JPH0211765A - Method of determining sputtering depth, sputtering yield or sputtering rate of metallic sample by glow discharge - Google Patents

Abstract

PURPOSE:To determine a sputtering depth, yield or rate by estimating ion energy and ion current value from discharge conditions, etc., for the glow discharge to execute sputtering and further comparing the same with the data in a high vacuum. CONSTITUTION:The glow discharge is executed with a metallic sample as a cathode and the surface of the sample is sputtered by gaseous ions and is etched in the depth direction. The intensity of the ion energy and the ion current value are respectively estimated by using the relation between the discharge voltage and the energy of the gaseous ion formed by the discharge or the relation between the discharge current and the ion current in the above mentioned glow discharge method. The sputtering yield of the desired sample by the above- mentioned ion energy is then estimated by applying the relation between the ion energy in a known high vacuum and the sputtering yield of the desired element to the glow discharge of a low yield. Further, sputtering depth or sputtering rate is determined by making calculation from this sputtering yield as well as the above-mentioned ion current value and sputtering time.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention is a method for etching a thin film on a solid surface layer or a substrate using a glow discharge ion spunter, and for performing quantitative analysis in the depth direction of a multilayer plating layer. This invention relates to a method for determining the sputter depth, sputter yield, or sputter speed of a metal sample in discharge emission spectroscopy.

[Conventional technology]

The present invention aims to reduce the sputter depth due to glow discharge of metal samples,
Regarding the method of determining sputtering yield (number of sample atoms sputtered per unit ion) or sputtering rate (sputtering depth per unit time), the king of these two methods is to convert them into each other via discharge time and the density and atomic weight of sample elements. Since it is possible, the sputter depth will be described below as a representative example.

Conventionally, the spunter depth of a sample has been determined using one of the following two methods. The first step is to prepare a sample with the same elemental composition as the sample of interest, set the discharge mode to select from constant voltage, constant current, or constant power, and use the type of discharge gas. (pressure settings, discharge voltage, discharge current, or discharge power settings) for a certain period of time or glow discharge, and determine the depth by directly measuring the surface shape of the sample before and after discharge with a stylus needle, or Alternatively, the change in mass of the sample before and after discharge is measured with a scale and converted into depth by calculation using the discharge area determined by the inner diameter of the discharge tube and the density of the sample. The second method is to prepare a sample consisting only of each element in advance for all the elements expected to be contained in the analysis sample, and perform glow discharge emission spectrometry analysis under the same discharge conditions as the analysis sample. , measure the luminescence intensity of each element, and calculate the luminescence amount per unit weight (
Divide the time profile of the luminescence intensity of each element in the analysis sample into minute time units, and calculate the integral value of the luminescence intensity per unit time. Divide by the amount of light emitted per unit weight of each element calculated previously to convert it to the weight of each element per unit time, and further divide by the discharge area and each density to find the spunter depth of each element. Furthermore, the sputter depth is determined by summing the sputter depth of each element (for example,
(Japanese Patent Application Laid-Open No. 60-185142).

[Problem to be solved by the invention]

In the first prior art method described above, it is not possible to determine the sputter depth for a sample whose elemental composition is unknown and the sputter depth per unit time is unknown.

On the other hand, in the second prior art method described above, it is essential to analyze the analytical sample under exactly the same analytical conditions as those under which the luminescence amount per unit weight was previously determined. However, depending on the composition of the sample, the voltage and current of the discharge may not be within the range allowed by the device, and the protection circuit may be activated and the discharge may be stopped, or the emission intensity may be too high or too low, resulting in the same It often happens that discharge cannot occur under certain conditions. In addition, with this method, even if the concentration of the element of interest in the sample, the type of constituent elements other than the element of interest, or the concentration of elements other than the element of interest change, the amount of luminescence per unit weight of the element of interest changes. It is assumed that you will not. In other words, for samples with compositions that have been confirmed to have no change in luminescence amount per unit weight,
Moreover, there is a problem in that it can only be applied to samples that can be analyzed under the same discharge conditions as those used when determining the amount of light emitted per unit weight.

The present invention can also be applied to samples whose sputtering speed is unknown, or whose composition is unknown whether the amount of luminescence per unit weight changes depending on other coexisting elements. The purpose is to provide a method for determining the

[Means to solve the problem]

-i, when determining the sputtering depth, the sputtering yield (number of sample atoms sputtered per unit ion) of the sample (element) under the sputtering conditions (ion species and ion energy), and The ion current value (the number of incident ions is determined from the elementary charge of one ion), the sputtering area, and the sputtering time must be known in the area.

