JPH0846000A - Method for testing reliability of semiconductor element - Google Patents
Method for testing reliability of semiconductor elementInfo
- Publication number
- JPH0846000A JPH0846000A JP17547294A JP17547294A JPH0846000A JP H0846000 A JPH0846000 A JP H0846000A JP 17547294 A JP17547294 A JP 17547294A JP 17547294 A JP17547294 A JP 17547294A JP H0846000 A JPH0846000 A JP H0846000A
- Authority
- JP
- Japan
- Prior art keywords
- oxide film
- electric field
- field strength
- time
- current density
- 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.)
- Pending
Links
Landscapes
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
Description
【0001】[0001]
【産業上の利用分野】本発明は半導体素子の信頼性試験
方法に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor device reliability test method.
【0002】[0002]
【従来の技術】半導体集積回路装置の高密度化、高集積
化、微細化の進行に伴い、それを構成するMOSトラン
ジスタあるいはMOSキャパシタで用いられるSiO2
膜(以下酸化膜と称する)は薄膜化の傾向にある。この
一方で、半導体集積回路の電源電圧は通常一定に保たれ
ることから、酸化膜に印加される電界強度は薄膜化に伴
って増大する傾向にある。この状況の下で、酸化膜の経
時絶縁破壊(TDDB:Time Dependent Dielectric Br
eakdown)が信頼性上重大な問題となってきている。こ
れは酸化膜に電界が印加された場合、電界の印加開始か
らある時間が経過した時点で酸化膜が破壊し、電気的な
絶縁性が失われて短絡が生じる現象である(この時間を
以下酸化膜寿命と称する)。このため、酸化膜寿命推定
のための経時絶縁破壊試験が半導体集積回路装置を設計
し開発する上で必須の項目となっており、その重要性は
さらに増大する傾向にある。従来の酸化膜寿命の推定方
法を以下に示す。2. Description of the Related Art With the progress of higher density, higher integration and miniaturization of semiconductor integrated circuit devices, SiO 2 used in MOS transistors or MOS capacitors constituting them.
A film (hereinafter referred to as an oxide film) tends to be thin. On the other hand, since the power supply voltage of the semiconductor integrated circuit is usually kept constant, the electric field strength applied to the oxide film tends to increase as the film thickness is reduced. Under these circumstances, the time-dependent dielectric breakdown (TDDB: Time Dependent Dielectric Br
eakdown) has become a serious issue in terms of reliability. This is a phenomenon in which, when an electric field is applied to the oxide film, the oxide film is destroyed at a certain time after the start of the application of the electric field, electrical insulation is lost, and a short circuit occurs. Called oxide film life). Therefore, the time-dependent dielectric breakdown test for estimating the lifetime of the oxide film is an essential item in designing and developing the semiconductor integrated circuit device, and its importance tends to further increase. The conventional method for estimating the lifetime of an oxide film is shown below.
【0003】図4および図5は従来の経時絶縁破壊から
酸化膜寿命を推定する方法を示したものである。図4は
酸化膜寿命の推定方法、図5は各ストレス電界強度にお
ける酸化膜寿命の測定結果の例である。FIG. 4 and FIG. 5 show a conventional method for estimating the life of an oxide film from the time-dependent dielectric breakdown. FIG. 4 shows an example of an oxide film life estimation method, and FIG. 5 shows an example of measurement results of the oxide film life at each stress electric field strength.
【0004】図4において横軸は酸化膜に印加される電
界強度、縦軸は酸化膜寿命を示している。図4におい
て、1は実使用時の最大電界強度、2は実使用時の酸化
膜寿命の推定値、3はストレス電界強度、4は酸化膜経
時絶縁破壊による50%累積故障時間の実測値、5は酸
化膜経時絶縁破壊による50%累積故障時間の推定値で
ある。In FIG. 4, the horizontal axis represents the electric field strength applied to the oxide film, and the vertical axis represents the oxide film life. In FIG. 4, 1 is the maximum electric field strength during actual use, 2 is the estimated value of the oxide film life during actual use, 3 is the stress electric field strength, 4 is the measured value of 50% cumulative failure time due to oxide film aging breakdown, 5 is an estimated value of 50% cumulative failure time due to dielectric breakdown with time of the oxide film.
【0005】図5において横軸は酸化膜に電界を印加す
るストレス時間、右の縦軸は酸化膜経時絶縁破壊による
累積故障率P、左の縦軸は累積故障率Pから計算された
ln(−ln(1−P))を示している。図5において
6は酸化膜経時絶縁破壊による50%累積故障時間の実
測値、7は各ストレス電界強度の下での累積故障時間の
時間依存性である。累積故障率Pを図5の様に横軸を時
間の対数、縦軸をln(−ln(1−P))のグラフ上
にプロットすることは一般にワイブルプロットと呼ば
れ、故障がワイブル分布に従う場合、ワイブルプロット
の結果は直線となる。ワイブルプロットは酸化膜の経時
絶縁破壊の累積故障率をグラフ化する方法として広く用
いられている。In FIG. 5, the horizontal axis represents the stress time for applying an electric field to the oxide film, the right vertical axis represents the cumulative failure rate P due to oxide film aging dielectric breakdown, and the left vertical axis represents ln (calculated from the cumulative failure rate P. -Ln (1-P)) is shown. In FIG. 5, 6 is a measured value of 50% cumulative failure time due to dielectric breakdown with time of the oxide film, and 7 is time dependence of the cumulative failure time under each stress electric field strength. It is generally called a Weibull plot to plot the cumulative failure rate P on a graph in which the horizontal axis is the logarithm of time and the vertical axis is In (-ln (1-P)) as shown in FIG. 5, and the failure follows the Weibull distribution. In this case, the result of the Weibull plot is a straight line. The Weibull plot is widely used as a method of graphing the cumulative failure rate of dielectric breakdown of an oxide film over time.
