JP4368824B2 - Glow discharge optical emission spectrometer and method - Google Patents

Description

  The present invention relates to a glow discharge emission spectroscopic analysis apparatus and method for analyzing generated light with a spectroscope while sputtering a sample.

  When a direct current or high frequency high voltage is applied between two electrodes in an argon (Ar) atmosphere with a gas pressure of about 500 to 1300 Pa, glow discharge occurs and Ar ions are generated. The generated Ar ions are accelerated by a high electric field, collide with the cathode surface, and knock out the substances present there. Although this phenomenon is called sputtering, the sputtered particles (atoms, molecules, ions) are excited in the plasma and emit light of a wavelength specific to the element when returning to the ground state. An analysis method for identifying elements by spectroscopically analyzing the emitted light is called a glow discharge emission spectroscopy method.

  By the way, in recent years, a sample of a thin film on a substrate has been analyzed using such an analysis method. As shown in FIG. 3, in the measurement result of such a sample, the periodic measurement intensity with respect to the sputtering time is obtained. A wave that is a major change is seen. This is because, for example, the light reflected in the surface of the substrate and the light reflected on the surface of the thin film are strengthened by the interference phenomenon as the thin film gradually becomes thinner. This is because the interference wave is weakened and becomes an interference wave, and the interference wave is measured together with the light generated in the plasma and directly incident on the spectroscope.

On the other hand, there is a method of calculating an average value of measured intensity within a wave period that is a periodic change and removing an interference wave as a corrected measured intensity (see Patent Document 1). And the analysis can be performed sufficiently accurately.
JP 2000-028530 A

  However, in recent years, since the thin film on the substrate can be made extremely thin, in such an analytical sample, when the strength is measured with respect to the sputtering time, only a part of the periodic change appears. Therefore, the correction method of Patent Document 1 that obtains the average of measured intensities in one period of the wave, that is, the time corresponding to the wavelength of the interference wave, cannot be applied.

  The present invention has been made in view of such problems, and a part of the periodic change in the measured intensity with respect to the sputtering time for a sample having a thin film on a substrate and the surface of the substrate or thin film being a mirror surface. An object of the present invention is to provide a glow discharge emission spectroscopic analysis apparatus and method capable of performing an accurate analysis even when the thickness of the thin film is very thin, which only appears.

  In order to achieve the above object, an apparatus according to the present invention includes a glow discharge tube having an anode tube, a thin film on a substrate, and a thin film of a sample having a mirror surface on the surface of the substrate and the thin film. However, in a glow discharge optical emission spectrometer that analyzes a sample based on the measured intensity of generated light, a sputtering time for a specific element in a thin film of a reference sample having a thin film of the same type as the thin film of the sample to be analyzed A reference data storage means for storing a measurement intensity having a periodic change of one cycle or more, and a measurement intensity with respect to a sputtering time for the specific element in a thin film of the analysis target sample. Measured intensity for the specified element in the thin film of the sample to be analyzed with respect to the measured intensity for the specified element in the thin film And it calculates the ratio to sputtering time, and correction means for the measured intensity corrected for sputtering time for the specific elements in the thin film of the sample to be analyzed.

  The method according to the present invention has a thin film formed on a substrate, and a substrate for analyzing the sample based on the measured intensity of light generated while sputtering the thin film of the sample having a mirror surface. In the discharge emission spectroscopic analysis method, in the reference sample thin film having a thin film of the same kind as the thin film of the sample to be analyzed, the measurement intensity with respect to the sputtering time for a specific element and having a periodic change of one cycle or more. Obtain the measurement intensity for the sputtering time for the specific element in the thin film of the analysis target sample in advance, and measure the specific element in the thin film of the analysis target sample for the measurement intensity for the specific element in the thin film of the reference sample The intensity ratio is calculated with respect to the sputtering time, and the And corrected measured intensity to the sputtering time for the specific elements.

