JPS59225312A - Plating thickness measuring method - Google Patents

Abstract

PURPOSE:To measure a plating thickness with good accuracy according to fluorescent X-ray spectroscopy by using four or more of standard specimens, by performing operation by a formula showing a calibration curve having a specific correction term. CONSTITUTION:Three standard specimens are properly selected from four or more standard specimens of which plating thicknesses (x) are known to measure fluorescent X-ray intensities (n) while X-ray intensity n<infinity>, background n0 and absorption coefficient mu of a specimen having a thickness regarded as an indefinite thickness are calculated by formula (1). The plating thickness (x) of the other standard specimen and values n<infinity>, n0 and mu are substituted for the formula (1) to calculate the calculation value of fluorescent X-ray intensity and the difference DELTAn with the actually measured fluorescent X-ray intensity is calculated to form formulae (2), (3). The fluorescent X-ray intensity of the specimen is calculated to be compared with the fluorescent X-ray intensity value of each standard specimen and the formula of the corresponding DELTAn is selected from the formula (3) and, for example, the plating thickness (x) is calculated by formula (5).

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method of measuring plating thickness by a fluorescent X-ray method using four or more standard samples.

Conventionally, when measuring plating thickness using the fluorescent X-ray method,
It has been 2t since plating thickness was measured from the following relational expression (1).

n=Cno9-Tno)(1-g-μ勺+1゜...,
(1) Here, n is the measured X-ray intensity when the plating thickness is 2, nLx
I is the X-ray intensity from a sample with a thickness that can be considered infinite;
fLo is the background and μ is the absorption coefficient. If there are three mark samples, substitute all six measured values into equation (1) and solve the fc three-dimensional simultaneous equation. L.

, noo, μ are determined. Conventionally, the number of samples is 4.
n even if there are more than one. , 71■, the value of μ is finally treated as a constant using, for example, the method of least squares n
It's here.

However, using the conventional method from three standard samples?
Find the values of L", ?L0, μ, and then do the same ')zc<,
Using the values of %Q and μ, calculate the difference between the fluorescent X-ray intensity obtained from formula (1) and the actual fluorescent X-ray intensity for several samples whose plating thickness is known. If you look at it, that is, ξ
When calculating the error when using the equation (1) as the entire calibration curve, it was found that the error changes systematically with respect to the plating thickness π.

(See FIG. 1) Therefore, the conventional method has the disadvantage that such systematic errors cannot be removed.

The present invention provides an extremely effective means for quickly eliminating the above-mentioned drawbacks, and aims to accurately measure plating thickness using a fluorescent X-ray method using four or more standard samples. shall be.

The film thickness measuring method according to the present invention will be described in detail below with reference to the drawings. The drawings show a thickness measuring device according to the present invention,
In the drawing, reference numeral 1 denotes a radiation source, and radiation X emitted from the radiation source 1 passes through a collimator 2 and is irradiated onto a sample 3 to be measured. Reference numeral 4 denotes a fluorescent X-ray x generated from the sample to be measured 3. 2 is a radiation detector that detects the radiation.
7 is a pulse height analyzer for analyzing the pulse height from the output of the amplifier 6; 8 is a CPU 19 for calculating all the outputs of the pulse height analyzer 7; and 8 is a display section for displaying the output of the 0PU8.

Next, a case will be described in which the plating thickness measuring device of the present invention having the above configuration is operated.

First, the parameters of the calibration curve are determined by measuring the fluorescent X-ray intensity from four or more standard samples whose plating thicknesses are known. To explain the procedure in detail, three standard samples are randomly selected from four or more standard samples and %'M is determined using the conventional method.
, %o, μ are calculated.

Next, by substituting the plating thicknesses of other standard samples and the values of n■, no, and μ, the calculated value of the fluorescent X-ray intensity at each plating thickness is obtained, and the fluorescent X-ray intensity is compared with the actually measured fluorescent X-ray intensity. Calculate the difference between If this value is Δn, equation (2) holds true.

n=(n■-RQri (1- to -μ勺+n0+△n(2
) A broken line consisting of several straight lines on the △n-z plane is determined from the value of n x △9 obtained in this way. Calculate the slope and intercept of these lines. For example, the following formula is determined.