【0006】酸化膜寿命を推定するため同一の形状、寸
法、製造プロセスにより形成された同等の酸化膜をキャ
パシタ絶縁膜とするMOSキャパシタが多数準備され、
複数の組に分けられる。各組に対してはそれぞれ図4に
示した実使用時の最大電界強度1よりも高いストレス電
界強度3を定電圧ストレスの形で印加する。このストレ
ス印加により各組の酸化膜は経時絶縁破壊を生じ、時間
と共に故障となる酸化膜の個数が増大する。In order to estimate the oxide film life, a large number of MOS capacitors having the same shape, size, and equivalent oxide film formed by the manufacturing process as a capacitor insulating film are prepared.
Divided into multiple sets. A stress electric field strength 3 higher than the maximum electric field strength 1 in actual use shown in FIG. 4 is applied to each set in the form of constant voltage stress. This stress application causes dielectric breakdown of the oxide films of each set with the lapse of time, and the number of oxide films that become defective increases with time.
【0007】この各ストレス電界強度の下での累積故障
率の時間依存性を図5に示すようにワイブルプロットす
る。経験的に酸化膜経時絶縁破壊の故障はワイブル分布
に従い、このため累積故障率の時間依存性のワイブルプ
ロットは通常直線となる。これから各ストレス電界強度
の下での累積故障率の時間依存性7に対して回帰直線を
求め、これを用いて各ストレス電界強度に対し累積故障
率が50%に達する時間、即ち50%累積故障時間の実
測値6を求める。The time dependence of the cumulative failure rate under each stress electric field strength is Weibull plotted as shown in FIG. Empirically, the failure due to aging of the oxide film follows the Weibull distribution, so that the Weibull plot of the time dependence of the cumulative failure rate is usually a straight line. From this, a regression line is calculated for the time dependence 7 of the cumulative failure rate under each stress electric field strength, and using this, the time when the cumulative failure rate reaches 50% for each stress electric field strength, that is, 50% cumulative failure The actual measurement value 6 of time is obtained.
【0008】この50%累積故障時間を酸化膜寿命とみ
なして図4に示すように50%累積故障時間の実測値6
を横軸電界強度、縦軸酸化膜寿命の片対数グラフにプロ
ットする。経験的には50%累積故障時間の実測値6は
このグラフ上で直線となる。これに基づいて50%累積
故障時間の実測値の回帰直線、即ち50%累積故障時間
の推定値5を求め、この直線を用いて実使用時の最大電
界強度1における実使用時の酸化膜寿命の推定値2を求
める。Assuming that the 50% cumulative failure time is the oxide film life, as shown in FIG. 4, the measured value of the 50% cumulative failure time is 6
Is plotted on a semi-logarithmic graph of electric field strength on the horizontal axis and oxide film life on the vertical axis. Empirically, the measured value 6 of the 50% cumulative failure time is a straight line on this graph. Based on this, the regression line of the measured value of the 50% cumulative failure time, that is, the estimated value 5 of the 50% cumulative failure time is obtained, and using this straight line, the oxide film life in the actual use at the maximum electric field strength of 1 in the actual use The estimated value 2 of is calculated.
【0009】拡散工程を終了したウエハ上のMOSキャ
パシタを用いて酸化膜の信頼性を短時間に評価するに
は、評価対象となる酸化膜をキャパシタ絶縁膜とする多
数のMOSキャパシタをウエハ状態のままで試験する。
この場合、オートプローバを用いてウエハ上のMOSキ
ャパシタに対して測定点を移動しながら、順次プロービ
ングして測定する。In order to evaluate the reliability of an oxide film in a short time by using a MOS capacitor on a wafer which has completed the diffusion process, a large number of MOS capacitors having an oxide film to be evaluated as a capacitor insulating film are in a wafer state. I will test it.
In this case, measurement is performed by sequentially probing while moving the measurement point with respect to the MOS capacitor on the wafer using an auto prober.
【0010】[0010]
【発明が解決しようとする課題】しかし従来の酸化膜寿
命の推定方法には以下の問題点があった。However, the conventional method of estimating the life of the oxide film has the following problems.
【0011】製造プロセス開発の際にプロセス条件は頻
繁に変更される。このプロセス条件の変更により酸化膜
寿命が変化する可能性があるため、酸化膜寿命の推定を
頻繁に行う必要がある。Process conditions are frequently changed during manufacturing process development. Since the oxide film life may change due to the change of the process conditions, it is necessary to frequently estimate the oxide film life.
【0012】あるいは工場で製品を量産する際に、装置
のトラブル等によるプロセス条件の変動により拡散工程
を終了したウエハ毎に酸化膜寿命が変動し、ときにはい
ちじるしく酸化膜寿命が短く信頼性上の基準を満たさな
くなる可能性がある。このような異常が発生した場合
に、市場に信頼性が不十分な製品を出荷することを防ぐ
ため、見落としなく早急に酸化膜寿命の推定値を測定す
ることが必要になる。Alternatively, when a product is mass-produced in a factory, the oxide film life varies for each wafer that has completed the diffusion process due to a change in process conditions due to equipment troubles, etc., and sometimes the oxide film life is extremely short and a reliability criterion. May not be met. When such an abnormality occurs, it is necessary to immediately measure the estimated value of the oxide film life without overlooking, in order to prevent the shipping of products with insufficient reliability to the market.
【0013】以上の場合において、拡散工程を終了した
全ウエハに対して、ウエハ毎に酸化膜寿命の推定値を短
期間で測定することが必要となる。In the above cases, it is necessary to measure the estimated value of the oxide film life for each wafer in a short period for all the wafers that have undergone the diffusion process.
