Electricity: measuring and testing – Impedance – admittance or other quantities representative of... – Lumped type parameters
Reexamination Certificate
2000-07-31
2001-11-06
Metjahic, Safet (Department: 2858)
Electricity: measuring and testing
Impedance, admittance or other quantities representative of...
Lumped type parameters
C324S762010, C438S017000
Reexamination Certificate
active
06313648
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for quantitating impurity concentration and recording medium storing record of program for quantitating impurity concentration in semiconductor devices and the like, and, particularly, to a method for quantitating impurity concentration and recording medium storing record of program for quantitating impurity concentration by which highly accurate measurement can be made.
2. Description of the Related Art
There have been 70 or more methods proposed for measuring an impurity distribution in the direction of the depth in the vicinity of the surface of semiconductor devices and the like. Among these methods, SIMS (secondary ion mass spectrometry) is widely used since the information of the type and amount of trace elements in the direction of the depth can be obtained with marked sensitivity. The details of SIMS are described, for example, in the literature “SOLID SURFACE ANALYSIS I, PP. 196-257, published by KODANSHA, 1995”. Incidentally, SIMS is an analytical method in which a sample is destroyed.
In the analysis of impurity distribution using SIMS, a primary ion is applied to a sample to quantitatively analyze a secondary ion emitted from the sample. Because of this, even in the cases where impurities in the sample are inert and the impurity concentration is high, an analysis of the inside apart from the interface of the sample such as semiconductor substrates can be made with high sensitivity. SIMS has a large advantage which is unrivaled by other measuring methods.
A typical nondestructive analysis method currently used includes, those, such as RBS (Rutherford backscattering method) making use of energy scatter or methods based on a pulse CV measurement making use of, as its measurement principle, variation of a depletion layer which phenomenon is peculiar to semiconductors. A method for measuring the distribution of carrier concentration by means of pulse CV measurement is often adopted as the nondestructive analysis method particularly in semiconductor fields because of its simplicity.
For example, a distribution of carrier concentration supported by the pulse CV characteristics can be measured in a method for calculating the distribution of carrier concentration based on a pulse CV measurement as described in the literature “E. H. NICOLLIAN and J. R. BREWS, MOS Physics and Technology, pp. 371-422, JOHN WILEY & SONS, Inc., 1982”.
The pulse CV measurement can be easily made plural times because it is a nondestructive measurement. Hence even if there is a measurement fluctuation in a measurement device, the measurement fluctuation can be eliminated by means of a statistical method.
However, the process in which a secondary ion is produced in the quantitative measurement of impurity concentration by using SIMS is not clarified because it is complicated and diversified. Therefore, only empirical techniques have been put to practical use. Specifically, techniques using a standard sample are mostly adopted in general. However, in these methods, tens and several of percents of errors are produced according to the qualities of the standard sample, calculating methods of RSF (relative sensitivity factor), daily error variations and the like. Moreover, SIMS is a destructive analysis method and hence it is impossible that the influence of a fluctuation of RSF is eliminated by a statistical method based on repeated plural measurements and hence the impurity concentration cannot be calibrated.
On the other hand, in the method for calculating the distribution of carrier concentration based on the pulse CV measurement, the distribution of carrier concentration is calculated from a variation in the width of a depletion region when voltage is applied to a sample. Therefore, when a plurality of impurities is present, only a total amount of a carrier effected by all of these impurities can be measured.
When an impurity is introduced in a high concentration, the range of the distribution of carrier concentration obtained based on the pulse CV measurement is made narrow, which makes it impossible to obtain the total profile of the impurity introduced. In addition, in the case where an impurity is inert, no coincidence between impurity concentration and carrier concentration can be brought about because the inert impurity does not contribute to the electric properties. Therefore, there is a demand for a quantitative method enabling simple, widespread and highly accurate measurement of impurities.
As aforementioned, in the measurements using SIMS, accurate impurity concentration can not be quantitatively measured when there is no standard sample. This is because of the following reasons. Specifically, the value obtained directly from SIMS is an ion number per unit time. However, since a process in which this ion is produced is not clarified, it is mostly necessary to use an empirical method such as a method utilizing a calibration curve and using a standard sample. Even in the case where the standard sample is used, the daily errors of the relative sensitivity function RSF are large and hence measurement fluctuations are present. Moreover, since SIMS is a destructive analysis method, fluctuations in the results of the measurement cannot be eliminated by a statistical method, which makes it difficult to measure impurities quantitatively.
Accordingly, to measure impurity concentration using SIMS, a standard sample designed to have definite distribution of impurity concentration is required. However, a standard sample usable for an essential element cannot always be obtained. Even in the case where the standard sample is present, measurement fluctuations causes a variation in the measured impurity concentration. Also, since SIMS is a destructive analysis method, elimination of fluctuations by a statistical method is difficult, giving rise to the problem that accurate impurity concentration is not necessarily measured.
In the method for calculating the distribution of carrier concentration from the profile obtained by pulse CV measurement, accurate impurity concentration in a widespread region cannot be measured when plural impurities are present in the measuring area, the impurities are inert or impurities are present in a high concentration.
This is because of the following reason. Specifically, when the distribution of carrier concentration is calculated from a pulse CV measurement, a variation in the width of a depletion region when voltage is applied is utilized. In order for the distribution of an impurity to coincide with the distribution of carrier concentration, it is necessary that an impurity which is to be a carrier is the same as a comparative impurity and this comparative impurity is only one material corresponding to this case. Also, all impurities present in the system must be active. An introduction of an impurity in a high concentration limits the range in which the accuracy of the measurement of the distribution of carrier concentration is secured. For example, the range of measurement of the distribution of carrier concentration obtained based on pulse CV measurement is limited to narrow regions. As a consequence, the distribution of carrier concentration at a desired depth cannot be obtained.
Therefore, if plural impurities have been introduced when it is intended to measure the distribution of carrier concentration by pulse CV measurement, it is difficult to measure each concentration of impurities separately. When an impurity is introduced in a high concentration, the distribution of carrier concentration only in the vicinity of the interface can be measured. Moreover in the case where the impurity is inert, the distribution of carrier concentration does not coincide with the distribution of the impurity concentration.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a method for quantitating impurity concentration and a recording medium in which a program for quantitating impurity concentration is recorded, the method being characterized in that it is provi
Deb Anjan K
McGinn & Gibb PLLC
Metjahic Safet
NEC Corporation
LandOfFree
Method for quantitating impurity concentration in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for quantitating impurity concentration in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for quantitating impurity concentration in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2614377