JP3389115B2 - Stacked structure inspection method and X-ray reflectivity device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention measures the X-ray reflectance of a thin film laminate having two or more layers formed on a substrate, analyzes the obtained reflection profile, and determines the film thickness of each layer of the laminate. TECHNICAL FIELD The present invention relates to a laminated structure inspection method and a film formation control method which are nondestructive in density and density and can be analyzed with high accuracy, and an apparatus using the same.

[0002]

2. Description of the Related Art In the fields of semiconductor integrated circuits, devices using the same, magnetic files, etc., semiconductor layers, insulating layers, and metal layers are laminated to form elements by patterning.

The film formed for the purpose of high performance and high performance of the element has become extremely thin and the number of laminated films has been increasing. Since the film thickness and the density of such a laminated body have a great influence on the characteristics of the element, it is necessary to improve the film formation controllability and to provide a highly accurate laminated structure evaluation method.

The X-ray reflectance method is an effective method capable of nondestructively analyzing the layer structure of a laminate. FIG. 10 shows a conventional reflectance measuring device, and FIG. 11 shows its analyzing method. After the X-ray from the X-ray source 1 is separated by the crystal spectroscope 3, it is incident on the sample 6 at an oblique incident angle θ, and the reflected X-ray from the sample 6 is detected by the detector 10.
Detect with.

The reflectance is measured from the θ / 2θ scan in which the rotary table 5 is driven by the control unit 11. The slits 2 and 4 limit the incident X-rays, and the slits 7 and 9 limit the reflected X-rays. In addition, the solar slit 8 has a reflection X
This is for improving the parallelism of the lines. The obtained reflectance is analyzed by the analysis device 12, and the results of the density and film thickness of each layer are output to the output device 13.

The analysis method shown in FIG. 11 is a generally used algorithm for optimizing the complex refractive index (proportional to the density), film thickness and interface width of each layer by the least squares method.

[0007]

The X-ray reflectivity method is a method for analyzing a vibrating structure (see FIG. 3) in a reflectivity profile caused by interference of X-rays reflected on the surface of the laminate and each interface. If the difference in refractive index (difference in density) between adjacent films is small, the intensity of reflected X-rays at the interface will be small, and analysis accuracy will be reduced for each film, and analysis will be impossible depending on the material composition of the thin film. There was a problem that said.

In order to overcome this problem, it is possible to utilize the effect of anomalous dispersion of the refractive index of X-rays by the constituent material of the laminated body.
As disclosed in Japanese Patent Application Laid-Open No. 0-38821, the number of laminated films is increased (5 layers or more) and the film itself is made extremely thin (several nm).
Inevitably, the problem that the analysis accuracy will decrease with the following).

Further, Japanese Patent Laid-Open No. 64-20405 reports a method of measuring the film thickness from the mutual phase of reflection intensity modulation in X-rays of two or more kinds of wavelengths. Since the phase shift due to the multiple scattering and absorption of X-rays at 1 is not taken into consideration, there is a problem that the film thickness can be simply obtained, but the accuracy is not sufficient. Furthermore, this method can be applied only to a single-layer film, not to a laminate having two or more layers.

The object of the present invention is to use X-rays of two or more wavelengths and to calculate the reflectance based on a laminated structure model.
A laminated structure inspection method that can evaluate the laminated structure with higher accuracy than the conventional method by an optimization method that minimizes the residual sum of squares with the ratio of reflectance measured at each wavelength, and the method were used. The equipment is provided.

Further, the magnetoresistive sensor mounted on the magnetic recording / reproducing apparatus is composed of a magnetic and non-magnetic thin film laminated body having a thickness of several nm, and the magnetic characteristics of the sensor strongly depend on the film thickness of each layer.

Therefore, the next object of the present invention is to control the film thickness by the present inspection method with high accuracy, and to perform magnetic recording / reproduction using a magnetoresistive sensor having stable performance by the present control method. The equipment is provided.