By the way, in the case of ion sputtering in a high vacuum, the energy of the ions is expressed as the product of the acceleration voltage and the ion charge, so the energy of the ions can be determined by the applied voltage. In addition, the ionic current can be directly measured using a Faraday cup. Therefore, by analyzing a sample with known film thickness by secondary ion mass spectrometry or Auger electron spectroscopy while sputtering with a constant ion current, data on the number of atoms sputtered per unit incident ion, that is, the sputtering yield, can be obtained. It is fully equipped for a wide range of sample (target) elements, incident ions, and energy ranges (see example in Figure 2).

However, in the case of glow discharge, it is difficult to measure ion energy and ion current values, so the sputtering yield of each element is also unknown. Therefore, the above method of determining the sputter depth from the sputter yield and ion current value could not be used.

The present invention is based on the following findings regarding the spanter phenomenon of metal samples caused by glow discharge.

(1) The glow discharge voltage and the energy of gas ions formed by glow discharge have a constant functional relationship E=k (V-V.) (1) where E is the ion energy, ■ is the glow discharge voltage, ■
. is the threshold voltage, k is in the proportionality coefficient, and this proportionality coefficient (
k) is the structure of the glow discharge tube and the type of gas used for discharge (Ar, He, NeKr, It□, N2.O
□, chloride gas, mixed gas, etc.) and pressure, and it can be experimentally confirmed that it does not depend on the type of sample metal.

(2) The discharge current consists of an ionic current derived from positive charges carried by gas ions, and a negative charge carried in the opposite direction to the ionic current by secondary electrons released by gas ions colliding with the sample. However, as shown in Figure 1, the ion current value is a constant value that does not depend on the sample element and is determined by the structure of the glow discharge tube, the set voltage, and the type and pressure of the discharge gas. Takes a value.

(3) The ratio between the ion current and the discharge current (the sum of the ion current and the secondary electron current) (ion current ratio) is a constant value that does not depend on the discharge voltage and is determined by the type of sample metal. As shown in Figure 3, this relationship shows that the number of sample atoms sputtered per unit current per unit time and the discharge voltage are (
This was discovered because there is a linear relationship (both in constant voltage mode and constant current mode).

(4) Using the ion energy determined in (1) above, the sputtering yield calculated from the relationship between incident ion energy and sputtering yield in ion sputtering in high vacuum is can be similarly applied. The correctness of this is that the glow discharge sputtering depth calculated using the sputtering yield calculated based on this and the ion current value calculated from (2) or (3) is experimentally calculated. This is proven by the good agreement with the glow discharge sputter depth.

(5) Therefore, using the ion current value obtained in (2) or (3) and the known sputtering yield in high vacuum obtained based on (4) above, the glow discharge time can be It can be converted to depth.

Next, means for achieving the object of the present invention will be explained.

First, A1. Si, Ti, V, Cr, Mn,
Fe, Co, old, CuZn, Zr, Nb, Mo
, Pd, Ag, Cd, In, Sn, Ta1
Prepare a sample made of one of the metal elements with a known sputtering yield in high vacuum, such as llI, Irpt, Δu, and Pb, and perform glow discharge in constant voltage mode at multiple set voltages for a fixed period of time. Measure the change in sample mass before and after and the discharge current.

Draw a relationship diagram between the set voltage and the mass reduction per unit time and unit current due to sputtering (see Figure 4), find the sputtering threshold voltage from its X-intercept, and subtract this from each set voltage to calculate each effective voltage. seek.

The above principle (1) k is constant regardless of the two elements, (3):
The coefficient k in equation (1), which is the relationship between the glow discharge voltage and the energy of gas ions formed by the glow discharge, was calculated by trial and error so that the ion current ratio does not depend on the voltage and is constant for each element. Alternatively, it is determined by graphical calculation method.

Next, regarding the sample consisting of a single metal element of interest,
Glow discharge is performed for a certain period of time at a specific set voltage, and the discharge current and mass change before and after sputtering are measured in the same way as above.

k and ■ found earlier. The ion energy corresponding to the set voltage is determined using The ion current value or ion current ratio (taking into account the discharge current) of the element of interest is determined by comparing it with the measured value of sample mass loss due to sputtering.