【0014】従来の酸化膜寿命の試験方法の中でウエハ
状態のままで試験をする場合においては、オートプロー
バを用いてウエハ上のMOSキャパシタに順次プロービ
ングして測定するために、同時に試験が可能なMOSキ
ャパシタは、プローブカードのプローブ針により同時に
電気的接続を取ることができる1個あるいは数個に限ら
れる。このため、多数のMOSキャパシタに複数のスト
レス電界強度を印加して累積故障率を測定していくには
長時間が必要とされ、特に低ストレス電界強度の測定に
おいてこの傾向はいちじるしく、全ウエハの酸化膜寿命
の推定をする際の大きな障害となっている。In the conventional method for testing the life of an oxide film, when the test is performed in the wafer state as it is, the test can be performed simultaneously since the MOS capacitors on the wafer are sequentially probed and measured using an autoprober. The number of such MOS capacitors is limited to one or several that can be electrically connected simultaneously by the probe needle of the probe card. Therefore, it takes a long time to measure a cumulative failure rate by applying a plurality of stress electric field strengths to a large number of MOS capacitors, and this tendency is remarkable especially in the measurement of a low stress electric field strength. This is a major obstacle in estimating the oxide film life.
【0015】測定時間を短くするためには試験時のスト
レス電界強度を高くする、あるいは測定に用いるMOS
キャパシタの数を少なくする等の方法がある。In order to shorten the measurement time, the stress electric field strength during the test is increased or the MOS used for the measurement is used.
There are methods such as reducing the number of capacitors.
【0016】しかしストレス時の電界強度を高くした場
合には、測定器の時間精度が不十分となるという問題、
および高電界により試験時に多量の発熱が生じ、酸化膜
の温度が予測不可能な上昇をし、経時絶縁破壊の時間が
本来の温度および電界強度における値よりも短くなると
いう問題により寿命推定に大きな誤差が生じる。このた
めストレス時の電界強度を高くして測定時間を短くし、
全ウエハの酸化膜寿命を推定することができなかった。However, when the electric field strength during stress is increased, the time accuracy of the measuring instrument becomes insufficient,
Also, a large amount of heat is generated during the test due to the high electric field, the temperature of the oxide film rises unpredictably, and the time-dependent dielectric breakdown time becomes shorter than the original value at the original temperature and electric field strength. There is an error. Therefore, the electric field strength during stress is increased to shorten the measurement time,
The oxide film lifetime of all wafers could not be estimated.
【0017】また、測定に用いるMOSキャパシタの数
を少なくした場合には、複数のストレス電界強度に対し
て累積故障率を求めることから、各ストレス電界強度に
割り当てられるMOSキャパシタの数は一層少なくな
る。同一ウエハ上のMOSキャパシタは同等であると仮
定されているにもかかわらず、実際には特性にばらつき
を持っている。各ストレス電界強度における累積故障率
は異なるMOSキャパシタから求められる。このため、
各ストレス電界強度に割り当てられるMOSキャパシタ
の数が少なくなるほど、各MOSキャパシタの特性のば
らつきが各ストレス電界強度における累積故障率の分布
の変位となって現われる。この結果、各MOSキャパシ
タの特性のばらつきが各ストレス電界強度における酸化
膜寿命の変位として現われ、これが最終的に酸化膜寿命
の推定値の誤差を大きくする。結論として測定に用いる
MOSキャパシタの数を少なくして測定時間を短くし、
全ウエハの酸化膜寿命を推定することができなかった。Further, when the number of MOS capacitors used for measurement is reduced, the cumulative failure rate is obtained for a plurality of stress electric field strengths, so that the number of MOS capacitors assigned to each stress electric field strength is further reduced. . Although it is assumed that the MOS capacitors on the same wafer are equivalent to each other, the characteristics actually vary. The cumulative failure rate at each stress electric field strength is obtained from different MOS capacitors. For this reason,
As the number of MOS capacitors assigned to each stress electric field strength decreases, the variation in the characteristics of each MOS capacitor appears as a displacement of the cumulative failure rate distribution at each stress electric field strength. As a result, variations in the characteristics of each MOS capacitor appear as a displacement of the oxide film life at each stress electric field strength, which ultimately increases the error in the estimated value of the oxide film life. In conclusion, reduce the number of MOS capacitors used for measurement and shorten the measurement time,
The oxide film lifetime of all wafers could not be estimated.
【0018】結局、ウエハ状態のままでオートプローバ
を用いてウエハ上のMOSキャパシタに順次プロービン
グして測定する方法では、実用上十分短い測定時間と実
用上十分高い精度を有する酸化膜寿命の推定値を実現す
ることができず、全ウエハの酸化膜寿命を推定すること
ができなかった。After all, in the method of probing the MOS capacitors on the wafer one after another using the autoprober in the wafer state as it is, the estimated value of the life of the oxide film has a sufficiently short measurement time in practice and a sufficiently high precision in practice. Could not be realized, and the oxide film life of all wafers could not be estimated.
【0019】本発明の目的は実用上十分短い測定時間
と、実用上十分高い推定精度を有する低コストの酸化膜
寿命の推定方法を提供することにより、拡散工程を終了
した全ウエハに対し酸化膜寿命の推定値を短期間で測定
することを可能とし、これにより製造プロセス開発の際
のプロセス条件の変更による酸化膜寿命の変化、あるい
は工場で製品を量産する際に、装置のトラブル等による
プロセス条件の変動による酸化膜寿命の変動を短期間で
見落とすことなくモニタリングすることを可能とするも
のである。An object of the present invention is to provide a low-cost method for estimating the lifetime of an oxide film, which has a sufficiently short measurement time for practical use and a sufficiently high estimation accuracy for practical use. It is possible to measure the estimated lifetime value in a short period of time, which allows changes in the oxide film lifetime due to changes in process conditions during development of the manufacturing process, or processes due to equipment troubles during mass production of products in the factory. This makes it possible to monitor changes in oxide film life due to changes in conditions in a short period of time without overlooking.