  According to the apparatus or method of the present invention, since a reference sample in which a periodic change occurs for one period or more, that is, a reference sample having a thickness equal to or greater than the wavelength of an interference wave caused by an interference phenomenon, the thin film of the reference sample is corrected. The measured intensity can be obtained accurately. Therefore, since the varieties of the thin film of the sample to be analyzed and the thin film of the reference sample are the same, the ratio of the measured intensity in the thin film of the sample to be analyzed to the measured intensity in the thin film of the reference sample is calculated. It is possible to obtain a corrected measurement intensity by removing the interference wave at. As a result, even if the thickness of the thin film of the sample to be analyzed is thinner than the wavelength of the interference wave caused by the interference phenomenon and the periodic change in measurement intensity is less than one period, the interference wave is accurately removed and corrected. The measured intensity can be obtained.

  In the apparatus of the present invention, the corrected measurement intensity with respect to the sputtering time for the specific element in the thin film of the reference sample stored in the reference data storage means was determined by an apparatus different from the glow discharge emission spectrometer. Preferably, the reference sample is corrected by the concentration distribution of the specific element in the thickness direction of the thin film. According to this configuration, even if the concentration of the specific element in the thin film of the reference sample is not constant in the thickness direction, the corrected measurement intensity is further corrected with the concentration distribution, so that the corrected measurement intensity of the analysis target sample is further increased. It will be accurate.

  The program according to the present invention causes a computer included in the glow discharge optical emission spectrometer of the present invention to function as the reference data storage unit and the correction unit.

  Hereinafter, a glow discharge emission spectroscopic analysis method according to an embodiment of the present invention will be described with reference to the drawings. First, an apparatus used for this method will be described. In this apparatus, as shown in FIG. 1, light S emitted from a Grimm glow discharge tube 1 that generates light having a wavelength specific to an element by sputtering utilizing glow discharge and transmitted through the window plate 13 is spectrally dispersed. The light enters the container 22. The spectroscope 22 measures the incident slit 24, the diffraction grating 26 that diffracts the light S incident from the incident slit 24 at different diffraction angles according to the wavelength, the exit slit 27 that passes the diffracted light, and the intensity of the diffracted light. A photomultiplier tube 28 is provided.

  Further, this apparatus uses a hollow anode type grim glow discharge tube 1 as shown in FIG. 2 as a glow discharge tube. The Grimm glow discharge tube 1 includes a support block (a support portion with which the sample 6 to be analyzed 6 or the reference sample 16 is abutted and an insulating portion in the present embodiment) 2 and an anode block 3 having an O-ring. It joins via the sealing members 11. The anode block 3 is integrally formed with a hollow anode tube 3d. The anode tube 3d is inserted into the support block 2 and close to the analysis surfaces (surfaces) 7a and 17a of the sample 6 to be analyzed or the reference sample 16. ing.

  Here, the samples 6 and 16 have thin films 7 and 17 such as TiAlN (titanium aluminum nitride) films on substrates 8 and 18 such as silicon wafers, and are analyzed with the reference sample 16. The thin films 7 and 17 of the target sample 6 are of the same type. The thin film 17 in the reference sample 16 has a thickness in which the above-described periodic change appears for one period or more. For example, if the thickness is 300 nm, an interference wave generated by an interference phenomenon with respect to any element of Ti, Al, and N. The wavelength of is one cycle or more. The analysis target sample 6 is particularly effective when the thickness is thin and the periodic change is less than one cycle, and the thickness is 200 nm, for example. Both the substrates 8 and 18 and the thin films 7 and 17 have mirror surfaces on the surfaces 8a and 18a and the surfaces 7a and 17a. The samples 6 and 16 are pressed against the support block 2 mainly by the cathode block 4 through a seal member 11 such as an O-ring having an annular shape surrounding the portions to be analyzed on the analysis surfaces 7a and 17a. .

  In this way, the opening of the inner space (glow discharge space) V of the support block 2 that accommodates the hollow anode tube 3d by the samples 6 and 16 is sealed, and this inner space V is evacuated (not shown). Thus, vacuuming is performed from the first and second vacuum exhaust holes 3b and 3c. Further, the anode block 3 has an argon gas supply hole 3a, and the inside V of the tube is a rare gas atmosphere (500 to 1300 Pa) of argon gas.