Note that the ion current ratio can also be determined by subjecting the elemental sample of interest to ion sputtering in a high vacuum and measuring the ion current and sample current (corresponding to discharge current). In this case, contrary to the above procedure, the ion current ratio is used to calculate the actual value of the sample mass reduction due to glow discharge sputtering using a diagram of the relationship between acceleration energy and sputtering yield (a diagram equivalent to Figure 2).
By comparing E and k in equation (1), it is also possible to determine.

In addition to the above, discoveries (1), (2), (
Based on 3), (4) and (5), the ion current value or ion current ratio can be expressed as E and the coefficient in equation (1), which is the relationship between the glow discharge voltage and the energy of gas ions formed by glow discharge. The sputtering yield can be determined from the known relationship between ion energy and sputtering yield of the metal element in high vacuum ion spacking.

Next, a method for determining the sputtering depth will be described. First, a case where glow discharge is performed in constant voltage mode will be described. A sample made of a single metal element is subjected to glow discharge in constant voltage mode, and the discharge current and discharge time are measured. Ion energy is determined from the set voltage using the previously determined relationship between discharge voltage and ion energy. On the other hand, the previously determined ion current value (or the ion current value from the actual discharge current value using the ion current ratio) is calculated, and the ion sputtering yield of the metal sample in high vacuum by the ion with the ion energy is calculated. The sputtering depth can be determined by calculating the mass loss of the sample due to sputtering from the known data, the discharge time, and the ion current value, and dividing this sputtering loss by the known discharge area and the density of the metal.

Next, the case of performing in constant current mode will be explained.

A sample made of a single metal element is subjected to glow discharge in constant current mode, and the discharge voltage and discharge time are measured.

The ion current value is determined by multiplying the set current value by the previously determined ion current ratio. Next, using the actual measured discharge voltage and the previously determined functional relationship between the discharge voltage and ion energy,
Find the ion energy value.

The mass loss due to the sputtering of the sample is calculated from the known data of the ion sputtering yield of the metal sample in a high vacuum by the ions of the ion energy, the discharge time, and the ion current value, and this sputtering loss is known. The sputtering depth can be determined by dividing the discharge area by the density of the metal.

When performing in constant power mode, the discharge voltage, discharge current, and discharge time are measured, and the sputtering depth can be determined in the same manner as described above.

The above explanation has been about samples made of a single metal, but in the case of samples made of an alloy of two or more metals, the approximate composition can be determined by other means (in the case of glow discharge optical emission spectroscopy, The approximate composition ratio is determined from the ratio of the emission intensities of each metal element), and the sputter yield of the alloy is determined by multiplying the sputter yield of each element determined by the method described above by the composition ratio. How to determine spanter depth.

The above has described a sample whose composition is uniform in the depth direction, but finally, a case of a sample whose composition changes in the depth direction (in the case of glow discharge optical emission spectrometry) will be described.

Draw the luminescence intensity-discharge time profile of each element contained, divide it into minute discharge time segments in which the composition can be approximated as uniform in the depth direction, calculate the spunter depth for each time segment using the method described above, and calculate each time segment. By integrating the values, the sputter depth corresponding to each discharge time can be determined.

[Effect]

As mentioned above, we will clarify the basic governing factors of sputtering caused by glow discharge and relate sputtering caused by glow discharge in a certain functional relationship to the sputtering yield in high vacuum, for which extensive data is already available. By discovering that this is possible, we have made it possible to determine the sputtering depth of a sample during glow discharge from the sputtering yield of the target element in high vacuum and the glow discharge conditions.

As described above, the sputtering yield and sputtering speed during glow discharge are similarly determined.

Examples will be explained below.

[Example 1] A commercially available glow discharge optical emission spectrometer (the inner diameter of the glow discharge tube is 4 mm, the inert gas for discharge is sulfur, the pressure is 9 mm)
．． 2 Torr), pure Ni with a total impurity content of 00 ppm or less was used as the sample, and
As shown in the table, 600.900.
The voltage was set at 1.200 V, glow discharge was performed for 70 to 220 seconds, and the discharge current and the mass of the sample before and after discharge were measured to determine the weight loss due to the spunter. Plot the weight loss due to sputtering per unit current and unit time on the Y-axis and the discharge voltage on the X-axis, and calculate the sputtering threshold voltage from the X-intercept.
o was determined (300 V). Next, at the discharge voltage (V), E was calculated assuming k = 0.3.