【0020】[0020]
【課題を解決するための手段】上記目的を達成するため
に、本発明の半導体素子の信頼性試験方法は、酸化膜に
複数の電界強度Eを印加して酸化膜中を流れる電流密度
Jを測定し、前記電流密度Jの測定値に対して、電界強
度の逆数1/Eに対する電流密度Jを電界強度Eの2乗
で除した値J/E2の対数の計算値の組を求め、前記各
計算値の組を直線あるいは一次式により近似し、前記直
線あるいは一次式を用いて酸化膜寿命を推定する電界強
度における酸化膜中を流れる電流密度を推定し、前記酸
化膜の破壊までの総電荷量Qbdを測定し、前記酸化膜
の破壊までの総電荷量Qbdを前記酸化膜寿命を推定す
る電界強度における酸化膜中を流れる電流密度により除
することにより酸化膜寿命を推定する。In order to achieve the above object, a semiconductor device reliability test method according to the present invention applies a plurality of electric field strengths E to an oxide film to determine a current density J flowing in the oxide film. The measured value of the current density J is measured, and a set of logarithmic values of the value J / E 2 obtained by dividing the current density J with respect to the reciprocal 1 / E of the electric field strength by the square of the electric field strength E is obtained, Approximate the set of each calculated value by a straight line or a linear expression, estimate the current density flowing in the oxide film at the electric field strength to estimate the oxide film life using the straight line or a linear expression, until the destruction of the oxide film The total charge amount Qbd is measured, and the oxide film life is estimated by dividing the total charge amount Qbd until the oxide film is destroyed by the current density flowing in the oxide film at the electric field strength for estimating the oxide film life.
【0021】また、破壊までの総電荷量Qbdの測定時
に酸化膜に定電流を印加し、破壊までの時間を測定する
ことにより破壊までの総電荷量を測定する。Further, a constant current is applied to the oxide film at the time of measuring the total charge amount Qbd until the breakdown, and the total charge amount until the breakdown is measured by measuring the time until the breakdown.
【0022】さらに、破壊までの総電荷量Qbdの測定
時に酸化膜に電流を階段波掃引して時間と共に増加させ
ることにより破壊までの総電荷量Qbdを求める。Further, when measuring the total amount of charge Qbd until breakdown, the total amount of charge Qbd until breakdown is obtained by sweeping a current through the oxide film in a stepwise wave and increasing it with time.
【0023】[0023]
【作用】上記本発明の酸化膜寿命の推定方法の構成によ
れば、酸化膜寿命の推定値を、従来多数のMOSキャパ
シタを用いて求めていたのに対して、1個のMOSキャ
パシタから求めることができる。According to the structure of the method for estimating the life of the oxide film of the present invention, the estimated value of the life of the oxide film is obtained from one MOS capacitor, whereas it is obtained from a large number of MOS capacitors in the past. be able to.
【0024】さらにストレス印加の時間は酸化膜に印加
する電界強度を測定器の時間精度が不十分となるという
問題、および高電界により試験時に多量の発熱が生じ
て、酸化膜の温度が予測不可能な上昇をし、経時絶縁破
壊の時間が本来の温度および電界強度における値よりも
短くなるという問題を生じない範囲内で高くし、一例と
して従来の酸化膜寿命の推定方法における複数のストレ
ス電界強度の中で最高のストレス電界強度と同程度とす
ることにより、従来の酸化膜寿命の推定方法における低
ストレス電界印加時よりも十分短くすることが可能であ
る。Furthermore, the time of stress application is such that the time accuracy of the measuring instrument for the electric field strength applied to the oxide film becomes insufficient, and a large amount of heat is generated during the test due to the high electric field, and the temperature of the oxide film is unpredictable. It is possible to increase it within a range that does not cause the problem that the time-dependent dielectric breakdown time becomes shorter than the original value at the temperature and the electric field strength. By setting the strength to be approximately the same as the highest stress electric field strength, it is possible to make it sufficiently shorter than when a low stress electric field is applied in the conventional oxide film life estimation method.
【0025】これに加えて酸化膜中を流れる電流密度J
を測定する時間はストレス印加の時間に比べて十分短
い。In addition to this, the current density J flowing in the oxide film
The time for measuring is sufficiently shorter than the time for applying stress.
【0026】以上の理由から本発明における酸化膜寿命
の推定方法における測定時間は、従来の酸化膜寿命の推
定方法のなかで、ウエハ状態のままでオートプローバを
用いてウエハ上のMOSキャパシタに順次プロービング
して測定する方法よりも数十分の一から数千分の一に短
縮され、ウエハ状態のまま、実用上十分短い測定時間で
酸化膜寿命の推定値を求めることが可能となる。For the above reasons, the measurement time in the method of estimating the life of an oxide film according to the present invention is the same as that of the conventional method of estimating the life of an oxide film, in which the MOS capacitors on the wafer are sequentially used in the wafer state using an autoprober. It is shortened to several tenths to several thousandths as compared with the method of measuring by probing, and it becomes possible to obtain the estimated value of the oxide film life in a wafer state in a practically short measurement time.
【0027】一方本発明による酸化膜寿命の推定方法に
おいては、同一のMOSキャパシタから破壊までの総電
荷量Qbdの測定値と各電界強度における電流密度Jの
推定値を求め、この2種類の量から各電界強度における
酸化膜寿命の推定値を求めることができる。このため従
来の酸化膜寿命の推定方法における問題、即ち各ストレ
ス電界強度における酸化膜寿命が異なるMOSキャパシ
タから求められるため、各MOSキャパシタの特性のば
らつきが各ストレス電界強度での酸化膜寿命の変位とな
って現われ、最終的に酸化膜の寿命推定の誤差が大きく
なるという問題を排除することができる。On the other hand, in the method of estimating the lifetime of the oxide film according to the present invention, the measured value of the total charge amount Qbd from the same MOS capacitor to the breakdown and the estimated value of the current density J at each electric field strength are obtained, and these two types of amounts are calculated. From this, the estimated value of the oxide film life at each electric field strength can be obtained. Therefore, since there is a problem in the conventional method of estimating the lifetime of an oxide film, that is, since it is required from the MOS capacitors having different oxide film lifetimes at each stress electric field strength, the variation in the characteristics of each MOS capacitor causes the variation of the oxide film lifetime at each stress electric field strength. The problem that the error in estimating the lifetime of the oxide film finally becomes large can be eliminated.
