keeping the process state of the heat treatment apparatus that heat-treats the thin plate-like heating object constant.
[0029]
First, the balance of the amount of power supply to the heating means will be described below. Here, as a device for performing the heat treatment method, a device having substantially the same configuration as a conventional heat treatment device is used. That is, temperature measuring means 5C, 5F, 5S, 5B, which are thermocouples, temperature controllers 6C, 6F, 6S, 6B and lamps 7C, 7F, 7S, 7B are arranged as shown in FIGS. The description will be made on the assumption that the wafer 1 as the thin plate-like heating target is heated. In addition, the wafer temperature referred to below does not mean that the temperature of the wafer is directly measured, but it is considered that the temperature of the wafer is indirectly measured. Therefore, the temperature measuring means 5C, 5F, 5S, 5B Is expressed as a wafer temperature.
[0030]
First, in a state where the susceptor 2, the susceptor ring 3, and the temperature measuring means 5C, 5F, 5S, and 5B, which are thermocouples, are considered to be sufficiently thermally coupled, a normal recipe (wafer temperature for each step, introduction Determine the gas flow rate, gas concentration, etc.) and start operation. In the following, for the sake of simplicity, a step of performing epitaxial growth will be described as an example.
[0031]
After the recipe is determined, in the step of supplying Si gas and performing epitaxial growth (deposition step), the wafer temperature in each zone is set to T_C, T_F, T_B, T_SA power command is issued by each control means (here, control means for performing PI control) to the lamps 7C, 7F, 7S, and 7B, which are heating means in each zone, so as to reach the target temperature. put out.
[0032]
FIG. 5A shows the average value of the difference between the power supplied to the lamps 7F, 7S, and 7B in the deposition step when the recipe and the production quality are equal, that is, when the heat treatment (epitaxial growth) is performed with all the setting conditions equal. (The power supplied to 7C is constant, which is indicated by the difference from the power supplied to the central part.) In this way, the operation was performed with exactly the same settings, and as shown in FIG. 5B, the temperature measurement values of each part (the temperature at the center was always constant. In order to clarify the difference, the power supplied to the lamp 7F for heating the front part of the wafer rises and the side part of the wafer is heated even though the average value of the difference from the center temperature is constant). It can be seen that the power supplied to the lamp 7S has decreased. In FIG. 5, a cassette containing a plurality of wafers 1 is defined as one batch. Hereinafter, the same applies.
[0033]
The above phenomenon is caused by a change in the contact state between the thermocouple as the temperature measuring means 5F for measuring the front part of the wafer in FIG. 2 and the point to be measured and the deterioration of the thermocouple shown in FIG. This is generated because the power supplied to the lamp 7F is increased so that the temperature of the front temperature control unit 6F becomes equal to the set value.
[0034]
In addition, as a result, the temperature of the front part of the wafer rises, the influence reaches the side part, and the temperature of the side part rises, so that the power supplied to the side part decreases. In this way, despite the fact that the temperature of each part is measured to be constant, the phenomenon that the power supplied to each corresponding part changes is confirmed not only in the front part but also in each part, and is independent for each zone. When using independent temperature control devices 6C, 6F, 6S, and 6B using the temperature measuring means 5C, 5F, 5S, and 5B, and when using a plurality of temperature measuring means 5C, 5F, 5S, and 5B, Is an unavoidable problem.
[0035]
FIG. 6 is a diagram showing a difference between the central part film thickness and the peripheral part film thickness of the wafer 1 after the epitaxial growth. (1), (2), and (3) in FIG. 6 are wafers manufactured at arrows (1), (2), and (3) in FIG. As shown in FIG. 6, the outer peripheral film thickness of (3) is larger than that of (1).
[0036]
The growth rate in epitaxial growth increases as the temperature increases. From this, the film thickness distribution in FIG. 6 shows that the temperature near the supply part (front part) where the gas concentration of the reaction gas is high is increased by the increase in the power supply to the lamp 7F as the heating means in the front part. This is considered to be a result of the epitaxial growth being further promoted in the outer peripheral portion.