The apparent ion current value of glow discharge was calculated from the ion sputtering yield data at this value of E, and the apparent ion current ratio was calculated (see Table 1). Although the apparent ion current ratios at each set voltage differ greatly, they should originally be the same value, and the fact that they do not match indicates that the assumed value of k is incorrect. By calculating the ion current ratio at each set voltage in the same way while changing the value of k little by little (for example, by 0.1), find the value of k that minimizes the difference in the value of the ion current ratio between the set voltages. 0.5 was obtained (see Table 1). Therefore, (1)
The formula can be written as:

E=0.5 (V-300) (2) Next, A
I, Si, Ti+ Cr, Fe, Cu, Mo
, Ta, and H metal plates as samples, constant voltage mode, set voltage 600.900.1200V, 50 to 50
Glow discharge was performed for 0 seconds, and the discharge current and sputter weight loss were measured (see Table 2). The sputtering threshold voltage of each metal was determined in the same manner as in the case of pure Ni.

The ion energy was calculated from the set voltage and the threshold voltage using equation (2), and the sputtering yield of each metal by Ar ions at each energy was determined from known data on sputtering in a high vacuum. The ion current value was calculated from the thus determined sputtering yield and the actually measured sputtering loss, and the ion current ratio of each metal was determined (see Table 2). Even if the set voltage is different, a constant ion current ratio can be obtained for the same metal. Further, the calculated ion current values at the same set voltage are approximately the same for each metal.

Next, use each of the above metal plates as analysis samples in constant voltage mode (600V) or constant current mode (50 to 75m8).
Glow discharge analysis was performed for 100 to 520 seconds at Calculated (3rd
table). When compared with the actual measured value of sputter depth, the third
As is clear from the table, there is a very good agreement between the two.
This shows that the principles (1) to (5) above are correct, and that the method of the present invention allows the sputter depth in glow discharge to be calculated with high accuracy.

[Example 2] Six kinds of high alloy steels containing Cr, N+, Mo+Cu were subjected to glow discharge under the conditions of a constant voltage of 600 cm and an Ar pressure of 9.2 Torr, and sputter etched for 200 seconds.

From the threshold voltage of each element determined in Example 1, the average threshold voltage was determined according to the sample composition, and the average ion energy was determined from equation (2). Next, the sputtering yield of each element at this energy was determined from literature, and the average sputtering yield was determined by further multiplying by the sample composition ratio. On the other hand, the average ion current ratio was obtained by multiplying the ion current ratio of each element by the sample composition ratio, and the ion current value was estimated. Hereinafter, the sputtering depth was determined by calculation in the same manner as in Example 1 (see Table 4). For comparison, the results of actually measuring the mass loss due to sputtering and determining the sputtering depth are also shown.

As is clear from Table 4, values very close to the actually measured values were obtained.

It also shows that principles (1) to (5) on which the present invention is based are correct.

〔Effect of the invention〕

In glow discharge, the current varies depending on the sample element in constant voltage mode, voltage in constant current mode, and voltage and current in constant power mode, so it is difficult to convert sputter time to sputter depth. was difficult. In addition, the ion energy changes depending on the discharge conditions, and
The inability to determine the ion current value made this difficult. In the present invention, sputtering in glow discharge can be reduced to the known sputtering yield in high vacuum through the relationship between discharge voltage and ion energy, discharge voltage, element and ion current value, and element and ion current ratio. Since it was made to relate to the following, it became possible to convert the glow discharge time to the glow discharge sputtering depth.

[Brief explanation of the drawing]

Figure 1 is a diagram showing the relationship between discharge voltage and ion current (sample; 1:
AI, 2:Si, 3:Ti, 4:Cr,
5: Fe, 6: Ni, 7: Cu, 8:
(Mo, 9:Ta, 10:W), Figure 2 shows the relationship between the total ion energy and the spanku yield of Cu in high vacuum, and Figure 3 shows the relationship between the total ion energy and the spanku yield of Cu in high vacuum. A diagram of the relationship between the number of atoms and discharge voltage for samples (old and Cu), Figure 4 shows IV Cr, Ni, Cu,
FIG. 3 is an explanatory diagram of the threshold voltage of Ta.

Claims (1)

[Claims] In the glow discharge method, which uses a glow discharge tube with the sample as a cathode to proceed with etching in the depth direction of the sample surface while sputtering gas ions on the sample surface, the relationship between discharge voltage and gas ion energy, and the discharge voltage or discharge current. The strength of the ion energy and the ion current value are estimated respectively using the relationship between The sputtering depth of a metal sample by glow discharge is estimated using the relationship between the sputtering yield of the element of interest, and further calculates the sputtering depth from the sputtering yield, the ion current value, and the sputtering time. difference,
How to determine sputter yield or sputter speed.