【0028】以上の理由から、本発明における酸化膜寿
命の推定方法における推定精度は、測定時間を短縮して
も低下することなく、従来の酸化膜寿命の推定方法と同
程度かそれ以上の精度を保つことができる。For the above reasons, the estimation accuracy of the method for estimating the life of an oxide film according to the present invention does not decrease even if the measurement time is shortened, and is as high as or better than that of the conventional method for estimating the life of an oxide film. Can be kept.
【0029】[0029]
【実施例】以下本発明を実施例により図面を参照して詳
細に説明する。DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the drawings by embodiments.
【0030】図1〜図3は本発明における酸化膜寿命の
推定方法を示したものである。図1は酸化膜寿命の推定
方法、図2は酸化膜中を流れる電流密度Jの電界強度依
存性の測定結果、図3は定電流密度ストレス印加時の酸
化膜破壊までの酸化膜印加電圧の時間変化の測定結果の
例である。1 to 3 show a method of estimating the life of an oxide film according to the present invention. FIG. 1 is a method for estimating the lifetime of an oxide film, FIG. 2 is a measurement result of electric field strength dependence of a current density J flowing in the oxide film, and FIG. 3 is a voltage of an oxide film applied until the oxide film is destroyed when a constant current density stress is applied. It is an example of the measurement result of time change.
【0031】図1において、図面上部の横軸は酸化膜に
印加される電界強度E、下部の横軸は酸化膜に印加され
る電界強度Eの逆数1/E、縦軸は酸化膜を流れる電流
の電流密度Jを酸化膜に印加される電界強度Eの2乗で
除した値J/E2をそれぞれ示している。In FIG. 1, the horizontal axis in the upper part of the drawing is the electric field strength E applied to the oxide film, the lower horizontal axis is the reciprocal 1 / E of the electric field strength E applied to the oxide film, and the vertical axis is the oxide film. A value J / E 2 obtained by dividing the current density J of the current by the square of the electric field strength E applied to the oxide film is shown.
【0032】図1において、11は電流密度Jの実測値
を電界強度Eの2乗で除した値、12はフォウラ−ノル
ドハイム(Fowler-Nordheim)プロットによるJ/E2の
推定値、13は破壊までの総電荷量Qbdと電流密度J
から求めた酸化膜寿命の推定値、14は実使用時の最大
電界強度、15は実使用時の酸化膜寿命の推定値であ
る。In FIG. 1, 11 is the value obtained by dividing the measured value of the current density J by the square of the electric field strength E, 12 is the estimated value of J / E 2 by the Fowler-Nordheim plot, and 13 is the breakdown. Total charge Qbd and current density J
The estimated value of the oxide film life obtained from the above, 14 is the maximum electric field strength during actual use, and 15 is the estimated value of the oxide film life during actual use.
【0033】図2において、横軸は酸化膜に印加される
電界強度E、縦軸は酸化膜を流れる電流の電流密度Jを
それぞれ示している。図2において、16は電流密度
J、17は電流密度Jの実測値である。In FIG. 2, the horizontal axis shows the electric field strength E applied to the oxide film, and the vertical axis shows the current density J of the current flowing through the oxide film. In FIG. 2, 16 is the current density J and 17 is the measured value of the current density J.
【0034】図3において、横軸は酸化膜に定電流密度
ストレスを印加するストレス時間t、縦軸は定電流スト
レスを印加中の酸化膜印加電圧Voxをそれぞれ示して
いる。In FIG. 3, the horizontal axis represents the stress time t for applying the constant current density stress to the oxide film, and the vertical axis represents the oxide film applied voltage Vox during the application of the constant current stress.
【0035】拡散工程を終了したウエハにおける酸化膜
寿命は、ウエハ上に形成されたMOSキャパシタを用い
て測定され、その測定値から図1に示した実使用時の酸
化膜寿命の推定値15が求められる。今回の酸化膜の膜
厚は12nm、酸化膜寿命の推定に用いたMOSキャパ
シタの酸化膜寸法は100μm×50μmである。The oxide film life of the wafer after the diffusion process is measured by using a MOS capacitor formed on the wafer, and the estimated value 15 of the oxide film life in actual use shown in FIG. 1 is obtained from the measured value. Desired. The thickness of the oxide film this time is 12 nm, and the dimensions of the oxide film of the MOS capacitor used for estimating the lifetime of the oxide film are 100 μm × 50 μm.
【0036】このウエハ上のMOSキャパシタを、ウエ
ハ状態のままでオートプローバと連動した測定器を用い
て、ウエハ上の測定点を移動しながら順次プロービング
して測定する。この測定器には電圧源、電流源、電圧
計、電流計、容量計が備わっている。The MOS capacitors on the wafer are sequentially probed while moving the measurement points on the wafer by using a measuring device linked to an auto-prober in the wafer state. This measuring instrument is equipped with a voltage source, current source, voltmeter, ammeter, and capacitance meter.
【0037】まずMOSキャパシタに対して、酸化膜の
容量がMOSキャパシタの容量として現われる電圧がM
OSキャパシタに印加され、MOSキャパシタの容量が
測定される。その容量と酸化膜寸法から酸化膜厚が計算
される。First, with respect to the MOS capacitor, the voltage at which the capacitance of the oxide film appears as the capacitance of the MOS capacitor is M.
Applied to the OS capacitor, the capacitance of the MOS capacitor is measured. The oxide film thickness is calculated from the capacitance and the oxide film size.
【0038】次に同じMOSキャパシタに電圧を印加す
ることにより、酸化膜に電界強度6〜9MV/cm程度
の電界が印加され、図2に示すように各電界強度におけ
る電流密度J16の実測値17が得られる。ここで酸化
膜に印加される電界強度は酸化膜印加電圧を先に求めた
酸化膜厚で除することにより求められる。Next, by applying a voltage to the same MOS capacitor, an electric field having an electric field strength of about 6 to 9 MV / cm is applied to the oxide film, and as shown in FIG. Is obtained. Here, the electric field strength applied to the oxide film is obtained by dividing the oxide film applied voltage by the oxide film thickness obtained previously.