[0037]
As is clear from this example, since the temperature measurement means 5C, 5F, 5S, and 5B have an unavoidable temperature measurement error, the temperature distribution control relying only on the temperature measurement means 5C, 5F, 5S, and 5B. Has limitations. Therefore, in order to maintain a constant in-plane distribution of wafer quality (film thickness, specific resistance), attention should be paid to energy (electric power, etc.) supplied to each zone having a close relationship with temperature. It is important to keep the power ratio supplied to each zone constant. This is because in the case of power measurement, a highly accurate measured value can be obtained without problems such as poor contact that occurs in the case of temperature measurement. That is, in the case of temperature measurement, as described above, there is a possibility that an error may occur due to the temperature measuring means, but in the case of power measurement, a correct measurement value can be obtained from the controller that controls the heating device. . Also, even when it is not possible to obtain from the controller, it is possible to easily obtain a correct measured value by connecting a known various power measuring device in front of the heating device.
[0038]
Although the deposition step has been described above, the heating step will also be briefly described.
[0039]
In the heat treatment furnace, there is always a step of raising and lowering the temperature in order to perform a predetermined heat treatment. In particular, when the temperature is raised from a low temperature (for example, 800 ° C.) to a high temperature (for example, 1100 ° C.), unnecessary thermal stress is generated in the wafer depending on the heating profile, which may greatly affect the quality of a heating object. is there. For example, in an epitaxial growth furnace, crystal dislocation called a slip occurs, and the occurrence may cause a loss of commercial value.
[0040]
This heating profile may vary greatly depending on the case where the temperature before the start of heating is different, or changes in the supplied gas temperature or gas flow rate. For this reason, even if a fixed set value is given, the temperature does not always rise in the same manner.
[0041]
In this case, the temperature control means determines a heating profile in advance and gives a command to the heating means to realize the heating profile, but if the actual wafer temperature at the start of heating is different, the temperature is increased. The power supply command value to the heating means differs depending on the difference. For example, if the actual wafer temperature at the start of heating is lower than the predetermined initial value of the heating profile, it is necessary to supply more power than the initial value until reaching the initial value. is there. As described above, by managing the power supply amount, the actual initial temperature can be managed. Similarly, since PI control or the like is performed based on the current temperature in order to satisfy the temperature rising profile, a temperature change during the temperature rising can be managed using the power supply amount. Further, since the gas temperature, the gas flow rate, and the gas concentration also cause the temperature change, by monitoring the gas temperature, the gas flow rate, and the gas concentration from the gas supply unit, the temperature change due to the gas can be easily managed.
[0042]
That is, in the temperature control by the heat treatment apparatus, it cannot be said that the heat treatment is always performed with the same temperature profile even if the recipe is the same. Therefore, data of each zone temperature, power, introduced gas temperature, gas flow rate and, if possible, gas concentration during the process are collected at a fixed sampling cycle, and each statistic at each step is compared with past statistic values. Further, the state of the heat treatment apparatus in the current step can be completely grasped by comparing the ratio with other related statistical values. This makes it possible to diagnose whether the same process is performed by the same recipe, that is, whether the heat treatment is performed with the same in-plane temperature distribution, and the same process can be maintained by performing control based on the diagnosis.
[0043]
In this way, multiple process data such as power, wafer temperature, gas flow rate, gas concentration, and gas temperature during heat treatment are collected and stored in real time, and statistical processing is performed for each step. Is compared with the past statistics of the same process data or other related statistics, the current process state is grasped, and based on the result, operation change instructions and the like are controlled.
[0044]
As a result, even when a measurement error occurs in the temperature measuring means, the in-plane temperature distribution of the wafer can be accurately maintained at a desired distribution without being affected by the error, and the specific resistance distribution and the film thickness distribution can be maintained. It is possible to perform an accurate heat treatment without quality defects such as deterioration and slip.
[0045]
[Heat treatment equipment]
Next, a heat treatment apparatus for performing the above heat treatment method will be described.
[0046]
FIG. 1 is a block diagram illustrating a configuration of a heat treatment apparatus according to the present embodiment. Hereinafter, a case where the heat treatment apparatus is applied to an epitaxial growth apparatus (epitaxial growth furnace) will be described as an example.
[0047]
The heat treatment apparatus 11 includes temperature measuring means 12C, 12F, 12S, 12B for measuring the temperature of each part of the wafer, heating means 13C, 13F, 13S, 13B for each part, and temperature measuring means 12C, 12F, 12S, 12B. In addition to the same structure as the conventional one such as temperature control means 14C, 14F, 14S, 14B for controlling each of the heating means 13C, 13F, 13S, 13B based on the detection result, data storage means 15, statistical analysis means 16, gain It has a decision and alarm generating means 17, power measuring means 18C, 18F, 18S, 18B, an alarm display 19, and power correcting means 20C, 20F, 20S, 20B.