[0013]

The feature of the laminated structure inspection method according to the present invention to achieve the above object is that X-rays are incident on a laminated body in which two or more thin films are formed on a substrate at a low angle θ. In the laminated structure inspection method for measuring the X-ray reflectance from the body and inspecting the layer structure of the laminated body by analyzing the reflectance, two or more kinds of the laminated bodies constituting the laminated body are incident on the X-rays. It is to use two or more kinds of elemental Kβ rays.

Further, X-rays are made to enter the thin film laminate formed on the substrate at a low angle θ, the X-ray reflectance from the laminate is measured, and the reflectance calculated based on an appropriate laminated structure model is used. By the optimization method that minimizes the residual sum of squares with the measured reflectance,
In the laminated structure inspection method for analyzing the laminated structure of the laminated body,
The reflectance is measured by using two kinds of wavelengths λ ₁ and λ ₂ as the X-ray wavelength, and the angle θ of each reflectance is defined as the scattering vector magnitude q.
After converting to (= 4π sin θ / λ), the q dependence of the reflectance ratio for the same q is calculated, and the dependence is optimized to inspect the layer structure of the laminate.

Further, two or more kinds of wavelengths for the X-ray wavelength λi =
The reflectance using 1 to n is measured, and the residual sum of squares of each wavelength χ
The optimization method that minimizes the sum Σχ ² i of ² i, is to inspect the layer structure of the laminate.

Further, an X-ray source for generating two or more types of characteristic X-rays used for measurement, a spectroscope, a driving base for the sample and the detector, a detector and a control unit are provided, and the sample installed on the driving base is provided. Irradiate X-rays and measure reflected X-rays from the sample X
The linear reflectance device includes a spectroscopic element that extracts only specific characteristic X-rays by two reflections, a drive unit that rotates and translates the spectroscopic element, and a control unit that controls the drive unit by the control unit. And an X-ray reflectance device provided with an analyzing means for performing an analysis based on the above-mentioned laminated structure inspection method.

Further, the feature of the film formation control method according to the present invention is that the film thickness of each film is 0.2 by using the above-mentioned laminated structure inspection method.
It is a film forming control method for controlling the surface roughness with a precision of 4 nm or less, the interface roughness of 4 nm or less, and the density of 4% or less.

Further, in a film forming apparatus for forming two or more thin films on a substrate, an X-ray source, a spectroscope, a sample and a detector driving stand for generating two or more kinds of characteristic X-rays used for measurement, An X that includes a detector and a control unit, irradiates a sample installed on the drive table with X-rays, and measures reflected X-rays from the sample.
The linear reflectance device includes a spectroscopic element that extracts only specific characteristic X-rays by two reflections, a drive unit that rotates and translates the spectroscopic element, and a control unit that controls the drive unit by the control unit. A film forming apparatus comprising the above X-ray reflectance device, wherein the transport time of the laminate between the X-ray reflectance device and the film forming device according to the control method is within 10 minutes. is there.

Furthermore, a magnetic recording medium, a magnetoresistive sensor for reading a magnetic signal written on the medium, a write head for writing a magnetic signal, the magnetoresistive sensor and the write head are mounted on a tip portion of the magnetic recording medium. In a magnetic recording / reproducing apparatus comprising an arm for driving in the radial direction of a recording medium, a control unit for controlling the magnetic recording medium and a driving unit for the arm, and a signal processing unit for processing a signal for reading and writing a magnetic signal, In the magnetic recording / reproducing apparatus, the thin film laminated body of the magnetoresistive sensor is constituted by the magnetoresistive sensor formed by the film forming apparatus.

[0020]

DETAILED DESCRIPTION OF THE INVENTION According to the present invention, X of two or more wavelengths is used.
Since the reflectance by the line is used, it is possible to evaluate the laminated structure with higher accuracy than the conventional reflectance of one wavelength.
The above reason will be described below.

Consider the X-ray reflectivity from a laminate in which an n-1 layer film is formed on a substrate. Complex refractive index n = 1-δ-i
β, the oblique angle θ of the incident X-ray, the X-ray wavelength λ, the film thickness t, and β
Using the approximation of ˜0, θ ² >> 2δ, θ << 1, the reflectance R is expressed by the following equation.