【0039】このようにして得られた電流密度J16の
実測値17から、図1に示すように、電流密度J17の
実測値17を電界強度Eの2乗で除した値J/E2の実
測値の対数を、電界強度Eの逆数1/Eに対してプロッ
トする、いわゆるフォウラ−ノルドハイムプロットを行
う。これは、電界を印加した際に酸化膜を流れる電流
は、主としてフォウラ−ノルドハイムトンネル電流で、
電流密度Jは酸化膜に印加される電界強度に対して、J
=AE2exp(−b/E)(A,bは定数)の形の依
存性を持つ。この依存性から主としてフォウラ−ノルド
ハイムトンネル電流により構成された酸化膜中を流れる
電流の電流密度Jを電界強度Eの2乗で除した値J/E
2の対数を電界強度Eの逆数1/Eに対してプロットし
た結果は直線となる。このプロットはフォウラ−ノルド
ハイムプロットと呼ばれ、この直線関係を用いて電流密
度Jの実測値を電界強度Eの2乗で除した値から、任意
の電界強度におけるフォウラ−ノルドハイムプロットに
よるJ/E2の推定値12を求める。このフォウラ−ノ
ルドハイムプロットによるJ/E2の推定値12を電界
強度Eの2乗E2を乗ずることにより、任意の電界強度
E、特に電流密度Jが微小なため実測値を得ることが不
可能な実使用時の最大電界強度4近傍の低電界強度時の
電流密度Jの推定値を求めることができる。From the measured value 17 of the current density J16 thus obtained, as shown in FIG. 1, the measured value J / E 2 obtained by dividing the measured value 17 of the current density J17 by the square of the electric field strength E is measured. A so-called Fowler-Nordheim plot is performed in which the logarithm of the value is plotted against the reciprocal 1 / E of the electric field strength E. This is because the current flowing through the oxide film when an electric field is applied is mainly the Fowler-Nordheim tunnel current,
The current density J is J with respect to the electric field strength applied to the oxide film.
= AE 2 exp (-b / E) (A and b are constants). From this dependence, the value J / E obtained by dividing the current density J of the current flowing in the oxide film mainly composed of the Fowler-Nordheim tunnel current by the square of the electric field strength E
The result of plotting the logarithm of 2 against the reciprocal 1 / E of the electric field strength E is a straight line. This plot is called a Fowler-Nordheim plot, and from this value obtained by dividing the measured value of the current density J by the square of the electric field intensity E using this linear relationship, J / Estimate 12 of E 2 is determined. By multiplying the estimated value 12 of J / E 2 by the Fowler-Nordheim plot by the square of the electric field strength E, E 2 , it is impossible to obtain an actually measured value because the arbitrary electric field strength E, especially the current density J is minute. It is possible to obtain an estimated value of the current density J at a low electric field strength near the maximum electric field strength 4 during possible actual use.
【0040】続いて同じMOSキャパシタの酸化膜に対
して電流密度J=1.0A/cm2定電流ストレスを印加
し、破壊までの総電荷量Qbdを測定する。定電流スト
レスを用いることにより、定電圧ストレスの場合に比べ
て、酸化膜の膜厚のばらつきによる測定時間のばらつき
を小さくできるという利点がある。このときの酸化膜印
加電圧の時間変化を図3に示す。定電流ストレスをMO
Sキャパシタの酸化膜に印加すると、定電流印加中の酸
化膜印加電圧の時間変化18に示すように、時間の経過
に従って酸化膜印加電圧の絶対値が酸化膜中への電子の
トラッピングにより徐々に増加する。この後、酸化膜印
加電圧は、ある時間の経過後、この例では18秒後に突
然大きく減少する。この酸化膜印加電圧の突然の大きな
減少が生じた時点が、酸化膜破壊が生じた時点19であ
る。定電流ストレス印加開始より酸化膜破壊が生じた時
点19までの時間tにストレス電流密度Jを乗じて、破
壊までの総電荷量Qbd=Jtを求める。この例では、
破壊までの総電荷量Qbd=18C/cm2である。こ
の破壊までの総電荷量Qbdは電界強度あるいは電流密
度Jに依存しない一定値となる。Then, a current density J = 1.0 A / cm 2 constant current stress is applied to the oxide film of the same MOS capacitor, and the total charge amount Qbd until breakdown is measured. The use of the constant current stress has an advantage that the variation in the measurement time due to the variation in the film thickness of the oxide film can be reduced as compared with the case of the constant voltage stress. FIG. 3 shows the time variation of the oxide film applied voltage at this time. MO for constant current stress
When applied to the oxide film of the S-capacitor, the absolute value of the oxide film applied voltage gradually changes over time due to the trapping of electrons into the oxide film as shown in time change 18 of the oxide film applied voltage during constant current application. To increase. After that, the applied voltage to the oxide film suddenly and largely decreases after a certain period of time, 18 seconds in this example. The time point when the oxide film applied voltage suddenly and drastically decreases is the time point 19 when the oxide film breakdown occurs. The time t from the start of the constant current stress application to the time 19 when the oxide film is destroyed is multiplied by the stress current density J to obtain the total charge amount Qbd = Jt until the breakdown. In this example,
The total amount of charge Qbd up to destruction is 18 C / cm 2 . The total charge amount Qbd until the breakdown has a constant value that does not depend on the electric field strength or the current density J.
【0041】この例に示した定電流ストレス印加による
破壊までの総電荷量Qbdの測定に代えて、ストレス電
流密度を階段波掃引し、時間の経過に従って増加させ、
破壊までの各段における時間と電流密度Jを積分して、
破壊までの総電荷量Qbdを求める方法も、広範囲にわ
たる破壊までの総電荷量Qbdの値を測定できることか
ら実用的な方法である。Instead of measuring the total amount of charge Qbd until breakdown by applying a constant current stress shown in this example, the stress current density is swept stepwise and increased with the passage of time.