[0048]
The data storage unit 15 is a unit for collecting and storing various kinds of data for performing statistical analysis in real time. The data to be stored include a normal recipe, a gas flow rate of a reaction gas (Si gas), a gas concentration, a gas temperature, temperatures detected by temperature measuring means 12C, 12F, 12S, and 12B, and power measuring means 18C, 18F, 18S, The electric power supplied to each of the heating means 13C, 13F, 13S, and 13B measured at 18B, the statistical value statistically processed based on the measured value, the processing information of the heating target, and the like. The information such as the gas flow rate, the gas concentration, and the gas temperature is information necessary for a heat treatment apparatus such as an epitaxial growth apparatus that requires introduction of a specific gas for the heat treatment. The type of data to be stored is selected according to the function of the processing device.
[0049]
The statistical analysis unit 16 performs a statistical process on the data stored in the data storage unit 15 in real time for each step, and performs a past statistical value of the same process data as the current statistical value or another related statistical value. This is an analysis device for comprehending the current process state by comparing with the above. Specifically, statistical processing is performed on the data stored in the data storage unit 15 by a method described later.
[0050]
The gain determining and alarm generating means (hereinafter referred to as “generating means”) 17 is a device for outputting an electric power correction gain, issuing an operation change instruction, and issuing an alarm generating instruction based on the result of the statistical analysis means 16. . The generating means 17 is connected to the alarm display 19, and via the power correcting means 20C, 20F, 20S, 20B, the respective temperature control means 14C, 14F, 14S, 14B and the respective power measuring means 18C, 18F. , 18S, 18B. In some cases, the generating unit 17 is directly connected to each of the temperature control units 14C, 14F, 14S, and 14B to correct the control target temperature of each of the temperature control units 14C, 14F, 14S, and 14B.
[0051]
The statistical analysis means 16 and the generation means 17 statistically process the data stored in the data storage means 15 and compare the processed statistics with past statistics or other related statistics to determine the current process state. And a control means for controlling the heat treatment apparatus 11 based on the result.
[0052]
Each of the power measuring means 18C, 18F, 18S, 18B is a device for measuring the power supplied to each of the heating means 13C, 13F, 13S, 13B. Specifically, each of the power measuring units 18C, 18F, 18S, and 18B calculates the power after the power from each of the temperature control units 14C, 14F, 14S, and 14B is corrected by the power correcting units 20C, 20F, 20S, and 20B. Measure.
[0053]
The alarm display 19 is a device for generating an alarm for a worker or displaying a warning.
[0054]
The power correction means 20C, 20F, 20S, and 20B are devices for correcting the power from each of the temperature control means 14C, 14F, 14S, and 14B based on the power correction gain from the generation means 17. Specifically, the power from each of the temperature control units 14C, 14F, 14S, and 14B is corrected based on the power correction gain output as an operation change instruction by the generation unit 17 based on the result of the statistical analysis unit 16. Then, it is supplied to each heating means 13C, 13F, 13S, 13B.
[0055]
Although not shown in FIG. 1, there is a thin plate-like heating target (wafer). Further, gas flow rate measuring means for measuring the flow rate of gas introduced to the surface of the wafer, gas concentration measuring means for measuring the concentration of gas introduced to the surface of the wafer, or introduced to the surface of the wafer Gas temperature measuring means for measuring the temperature of the gas.
[0056]
The statistical analysis means 16 specifically performs the following processing. That is, a statistical analysis value for each step in the recipe, for example, an average value and a variance value are calculated. Here, since the variance is calculated, data of at least three points is required.
[0057]
FIG. 7 shows a processing flowchart when a recipe is determined. First, when a recipe is determined, the wafer temperature T for each zone is determined so that a desired film thickness, film thickness distribution, specific resistance, and specific resistance distribution can be obtained in a state where defects such as slip are not generated._C, T_F, T_S, T_BIs determined (S1). Determined wafer temperature T_C, T_F, T_S, T_BP supplied to each heating means when processing a wafer with_C0(T), P_F0(T), P_S0(T), P_B0(T), measured temperature T_C0(T), T_F0(T), T_S0(T), T_B0(T), gas flow rate Q₀(T), gas concentration C_TCSO(T), gas temperature T_Q(T) is stored in the data storage unit 15. Here, (t) means time-series data.