Here AI , A2 , , B1 , Bt , , a
is the constant 1×Ip”Zl. is the plating thickness of the standard sample.

Next, the strong fluorescence X-ray history of the measurement sample whose plating thickness is unknown is measured and the plating thickness is calculated from equations (2) and (3). Specifically, first, the total fluorescent X-ray intensity of the measurement sample is measured,
By comparing the fluorescence ). For example, if it is known that the plating thickness to be determined is between π2 and z3, the formula used here is formula (4).

Δ71=A, π0Bt ('1) Substituting equation (4) into equation (2), we obtain equation (5).

n= Cn0O−n6 ) (1e−μ”) +7t
o + A2 Z + B2°. , (5) Kokote? L” y 7Lo # P t A2 t
Bt is an n-* constant determined from the measurement results of the standard sample, and n is the measured value of the fluorescence intensity of the sample. Plating thickness 2
is obtained by completely solving equation (5).

The standard sample used in the plating thickness measurement method according to the present invention is:
The curve is selected so that the systematic error caused by the conventional method is approximated by a broken line with a small number of break points. Further, even in the case where the plating thickness of a standard sample that can reduce the error is too thin to be used as a model sample, the plating thickness measuring method according to the present invention can be used.

In this case, parameters uniquely determined by the material, the type of plating metal, and the plating thickness of the standard sample are stored in advance or input at the time of measurement. If this parameter f x p and n, then △nff: The values of the slope A and the intercept B of the straight line are determined as follows. FIG. 3 shows the relationship between the correction term Δn and the plating thickness X on the Δn-π plane. Here, X□p”2 is the plating thickness of the standard sample.The plating thickness determined by the conventional method is
If it is greater than or equal to -1 or less, connect the point Δn in Zl and the point △n in ZtK with a line segment and calculate the slope as A1△
Let n-intercept-IB. When the plating thickness determined by the conventional method is smaller than XpJ:, the point △1 in Zl and w
The slope f of the line segment connecting the point where the straight line passing through the point Δn in 2VC intersects with the Xp=00 straight line and the origin is set to zero.

As described above, the plating and thickness measurement method according to the present invention includes the systematic errors caused by the conventional method by adding a correction term ff that changes linearly to the conventional method 2U). ', tyx Even when it is not possible to actually produce a target sample, by using the parameter Xp, the plating thickness can be accurately measured from a few target samples.

[Brief explanation of drawings]

Figure 1 shows the actual fluorescent X-ray intensity and the conventional method, that is, equation (1).
The difference in fluorescence X-ray intensity calculated by nf is plotted against plating thickness X. Figure 2 is a block diagram of a device related to an embodiment of the present invention. 1°. Radiation generating section 21. Collimator 30. Sample to be measured 40. Radiation detector 6°, linear amplifier 7°. Wave height analyzer 8 degrees, CPU 9. FIG. 3 of the display section is an explanatory diagram of the case where the relational expression between the correction term 4n and the plating thickness Z(1) is determined using the parameter Xpf. Applicant Daini Seikosha Co., Ltd. Attorney Mogami, Patent Attorney

Claims (1)

[Claims] In the plating thickness measurement method using the fluorescent X-ray method, the first step is to measure and remember the fluorescent X-ray intensity from a sample with a known plating thickness of 4 or more, and to measure the background and plating thickness. The second step of determining the thickness relationship is υ, which is an equation that expresses a calibration curve with n correction terms obtained by this second step, n=("Tho) (1-e-μ勺+nQ)
+A Z +B (Here n is the measured X-ray intensity, nOo is the value representing the X-ray intensity from a sample whose plating thickness is almost infinite, 'nO is the measured X-ray intensity from a sample with zero plating thickness. , μ is an absorption coefficient, and A and B are constants) to calculate the plating thickness.