Integrating the time and current density J at each stage until breakdown,
The method of obtaining the total charge amount Qbd until the breakdown is also a practical method because the value of the total charge amount Qbd until the breakdown can be measured over a wide range.
【0042】このようにして求めた酸化膜の破壊までの
総電荷量Qbdを、先に求めた任意の電界強度における
電流密度J、その中でも特に低電界のため電流密度Jが
微小で実測値を得ることが不可能な実使用時の最大電界
強度14の推定値によって除し、各電界強度における破
壊までの総電荷量Qbdと電流密度Jから求めた酸化膜
寿命の推定値13、その中でも特に実使用時の最大電界
強度14における実使用時の酸化膜寿命の推定値15が
得られる。The total amount of electric charge Qbd until the oxide film is broken, which is obtained in this way, is calculated by measuring the current density J at an arbitrary electric field strength obtained previously, and particularly the electric current density J is very small because it is a low electric field. It is divided by the estimated value of the maximum electric field strength 14 at the time of actual use, which is impossible to obtain, and the estimated value 13 of the oxide film life obtained from the total charge amount Qbd to the breakdown and the current density J at each electric field strength, among them, An estimated value 15 of the oxide film life in actual use at the maximum electric field strength 14 in actual use is obtained.
【0043】本発明の実施例における試験時間は、拡散
工程を終了した1ロット約50枚のウエハに対し各ウエ
ハ5点の測定点がある場合、約1.5時間という短時間
で全測定が終了し、各ウエハ上の各測定点の酸化膜寿命
の推定値が得られる。この測定時間は、一つの酸化膜寿
命の推定値を得るのに必要なMOSキャパシタの個数が
1個だけであり、かつ1個当りの測定時間を短縮できる
ため、従来の酸化膜寿命の推定方法による場合の数十分
の一であり、飛躍的な測定効率の向上が実現される。The test time in the embodiment of the present invention is such that when there are 5 measurement points on each wafer for about 50 wafers in one lot after the diffusion process, all the measurements are completed in a short time of about 1.5 hours. After that, an estimated value of the oxide film lifetime at each measurement point on each wafer is obtained. This measurement time requires only one MOS capacitor to obtain an estimated value of one oxide film life, and the measurement time per one can be shortened. This is a few tenths of the case described above, and a dramatic improvement in measurement efficiency is realized.
【0044】一方で本実施例の酸化膜寿命の推定方法に
おいては、各電界強度における酸化膜寿命の推定値は、
1個のMOSキャパシタから測定された破壊までの総電
荷量Qbd、および各電界強度における電流密度Jの推
定値から求められる。このため、酸化膜寿命の推定精度
は測定時間が短縮されたにもかかわらず低下しない。On the other hand, in the oxide film life estimation method of this embodiment, the estimated value of the oxide film life at each electric field strength is
It can be obtained from the total amount of charge Qbd measured from one MOS capacitor until destruction and the estimated value of the current density J at each electric field strength. For this reason, the estimation accuracy of the oxide film life does not decrease even though the measurement time is shortened.
【0045】[0045]
【発明の効果】本発明によれば、酸化膜寿命の推定値を
得るための測定時間を推定精度の低下を伴うことなく、
大幅に短縮することができる。これにより拡散工程を終
了した全ウエハに対して、酸化膜寿命の推定値を短期間
で測定することを実現する。According to the present invention, the measurement time for obtaining the estimated value of the oxide film life can be reduced without degrading the estimation accuracy.
It can be greatly shortened. As a result, it is possible to measure the estimated lifetime of the oxide film in a short period for all the wafers that have undergone the diffusion process.
【0046】これにより製造プロセス開発の際にはプロ
セス条件の変更による酸化膜信頼性の変化を短期間で評
価でき、開発効率の大幅な向上および開発コストの削減
という効果が得られる。As a result, changes in oxide film reliability due to changes in process conditions can be evaluated in a short period of time during the development of the manufacturing process, and the effects of greatly improving development efficiency and reducing development costs can be obtained.
【0047】一方、工場で製品を量産する際には、装置
のトラブル等によるプロセス条件の変動による酸化膜寿
命の変動を低コストかつ短期間で、全ウエハについてモ
ニタリングをすることが可能となることから、低コスト
で製品の信頼性の向上が実現できるという効果が得られ
る。On the other hand, when the products are mass-produced in the factory, it is possible to monitor the fluctuation of the oxide film life due to the fluctuation of the process conditions due to the trouble of the apparatus or the like for all the wafers at low cost and in a short period of time. Therefore, the effect that the reliability of the product can be improved at low cost can be obtained.
【図1】本発明の一実施例の酸化膜寿命の推定方法を示
す図FIG. 1 is a diagram showing a method of estimating an oxide film lifetime according to an embodiment of the present invention.
【図2】本発明の一実施例の酸化膜中を流れる電流密度
Jの電界強度依存性の測定結果を示す図FIG. 2 is a diagram showing measurement results of electric field strength dependence of a current density J flowing in an oxide film according to an embodiment of the present invention.
【図3】本発明の一実施例の定電流密度ストレス印加時
の酸化膜破壊までの酸化膜印加電圧の時間変化を示す図FIG. 3 is a diagram showing a time change of an oxide film applied voltage up to oxide film breakdown when a constant current density stress is applied according to an embodiment of the present invention.
【図4】従来の酸化膜寿命の推定方法を示す図FIG. 4 is a diagram showing a conventional method for estimating the lifetime of an oxide film.
【図5】従来の各ストレス電界強度における酸化膜寿命
の測定結果を示す図FIG. 5 is a diagram showing the measurement results of the oxide film lifetime at each conventional stress electric field strength.