[0058]
The statistical analysis unit 16 calculates a statistic D3 in the process S2 based on the process data D2 stored in the data storage unit 15, associates the statistic with the manufacturing instruction D1, and stores it in the data storage unit 15 ( S2).
[0059]
For example, the average value and the standard deviation are calculated by the equations (1) to (8) as the statistics of the power.
[0060]
(Equation 1)

here,
P_C0AVE： Average power supply to central heating means
P_F0AVE: Average power supply to front heating means
P_S0AVE: Average power supply to side heating means
P_B0AVE: Average power supply to rear heating means
P_C0STD： Standard deviation of power supply to center heating means
P_F0STD: Standard deviation of power supply to front heating means
P_S0STD： Standard deviation of power supply to side heating means
P_B0STD: Standard deviation of power supply to rear heating means
N: Number of sampling points
(J): Each sampling point
Since the ratio between the zones is important for the power, the ratio of the average value to the center is also η_F0, Η_B0, Η_S0Calculate as and save.
[0061]
η_F0= P_F0AVE/ P_C0AVE(9)
η_S0= P_S0AVE/ P_C0AVE(10)
η_B0= P_B0AVE/ P_C0AVE(11)
On the other hand, when the recipe is determined, some allowable ranges are set in advance for these values. For example, upper and lower limit values 判定 す る for determining whether to automatically set a power correction instruction to power correction units 20C, 20F, 20S, and 20B._CAnd upper and lower limit values to determine whether to generate an alarm for the operator._ASet in advance. In addition, here, in order to simplify the description, the upper and lower limit values △_C, △_AThe same value is set and used for all the zones, and this set value may be the same value for each zone or an individual value depending on various conditions. The size also varies depending on various conditions.
[0062]
After the adjustment, switch to normal operation and start operation. FIG. 8 shows a processing flow of the present heat treatment apparatus during normal operation.
[0063]
The processing contents up to S5 are almost the same as those in FIG. 7 described above._CAVE, P_CSTD, P_FAVE, P_FSTD, P_SAVE, P_SSTD, P_BAVE, P_BSTDIs calculated in the same manner as in the above equations (1) to (8).
[0064]
here,
P_CAVE： Average power supply to central heating means
P_FAVE: Average power supply to front heating means
P_SAVE: Average power supply to side ripening means
P_BAVE: Average power supply to rear heating means
P_CSTD： Standard deviation of power supply to center heating means
P_FSTD: Standard deviation of power supply to front heating means
P_SSTD： Standard deviation of power supply to side heating means
P_BSTD: Standard deviation of power supply to rear heating means
Next, as described above, the ratio of the average value to the center is η_F, Η_B, Η_SAre calculated in (12) to (14).
[0065]
η_F= P_FAVE/ P_CAVE(12)
η_S= P_SAVE/ P_CAVE(13)
η_B= P_BAVE/ P_CAVE(14)
Next, in the statistic comparison test processing S6, the power ratio η between zones of the same manufacturing instruction, the same recipe, and the same product type_F0And η_F, Η_S0And η_S, Η_B0And η_BCompare. Here, for simplicity of explanation, η_FOnly the following will be described.
[0066]
η_F0− △_C<Η_F<Η_F0+ △_CIf so, it is determined that the processing has been performed in the same state as that at the time of adjustment, and nothing is performed, and the process returns to S3 (S7).
[0067]
η_F0− △_C> Η_F, And η_F0− △_A<Η_F
Or
η_F0+ △_C<Η_F, And η_F0+ △_A> Η_F
If so, correct it. When the power correction gain is determined (S8) and automatic correction is performed (S10), η_F0A correction command is issued so as to approach (S11). In this case, the electric power is η for each of the temperature control means 14C, 14F, 14S, and 14B._F0Target temperature T_FMay be corrected.
[0068]
If automatic correction is not performed in S10, guidance is displayed to the operator on the alarm display 19 (S12), and the change itself is performed by the operator.
[0069]
Also, η_F0− △_A> Η_FOr η_F0+ △_A<Η_FIf so, an alarm is generated on the alarm display 19 (S9), and a message requesting the operator to take action is displayed (S12).
[0070]
When the power correction gain is determined by the generation means 17 (S8) and automatic correction is not performed or an alarm is generated (S9), guidance is output to the worker, and the worker completes the countermeasure. Until the processing is stopped.