[0022]

[Equation 1]

[Mathematical formula-see original document] [gamma] in the equation (1) is a quantity representing the phase of the reflected X-ray and is a function of the film thickness. In Equation 1, the first sum term is the reflection from the surface and the interface, and information about the film thickness cannot be extracted from this term.

On the other hand, the second sum term is due to the interference of X-rays reflected on the surface and the interface, and a vibrating structure appears in the reflectance (see FIG. 3). This vibrating structure contains information on the film thickness and density of each film of the laminate.

Conventionally, the reflectance including the vibrating structure measured at one wavelength is analyzed. As is clear from Equation 1, the reflectance is attenuated by the −4th power of θ, so a slight difference in the vibration structure will be analyzed. On the other hand, rewriting Equation 1 using the variable transformation of the scattering vector magnitudes q = 4π sin θ / λ and ξ = (4π / λ) ² δ gives the following equation.

[0026]

[Equation 2]

Due to this variable change, since λ is not directly included in the above equation, the periods of the vibrating structure of the reflectance measured at each wavelength are almost the same (see FIG. 5).

Therefore, by taking the ratio of the reflectances R ₁ and R ₂ measured at the _two X-ray wavelengths λ ₁ and λ ₂ , the −4th power attenuation term is eliminated and the vibration structure can be extracted. Therefore, the accuracy is improved (see FIG. 6).

FIG. 1 shows an analysis algorithm according to the present invention. The reflectance measured at two wavelengths is the magnitude of the scattering vector q
Then, the structure of each laminated film is optimized by fitting the ratio by the method of least squares. In this case, the ratio of reflectance at each wavelength is calculated using the X-ray scattering theoretical formula.

Another analysis algorithm of the present invention with improved analysis accuracy is shown in FIG. The point that the reflectance measured at two wavelengths is directly analyzed is the same as the conventional analysis of reflectance measured at one wavelength. However, it differs from the conventional method in that the sum of the residual sums of squares calculated at each wavelength is minimized as the optimization condition judgment. Since the refractive index depends on the wavelength,
The two wavelengths have different amplitudes of the vibrating structure. In the present invention, the accuracy can be improved because the different vibration structures are analyzed together.

Embodiments of the present invention will be described below with reference to the drawings.

Example 1 First, it was confirmed by computer simulation that the accuracy of the inspection method according to the present invention was improved as compared with the conventional method. The structure of the laminate is of three types shown in Table 1.

[0033]

[Table 1]

Calculation of the reflectance from the laminate is described in L. G.
Paratt (Phys. Rev. 95, 359 (195
4)) and L.S. Nevot and P.N. Croce (Rev.
Phys. Appl, 15, 761 (1980)), using the theoretical elements of the laminate, Cu and Co, which are Kβ rays, Cu-Kβ (0.13922 nm) and Co.
Calculated at two wavelengths of -Kβ (0.16208 nm).

In the calculation, the complex refractive index of each layer is a reference value (S
Asaki, KEK Report, 88-14) was used, and each interface width was calculated as 0.5 nm.

FIG. 3 shows the reflectance (Cu-K) from each laminate.
β) is shown. Solid lines, dotted lines, and broken lines are laminated bodies a, b, and c, respectively.
Therefore, there is only a slight difference in the reflectance of each laminate. It is obvious that high-precision analysis is difficult to analyze this reflectance by the conventional method.

On the other hand, FIG. 4 shows the ratio of reflectances according to the present invention for two wavelengths. The vertical axis represents Co-Kβ reflectance / Cu-Kβ
The reflectance and the horizontal axis are the magnitude q of the scattering vector. solid line,
Dotted lines and broken lines are laminated bodies a, b, and c, respectively. In Figure 4,
The oscillating portion of the reflectance is extracted, and its amplitude is emphasized compared to FIG.