11 値 12 推定値 13 推定値 14 最大電界強度 15 推定値 11 value 12 estimated value 13 estimated value 14 maximum electric field strength 15 estimated value
Claims (3)
化膜中を流れる電流密度Jを測定し、前記電流密度Jの
測定値に対して、電界強度の逆数1/Eに対する電流密
度Jを電界強度Eの2乗で除した値J/E2の対数の計
算値の組を求め、前記各計算値の組を直線あるいは一次
式により近似し、前記直線あるいは一次式を用いて酸化
膜寿命を推定する電界強度における酸化膜中を流れる電
流密度を推定し、前記酸化膜の破壊までの総電荷量Qb
dを測定し、前記酸化膜の破壊までの総電荷量Qbdを
前記酸化膜寿命を推定する電界強度における酸化膜中を
流れる電流密度により除することにより酸化膜寿命を推
定することを特徴とする半導体素子の信頼性試験方法。1. A plurality of electric field intensities E are applied to an oxide film to measure a current density J flowing in the oxide film, and the current density J is a reciprocal of the electric field intensity with respect to the measured value of the current density J. J is divided by the square of the electric field strength E to find a set of logarithmic values of J / E 2 , and each set of the calculated values is approximated by a straight line or a linear expression, and oxidation is performed using the straight line or the linear expression. The current density flowing in the oxide film at the electric field strength for estimating the film life is estimated, and the total charge amount Qb until the oxide film is destroyed.
The oxide film life is estimated by measuring d and dividing the total charge amount Qbd until the oxide film is destroyed by the current density flowing in the oxide film at the electric field strength for estimating the oxide film life. Reliability test method for semiconductor devices.
化膜に定電流を印加し、破壊までの時間を測定すること
により破壊までの総電荷量を測定することを特徴とする
請求項1記載の半導体素子の信頼性試験方法。2. The total amount of charge until breakdown is measured by applying a constant current to the oxide film at the time of measuring the total amount of charge Qbd until breakdown, and measuring the time until breakdown. A method for testing reliability of a semiconductor device as described above.
化膜に電流を階段波掃引して時間と共に増加させること
により破壊までの総電荷量Qbdを求めることを特徴と
する請求項1記載の半導体素子の信頼性試験方法。3. The total charge amount Qbd until breakdown is obtained by sweeping a current through the oxide film in a stepwise wave to increase with time when the total charge amount Qbd before breakdown is measured. Reliability test method for semiconductor devices.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP17547294A JPH0846000A (en) | 1994-07-27 | 1994-07-27 | Method for testing reliability of semiconductor element |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP17547294A JPH0846000A (en) | 1994-07-27 | 1994-07-27 | Method for testing reliability of semiconductor element |
Publications (1)
Publication Number | Publication Date |
---|---|
JPH0846000A true JPH0846000A (en) | 1996-02-16 |
Family
ID=15996661
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP17547294A Pending JPH0846000A (en) | 1994-07-27 | 1994-07-27 | Method for testing reliability of semiconductor element |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPH0846000A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6049213A (en) * | 1998-01-27 | 2000-04-11 | International Business Machines Corporation | Method and system for testing the reliability of gate dielectric films |
US6188234B1 (en) | 1999-01-07 | 2001-02-13 | International Business Machines Corporation | Method of determining dielectric time-to-breakdown |
CN114970157A (en) * | 2022-05-26 | 2022-08-30 | 北京航空航天大学 | Method for predicting test life of small sample of electronic product under voltage stress |
-
1994
- 1994-07-27 JP JP17547294A patent/JPH0846000A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6049213A (en) * | 1998-01-27 | 2000-04-11 | International Business Machines Corporation | Method and system for testing the reliability of gate dielectric films |
US6188234B1 (en) | 1999-01-07 | 2001-02-13 | International Business Machines Corporation | Method of determining dielectric time-to-breakdown |
CN114970157A (en) * | 2022-05-26 | 2022-08-30 | 北京航空航天大学 | Method for predicting test life of small sample of electronic product under voltage stress |
CN114970157B (en) * | 2022-05-26 | 2024-05-28 | 北京航空航天大学 | Method for predicting service life of small sample test of electronic product under voltage stress effect |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7579859B2 (en) | Method for determining time dependent dielectric breakdown | |
US6077719A (en) | Semiconductor device evaluation method, method of controlling the semiconductor device production processes and recording medium | |
US20110031981A1 (en) | Valuation method of dielectric breakdown lifetime of gate insulating film, valuation device of dielectric breakdown lifetime of gate insulating film and program for evaluating dielectric breakdown lifetime of gate insulating film | |
US5485097A (en) | Method of electrically measuring a thin oxide thickness by tunnel voltage | |
US3889188A (en) | Time zero determination of FET reliability | |
EP1279966A2 (en) | Apparatus and method for determining the reliability of integrated semiconductor circuits at high temperatures | |
US6198301B1 (en) | Method for determining the hot carrier lifetime of a transistor | |
US5793212A (en) | Method of measuring the breakdown charge of a dielectric film | |
US3882391A (en) | Testing the stability of MOSFET devices | |
JPH0846000A (en) | Method for testing reliability of semiconductor element | |
Olivo et al. | Influence of localized latent defects on electrical breakdown of thin insulators | |
US7155359B1 (en) | Determination of device failure characteristic | |
JP4844101B2 (en) | Semiconductor device evaluation method and semiconductor device manufacturing method | |
JP3239617B2 (en) | Semiconductor device reliability test method | |
JP4640203B2 (en) | Semiconductor wafer evaluation method | |
US6377067B1 (en) | Testing method for buried strap and deep trench leakage current | |
JP3019564B2 (en) | Evaluation method of insulating film | |
JP2584093B2 (en) | Insulation film reliability evaluation method | |
JP3859357B2 (en) | Insulating film evaluation method | |
JPH08153764A (en) | Method for testing reliability of semiconductor element | |
Li et al. | Effective defect density for MOS breakdown: Dependence on oxide thickness | |
Vollertsen | Statistical modelling of time dependent oxide breakdown distributions | |
JP2002141388A (en) | Evaluation method for semiconductor device and evaluation device therefor | |
Andreev et al. | Programmable set to monitor charge state change of MIS devices under high-fields | |
Hirayama et al. | Time dependent dielectric breakdown measurement of high pressure low temperature oxidized film |