[0071]
On the other hand, in the case of dispersion, processing is performed as follows. Regarding the variance, the relation with the past data is particularly important, and comparison is performed individually for each zone. Normally, when the epitaxial growth is performed at a constant temperature, the dispersion is very small. However, when the gas flow rate fluctuates (when there is no gas flow rate measuring means) or when an unexpected device abnormality occurs, the dispersion is very small. growing.
[0072]
The calculation of the dispersion value is particularly important in a heating device that measures the temperature with a radiation thermometer from outside the chamber made of quartz. For example, when the horizontality of the wafer is lost, the fluctuation of the temperature according to the rotation of the wafer is detected from the relation of the visual field of the radiation thermometer. The temperature control unit issues a periodic change instruction of the supply power to the heating unit in order to suppress the fluctuation. As a result, the variance of the supplied power increases. If the dispersion is managed, the horizontality of the wafer can be easily monitored, and the operator can be instructed to reassemble the apparatus.
[0073]
Further, if a frequency analysis result such as FFT is used as a statistic, since the rotation cycle of the susceptor and the cycle of power fluctuation are the same, the horizontality of the wafer can be monitored in the same manner as described above.
[0074]
The management method for dispersion is the same as the power ratio described above.
[0075]
Dispersion of power supplied to each heating means
P_CSTD
P_FSTD
P_SSTD
P_BSTD
And an allowable range is set for it. For example, the upper and lower limit values when an alarm is_ASTDAnd Note that, here, for simplicity of explanation, △_ASTDUse the same value in all zones as_ASTDIs a value corresponding to various conditions. That is, △_ASTDMay be the same value in each zone or may be different values individually. The size also varies depending on various conditions.
[0076]
Where P_F0STDAnd P_FSTDAnd P_F0STD− △_ASTD<P_FSTD<P_F0STD+ △_AIf so, it is determined that the processing has been performed in the same state as at the time of adjustment, and no particular operation is performed.
[0077]
P_F0− △_A> P_FOr P_F0+ △_A<P_FIf this is the case, an alarm is issued, and the operator is requested to perform an abnormality check on the apparatus, such as a change in the gas flow rate or a check on the installation status of the susceptor.
[0078]
At this time, if the same statistical value processing is performed for the gas flow rate, the variance value of the power and the variance value of the gas flow rate are used. Since it is clear that there is, the operator is requested to confirm the change in the gas flow rate.
[0079]
As a result, even when a measurement error occurs, accurate heat treatment can be performed without being affected by the error.
[0080]
[Example]
FIG. 9 shows the result when the present heating apparatus is used as an epitaxial growth apparatus. In this example, the power ratio η is used as a statistic._FHere is an example in which the result was effective.
[0081]
Here, (1) the thermocouple was replaced in the eighth batch, and (2) control (management) was started in the twelfth batch. Here, when the thermocouple was replaced, the front temperature was measured to be low due to a faulty installation or an error in the thermocouple, so the front temperature control unit issued a larger power supply command to the front heating unit, and as a result, In addition, the front part is more heated, and the outer peripheral part film thickness increases. Here, when the control is turned on in (2), η_FExceeds the management range UCL2 for alarm generation, the operator immediately issues a "heating device confirmation" alarm and η_FChart was displayed.
[0082]
As a result, the operator judged that there was a problem with the thermocouple replaced in (1), and set a new thermocouple. As a result, η_FEntered between the control start management ranges UCL1 and LCL1, and the film thickness distribution also decreased.
[0083]
Next, at the 22nd batch, η_FHas exceeded UCL1, the generation means 17 makes a correction of about 5% to the front power correction means 20F, so that the film thickness distribution decreases immediately from the next batch.
[0084]
Normally, when the same product is continuously operated, quality such as film thickness may not be measured until a plurality of batches are completed. In the case of this embodiment, the 18th to 25th batches are of the same type, and the film thickness measurement is not normally performed. Therefore, in the case where the heating apparatus is not the heating apparatus of the present invention, it is possible that many defects were not noticed until the end of 25 batches without noticing the change in the film thickness distribution deteriorated in 22 batches. However, according to the invention, η_FBy controlling this, the change in the chamber was grasped, and the power supplied to the front part was reduced, so that the deterioration of the film thickness distribution could be suppressed to a minimum of only one of 22 batches.