Further, it can be clearly seen that the amplitude of vibration differs depending on the laminated body. Since this vibration structure includes information on the layer structure, the inspection method according to the present invention (see FIG.
It is apparent that the layer structure of the laminated body can be analyzed with higher accuracy than in the conventional case according to the reference).

Example 2 It was confirmed by computer simulation that the analysis method according to the present invention is more accurate than the conventional method. The structure of the laminated body is Co (4) /
Cu (2) / Co (4) / Ta (5) / Si substrate, and the value in parentheses is the film thickness (nm).

FIG. 5 shows Cu-Kα (0.15406 nm).
And the reflectance calculated at the wavelengths of Cu-Kβ are shown by a solid line and a dotted line, respectively. By changing the wavelength, a slight difference in reflectance can be observed. In the conventional method, this reflectance is analyzed individually.

On the other hand, according to the present invention, an algorithm for minimizing the sum of residual square sums calculated at each wavelength (see FIG. 2).
Since this is used, the accuracy of analysis is improved because optimization is performed while maintaining the consistency of the vibrating portion of the reflectance measured at each wavelength.

That is, in the reflectance at each wavelength, the film thickness and the interface width of each layer are common, and the complex variable index having wavelength dependency is optimized at each wavelength. Therefore, common variables (film thickness, interface width) are used.
Can be consistent. Therefore, it has been clarified that the analysis method according to the present invention is more accurate than the conventional method.

Further, in FIG. 6, Cu-Kβ reflectance / Cu-
The reflectance ratio of the Kα reflectance is also shown. The oscillating portion of the reflectance is extracted, and its cycle and amplitude show a complicated shape. This is in sharp contrast to the reflectance shown in FIG. 5, and it can be seen that the slight difference in reflectance is more emphasized. That is, it is shown that the analysis method based on the reflectance ratio according to the present invention is also effective.

[Embodiment 3] FIG. 7 shows X of an embodiment of the present invention.
It is a block diagram of a linear reflectance device. X-rays having two or more kinds of wavelengths from the X-ray source 1 are extracted by the crystal spectroscope 3 and then incident on the sample 6. Reflected X-rays from the sample 6 are slits 7 and 9 and a solar slit 8
After shaping and parallelizing with, the measurement is performed with the detector 10.

Next, each function will be described. In the X-ray composite target 20, a rotating anticathode using pure metal of Co and Cu was used in the rotating part irradiated with the electron beam from the filament 21.

The operating conditions of the X-ray source are a tube voltage of 45 kV and a tube current of 200 mA. From this X-ray source, Co-Kα,
Characteristic X-rays of Kβ and Cu-Kα, Kβ are emitted.
At this time, the strengths of Co-Kβ, Cu-Kα, and Cu-Kβ are about 500 kcps, 6000 kcps, and about 100 kcps, respectively.
It was 0 kcps. The crystal spectrometer 3 has Si (111)
The channel-cut type crystal was used. The control unit 11 has a function of rotating and translating the crystal.

In order to extract the X-rays of the Co-Kβ wavelength from the X-ray source 1, the oblique angle of the X-rays with respect to the crystal is 14.98.
The sample 6 was irradiated with X-rays which were dispersed by adjusting the slit 4 position.

Further, the X-ray of the Cu-Kβ wavelength from the X-ray source 1
In order to extract the X-ray, the visual oblique angle to the X-ray crystal is set to 1
The angle was set to 2.83 °, and the crystal was translationally driven to adjust the slit 4 previously set so that the spectroscopic X-ray was transmitted.

As a result, the incident angle and the irradiation position of the X-ray of each wavelength on the sample could be matched. The rotary table 5 is composed of two axes of θ and 2θ, and the sample holding stage provided on the θ table is driven by the control unit 11 to make the sample incident X.
Can be aligned parallel to the line. As the detector 10, a scintillation counter having an opening diameter of 1 inch was used. Each drive unit can be controlled by the control unit 11.

Next, the measurement and analysis procedure will be described. First, the sample is retracted from the optical axis, the crystal spectroscope 3 is adjusted so that X-rays of each wavelength can be extracted, and the set angle and translational position are recorded.