  The Grimm glow discharge tube 1 generates a glow discharge by applying a high-frequency or direct-current voltage between an anode block 3 and a cathode block 4 by a power supply unit (feeding means) 12, and samples 6 and 6 through the cathode block 4. Samples 6 and 16 are sputtered by applying a negative voltage to 16 and causing argon cations generated by the occurrence of glow discharge to collide with the analysis surfaces 7a and 17a of the sample. The coolant is introduced into the jacket from a coolant introduction path (not shown) of the cathode block 4 and fed to the coolant discharge path, thereby cooling the samples 6 and 16 and the hollow anode tube 3d through the cathode block 4. is doing.

  In addition, this apparatus includes an analyzer 15 including a reference data storage unit 31 and a correction unit 32 as shown in FIG. More specifically, the analyzer 15 is a computer and functions as the reference data storage unit 31 and the correction unit 32 by a program. The reference data storage means 31 has a measured intensity with respect to the sputtering time for a specific element in the thin film 17 of the reference sample 16 having a thin film made of the same kind as the thin film 7 of the sample 6 to be analyzed, and the periodic change is one cycle or more. The measured intensity is stored. The correction means 32 obtains the measured intensity with respect to the sputtering time for the specific element in the thin film 7 of the sample 6 to be analyzed, and the specific element in the thin film of the sample to be analyzed with respect to the measured intensity of the specific element in the thin film 17 of the reference sample 16. The ratio of the measured intensities is calculated with respect to the sputtering time, and the measured intensity corrected for the sputtering time for the specific element in the thin film 17 of the sample 16 to be analyzed is obtained.

  Next, the method of this embodiment using this apparatus will be described. First, as described below, the measurement intensity with respect to the sputtering time is obtained in advance for the reference sample 16 and stored in the reference data storage unit 31. Specifically, first, as shown in FIG. 2, the analysis surface 17a of the reference sample 16 is brought into contact with the support block 2, the cathode block 4 is pressed and brought into conductive contact, and the reference sample 16 is held. Further, when the inner space V of the support block 2 is evacuated by a decompression means (not shown) and is brought into a rare gas atmosphere of argon (500 to 1300 Pa), the analysis surface 17a of the reference sample is caused by the atmospheric pressure applied to the back surface 16e. Is pressed against the support block 2 through the seal member 11 and is in close contact therewith.

  When a high frequency or direct current voltage is applied by the power supply unit (feeding means) 12 between the anode tube 3d and the reference sample 16 through the anode block 3 and the cathode block 4, glow discharge is generated, and argon cations are generated. Generated. The reference sample 16 is sputtered by the Ar ions, and the generated light S passes through the window plate 13 and travels toward the diffraction grating 26 of the spectrometer 22 through the entrance slit 24 of FIG. The diffraction grating 26 diffracts light having a predetermined wavelength and enters the photomultiplier tube 28 through the exit slit 27. The photomultiplier tube 28 measures the intensity of incident light.

  As a result, for example, as shown in FIG. 3A, a change in the measured intensity with respect to the specific element with respect to the sputtering time can be obtained and stored in the reference data storage unit 31 in FIG. FIG. 3A shows the measured intensity of only one specific element for simplification, but in practice, each element Al, Ti, N constituting the thin film 17 with respect to the sputtering time is shown. Measurement intensity is obtained. Therefore, the reference data storage means 31 stores the measured intensity with respect to the sputtering time corresponding to FIG. 3A for each element of the reference sample 16.

  Next, for the sample 6 to be analyzed having the thin film 7 of the same kind as the thin film 17 of the reference sample 16, the intensity of light emission by glow discharge is measured in the same manner as the reference sample 16, and each element with respect to the sputtering time is measured. Obtain changes in measured intensity. Here, when the thin film of the sample 6 to be analyzed is thin, for example, 200 nm, as shown in FIG. 3B, the wave of the periodic change in the intensity of the specific element with respect to the sputtering time is less than one cycle. It becomes. Therefore, even if this interference phenomenon is removed to correct the measurement intensity, the average value of the measurement intensity cannot be calculated as in the method described in Patent Document 1 described above.

  On the other hand, in the present invention, even if the wave of the periodic change in the intensity of the specific element with respect to the sputtering time is less than one cycle, the correction means 32 (FIG. 1) has corrected by removing the interference phenomenon. Find strength. In summary, the correction process is a measurement of the specific element in the analysis target sample 6 (FIG. 2) shown in FIG. 3B with respect to the measurement intensity of the specific element in the reference sample 16 (FIG. 2) shown in FIG. Determine the intensity ratio. Hereinafter, the processing necessary for obtaining this ratio will be described in detail.