[0085]
[Modification]
(1) In the above embodiment, the statistical value of the power has been mainly described. However, the data to be statistically processed is not limited thereto, and other data may be statistically processed and used. Also in this case, although the degree is different, almost the same operation and effect as the above embodiment can be obtained.
[0086]
(2) In the above embodiment, as the process data, the power supplied to the heating means 13C, 13F, 13S, and 13B during the heat treatment, the wafer temperature, the gas flow rate, the gas concentration, and the gas temperature have been described as examples. However, all process data that can be statistically processed can be used. Also in this case, although the degree is different, almost the same operation and effect as the above embodiment can be obtained.
[0087]
(3) Other related statistical values include a covariance value, a correlation coefficient, and the like representing a relationship between data.
[0088]
【The invention's effect】
As described above, according to the heat treatment method and the heat treatment apparatus of the present invention, the following effects can be obtained.
[0089]
A change in the process state that may occur during actual operation, such as a measurement error occurring in the temperature measuring means, can be reliably detected, and accurate heat treatment can be performed without being affected by the error. For example, a constant process state can be maintained, and the in-plane temperature distribution of the thin plate-like heating target can be maintained constant.
[0090]
As a result, for example, in an epitaxial growth furnace, it is possible to perform a heat treatment that does not deteriorate the film thickness distribution or the specific resistance distribution caused by a defect in the temperature distribution, and that does not generate slip caused by thermal stress.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a heat treatment apparatus according to an embodiment of the present invention.
FIG. 2 is an explanatory diagram schematically showing an example of a temperature measuring point in an epitaxial growth apparatus.
FIG. 3 is a block diagram illustrating an arrangement example of a temperature measuring unit, a lamp, and a temperature control device.
FIG. 4 is a table showing an example of a temperature measurement error due to deterioration of a thermocouple.
FIG. 5 is a graph showing a relationship between a change in power supplied to a heating unit and an actual temperature of each unit.
FIG. 6 is a graph showing a change with time of a film thickness distribution.
FIG. 7 is a flowchart showing processing when a recipe is determined.
FIG. 8 is a flowchart showing processing during normal operation.
FIG. 9 is a graph showing changes such as a film thickness and a temperature difference in a specific example.
[Explanation of symbols]
11: heat treatment apparatus, 12C, 12F, 12S, 12B: temperature measurement means, 13C, 13F, 13S, 13B: heating means, 14C, 14F, 14S, 14B: temperature control means, 15: data storage means, 16: statistical analysis Means, 17: gain determination and alarm generation means, 18C, 18F, 18S, 18B: power measurement means, 19: alarm display, 20C, 20F, 20S, 20B: power correction means.

Claims (4)

In the heat treatment method of dividing the thin plate-shaped heating object into a plurality of zones, and individually controlling and heat-treating each zone,
One or more data of the process data during the heat treatment is collected and stored in real time, and statistical processing is performed for each step. The current statistical value during the heat treatment and the past statistical value of the same process data Alternatively, a heat treatment method characterized by grasping a current process state by comparing with other related statistical values, and controlling based on the result. The heat treatment method according to claim 1,
As the statistical value, one or both of an average value and a variance value are used, and a power average value ratio between the respective zones is maintained constant for each step, and the variance value is managed to a certain value or less. Heat treatment method. In a heat treatment apparatus including a heating unit that heats a thin plate-shaped heating target that is divided into a plurality of zones and is individually controlled for each zone,
Power measuring means for measuring the power supplied to the heating means, temperature measuring means provided in each of the zones or some of the zones for measuring the temperature of the zones, and introduced to the surface of the thin plate-shaped object to be heated Gas flow measuring means for measuring the flow rate of the gas to be heated, gas concentration measuring means for measuring the concentration of the gas introduced to the surface of the thin plate-shaped heating object, or gas introduced to the surface of the thin plate-shaped heating object One or more of gas temperature measuring means for measuring the temperature of
Storage means for storing the measurement results of each of the above means,
A control for statistically processing the data stored in the storage means, comparing the processed statistic with a past statistic or other related statistic to grasp a current process state, and controlling based on the result. And a heat treatment apparatus. The heat treatment apparatus according to claim 3,
As the statistical value, one or both of an average value and a variance value are used, and a power average value ratio between the respective zones is maintained constant for each step, and the variance value is managed to a certain value or less. Heat treatment equipment.

Images