1) Identification of boundary between thin film and substrate Since thin films 7 and 17 (FIG. 2) become thinner as sputtering progresses, the thickness of the thin film decreases as the sputtering time elapses. Therefore, in the lapse of the sputtering time in FIGS. 3 (a) and 3 (b), when the thickness of the thin film disappears completely, that is, at the boundary between the thin films 7, 17 (FIG. 2) and the substrates 8, 18 (FIG. 2). Exists. The boundary point is, for example, an inflection point in the vicinity where the intensity is greatly reduced, that is, a point at which the second derivative value becomes 0, at time t _{bb in} FIG. 3A and at time t _bs in FIG. is there. It also identifies the time points of the peaks and valleys of the waveform due to interference. This is specified, for example, by parabolic fitting or as the time when the first derivative becomes zero.

2) Adjustment of time axis Next, the time axis which is the horizontal axis of the graphs in FIGS. 3A and 3B is adjusted. In this way, the time axis is adjusted when the sputtering rate is different between the reference sample and the sample to be analyzed, and the elapsed sputtering time in FIGS. 3 (a) and 3 (b) has the same film thickness. This is because it does not correspond to the decrease, and it is made to correspond by adjusting.

Here, the time axis is adjusted with reference to the time points t _bb and t _bs of the boundaries obtained in the above 1), that is, the time points of the boundaries are aligned. Specifically, the time interval T _b between the boundary time t _bb for the reference sample of FIG. 3A and the waveform bottom t t _mb closest to this time t _bb , and the analysis of FIG. The time interval T _s between the time point t _bs at the boundary of the target sample and the time point t _ms at the valley bottom of the waveform closest to this time point t _bs is obtained, and the ratio of T _{s to} T _{b as} shown in the following equation (1) the the time axis adjustment factor C _T. Incidentally, if the thickness of the analysis sample is no valley in the waveform to a very thin, time axis adjustment factor C _T using a time interval between peaks of the point and the boundary point of the waveform in place of the valley of the waveform Find it.

C _T = T _s / T _b (1)

The time axis of the reference sample by using the time axis adjustment coefficient C _T can be adjusted to the time axis of the analysis sample. That is, in the time variation of the measured intensity of the reference sample shown in FIG. 3 (a), by multiplying the time axis adjustment coefficient C _T at a given time may correspond to the analysis of the target sample time. In addition, if this time-axis adjustment coefficient _CT is calculated about one of the elements which comprise a thin film, it can be utilized also in the correction process of another element.

3) Adjustment of swell amplitude Next, the measurement intensity on the vertical axis of the graphs of FIGS. 3 (a) and 3 (b) is adjusted. That is, the height of the waveform shown in FIGS. 3A and 3B, that is, the swell amplitude is adjusted. The reason why the swell amplitude is adjusted in this way is to match the difference in amplitude caused by the difference in measurement conditions between the reference sample and the sample to be analyzed.

Here, the adjustment of the swell amplitude is performed according to the following procedure. First, the intensity variation I _b for the reference sample shown in FIG. 3 (a) is expressed by the following equation (2).

I _b = a _b f (x) + b _b (2)

However,
f (x): a function indicating a waveform, and the difference between the maximum value and the minimum value is 1.
a _b : (Intensity of the peak of the waveform closest to the boundary time point in the reference sample) − (Intensity of the valley bottom of the waveform closest to the boundary time point in the reference sample)
b _b : (intensity of the valley bottom of the waveform closest to the boundary time point in the reference sample) + a _b / 2

The intensity variation I _s for the analysis sample shown in FIG. 3 (b) is expressed by the following equation (3).

I _s = a _s f (x) + b _s (3)

However,
f (x): a function indicating a waveform, which is the same as the function in the reference sample.
a _s : (intensity of the peak of the waveform closest to the time of the boundary in the sample to be analyzed) − (intensity of the valley bottom of the waveform closest to the time of the boundary in the sample to be analyzed)
b _s : (intensity of valley bottom of waveform closest to boundary time point in analysis sample) + a _s / 2

Here, the fact that match the undulation amplitude of the reference sample undulation amplitude of the sample to be analyzed is adjusted undulation amplitude of the reference sample is that it is a _s. Therefore, the measured intensity I _b ′ composed of the adjusted swell amplitude of the reference sample is expressed by the following equation (4).

I _b ′ = a _s f (x) + b _b (4)

  Substituting f (x) obtained from Equation (2) into Equation (4) yields Equation (5) below.

I _b ′ = AI _b + B (5)

However,
_{A =} a s / a _b
B = b _b (1-A)

Since the measurement intensity I _b ′ of the reference sample having the undulation amplitude thus adjusted can be obtained, the undulation amplitude of the reference sample can be matched with the undulation amplitude of the sample to be analyzed. That is, by applying Equation (5) to the measurement intensity of the reference sample shown in FIG.

4) Standardization of the measurement intensity of the reference sample Next, obtain the average value of the measurement intensity of the reference sample. Specifically, the average value I _ab of the measured intensity from the time point t _ab after a lapse of the sputtering start time in FIG. 3A to the time point t _{bb of} the boundary obtained in the above 1) is obtained. As described above, the average value I _ab of the measurement intensity can be obtained in the reference sample because, as shown in FIG. 3A, the reference sample has a periodic change wave of one or more cycles as shown in FIG. It is because it has thickness.

After obtaining the average value I _ab of the measurement intensity, the measurement intensity of the reference sample shown in FIG. 3A is divided by this average value I _ab to make the average value 1, and the measurement intensity of the specific element in the reference sample To standardize. The process of dividing the measured intensity by the average value I _ab is performed every predetermined time. For example, each intensity measured during sputtering is divided, that is, performed every time the intensity is measured. This predetermined time interval is, for example, 20 ms.

5) Calculate the ratio of the measured intensity In the state where the processes 2) to 4) are performed, the ratio of the measured intensity of the sample to be analyzed to the measured intensity of the reference sample is obtained. Specifically, the time points t _bb and t _bs of the respective boundaries between the reference sample and the analysis target sample are set as the start time points, and during the correction time Tad until the time point t _as a little after the sputtering start time of the analysis target sample, A ratio of measured intensities is obtained every predetermined time in a direction going back the sputtering time. The predetermined time for obtaining the intensity ratio is also the time when the intensity is measured during the sputtering as in the case of obtaining the average value _Iab . Here, the correction time Tad is for reference samples as described in 2) above, a correction time Tad of a state performing the adjustment of the time axis is multiplied by the time axis adjustment factor C _T. Further, the measurement intensity of the reference sample for obtaining the ratio of the measurement intensity is the measurement intensity I _b ′ composed of the adjusted swell amplitude as described in the above 3). In this embodiment, the waveform of the reference sample is adjusted with respect to the time axis and the waviness amplitude, but the waveform of the sample to be analyzed may be adjusted.

The ratio obtained as described above is the corrected measurement intensity of the sample to be analyzed. The reason why the intensity of the sample to be analyzed can be derived simply by obtaining the ratio in this way is that the average intensity I _ab of the reference sample is set to 1 by the standardization of the measurement intensity of the reference sample in 4) above.

As described above, the measurement intensity correction target is only the correction section Tad, and the above correction is not applied after the boundary time _tbs . This is because the measured intensity near the boundary is easily affected by the surface state of the substrate and exhibits different characteristics. Examples of the correction of the measured intensity in the vicinity of the boundary, the corrected measured intensity I _c can be expressed by the following equation (6).

I _c = α (I−I ₄₀ ) + I ₄₀ (6)

here,
α = (I _s −I ₄₀ ) / (I _top −I ₄₀ ) (7)

However,
I _s : Corrected measurement intensity at the peak of the waveform closest to the boundary time point I ₄₀ : Intensity of 40% of I _s I _top : Measurement intensity at the peak of the waveform closest to the boundary time point I: Measurement intensity

The equation (6) is the measured intensity can be applied to the point where the I _40, thereafter not adding the correction to the measured intensity. Further, instead of applying the equation (6) to correct the measurement intensity near the boundary, the measurement intensity near the boundary can be not particularly corrected.

6) Correction of density gradient It is desirable to arbitrarily correct the density gradient after the correction processes 1) to 5) above. Specifically, since the concentration of the specific element is usually not constant in the thickness direction of the film, the correction strength obtained in the above 5) is further corrected. Specifically, the concentration distribution of the specific element in the thickness direction of the thin film of the reference sample is obtained in advance by a method different from the glow discharge optical emission spectrometry, for example, secondary ion mass spectrometry (SIMS). Based on the density distribution thus obtained, that is, the density gradient, the intensity obtained in the above 5) is corrected.

  An example in which the intensity of the sample to be analyzed is corrected by the correction procedure shown in the above 1) to 6) will be described. FIG. 4A shows a change in the measured intensity before correction with respect to the sputtering time, and FIG. 4B shows a change in the measured intensity after correction with respect to the sputtering time. Here, the thin film is TiAlN, the thin film of the reference sample has a thickness of 300 nm, and the thin film of the sample to be analyzed has a thickness of 200 nm. The waveform due to interference appearing in FIG. 4A is removed in FIG. 4B.

  As described above, the measurement intensity corrected by removing the intensity due to the interference with respect to the specific element of the sample to be analyzed can be obtained, and thus the exact concentration of the specific element can be obtained.

It is a front view which shows the apparatus used for the glow discharge emission-spectral-analysis method of one Embodiment of this invention. It is a longitudinal cross-sectional view which shows the glow discharge tube of an analyzer same as the above. (A) is a figure which shows the change of the measurement intensity | strength about the specific element with respect to sputtering time in a reference | standard sample, (b) is a figure which shows the change of the measurement intensity | strength about the specific element with respect to sputtering time in a sample to be analyzed. (A) is a figure which shows the measurement intensity | strength before correction | amendment about the specific element with respect to sputtering time, (b) is a figure which shows the measurement intensity | strength after correction | amendment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Glow discharge tube 3d Anode tube 6 Sample to be analyzed 7 Thin film of sample to be analyzed 8, 18 Substrate 16 Reference sample 17 Thin film of reference sample 31 Reference data storage means 32 Correction means

Claims (4)

Sample analysis based on the measured intensity of light generated while sputtering a thin film of a sample having a glow discharge tube having an anode tube, a thin film on the substrate, and the surface of the substrate and the thin film being mirror-like In a glow discharge optical emission spectrometer that performs
Reference data storage means for storing a measurement intensity with respect to the sputtering time for a specific element and having a periodic change for one period or more in a reference sample thin film having a thin film of the same kind as the thin film of the sample to be analyzed ,
In the thin film of the analysis target sample, the measurement intensity with respect to the sputtering time for the specific element is obtained, and the ratio of the measurement intensity for the specific element in the thin film of the analysis target sample to the measurement intensity for the specific element in the thin film of the reference sample is sputtered. A glow discharge emission spectroscopic analysis apparatus, comprising: a correcting unit that calculates the time and calculates a corrected measurement intensity with respect to a sputtering time for the specific element in the thin film of the sample to be analyzed. In claim 1,
The thickness of the thin film of the reference sample, which was obtained by using a device different from the glow discharge emission spectroscopic analysis device for the corrected measured intensity with respect to the sputtering time for the specific element in the thin film of the reference sample stored in the reference data storage means A glow discharge optical emission spectrometer that is corrected by the concentration distribution of a specific element in the direction.   A program for causing a computer included in the glow discharge emission spectroscopic analysis apparatus according to claim 1 or 2 to function as the reference data storage unit and the correction unit. In a glow discharge emission spectroscopic analysis method for analyzing a sample based on the measurement intensity of light generated while sputtering a thin film of a sample having a thin film on a substrate and the surface of the substrate and the thin film being a mirror surface,
In the thin film of the reference sample having a thin film of the same type as the thin film of the sample to be analyzed, the measurement intensity with respect to the sputtering time for the specific element and having a periodic change for one period or more is obtained in advance.
In the thin film of the analysis target sample, the measurement intensity with respect to the sputtering time for the specific element is obtained, and the ratio of the measurement intensity for the specific element in the thin film of the analysis target sample to the measurement intensity for the specific element in the thin film of the reference sample is sputtered. A glow discharge emission spectroscopic analysis method, characterized in that it is calculated with respect to time and is a corrected measurement intensity with respect to sputtering time for the specific element in the thin film of the sample to be analyzed.

Images