Next, after returning the sample 6 to the optical axis and aligning the sample so that it is parallel to the X-ray, the sample is rotated to a set angle (θ) and the 2θ arm is set to a predetermined angle (2θ). Rotate.

In the measurement, the reflected X-ray intensity is measured by θ-2θ scanning at a step angle of 0.01 ° of the sample. First C
The reflectance was measured at the o-Kβ wavelength.

Next, the sample and the detector were returned to the initial setting values, the crystal spectroscope 3 was set to the Cu-Kβ wavelength, and the reflectance was measured. The scanning angle range of the sample is 0.15 ° to 3.
It was set to 0 °.

The drive of the rotary table 5 was controlled by the control unit 11, and the signals from the detector 10 were sequentially recorded in the analysis device 12 via the control unit 11. The sample is Ta
(30) / CrMnPt (300) / Co (30) / C
u (23) / Co (10) / NiFe (50) / Ta
In (50) / substrate, the value in parentheses is the design film thickness (Å).

As the wavelength of the incident X-ray, Kβ rays of Co and Cu, which are the constituent elements of the sample, were used. The analysis uses the calculation code based on the algorithm shown in FIG.
The above was performed, and the result was output to the output device 13.

The layer structure model used in the analysis is oxide layer / T
a / CrMnPt / Co / Cu / Co / NiFe / reaction layer / Ta / interface layer / processed layer / substrate, the refractive index of each layer,
The film thickness and interface width were optimized.

FIG. 8 shows the reflectance measured at each wavelength. In this analysis, an algorithm that minimizes the sum of the reflectance measured at each wavelength and the calculated reflectance was used. For comparison, Co
Only the reflectance at −Kβ was analyzed using the usual algorithm.

FIG. 9 shows the ratio (χ ² / χ) of the residual sum of squares χ ^{2 to} the minimum value when the film thickness of Co on the substrate side is fixed to a value deviated from the optimum value and other variables are optimized again. ² min) is shown for two wavelengths and one wavelength.

It can be seen that the χ ² distribution for one wavelength is almost flat and does not easily converge to the minimum value. On the other hand, χ at 2 wavelengths
It can be seen that the ² distribution has a downwardly convex shape and tends to converge to the minimum value in the optimization. From this, it has been found that the accuracy is improved when the analysis is performed with two wavelengths.

Next, the error including the measurement and analysis by the two-wavelength method was examined. The above sample was 3 with Co-Kβ
The reflectance data measured twice with Cu-Kβ was combined and analyzed using an algorithm that minimizes the sum of the measured reflectance and the calculated reflectance at each wavelength. as a result,
Table 2 shows the average values and dispersions of ξ, the film thickness d, and the interface width σ, in which the refractive index is changed.

From this, the refractive index (proportional to the density) is about 0.4%, and the Cu film thickness is about 0.2Å (= 0.02n).
m), CrMnPt interface width is about 0.4Å (= 0.04n)
It was found that analysis can be performed with the accuracy of m).

[0062]

[Table 2]

[Embodiment 4] Next, a film formation control method and a magnetoresistive sensor will be described. An RF sputtering device is used as a film forming device, and CrMnPt, Co, Cu,
The sputter rate of each target of NiFe and Ta was calibrated.

First, using the sputtering rate incorporated in the control unit of the sputtering apparatus, 1, 2,
A film having a designed film thickness of 4, 8 nm was formed. CrMnPt was formed with a designed film thickness of 10, 20, 40, and 80 nm.

Next, these calibration laminates were measured and analyzed using the X-ray reflectivity apparatus of Example 3, and the film thickness of each layer was analyzed. From the analyzed film thickness and film formation time, the sputter rate of each target was fitted by linear regression for calibration. In this operation, the sputtering rate was calibrated under each film forming condition in which the RF power of the RF sputtering device and the pressure in the chamber were changed.

By these operations, the film thickness of 0.2 nm or less,
It became possible to control the interface unevenness with an accuracy of 0.4 nm or less and a density of 4% or less. That is, since the analysis accuracy of the two-wavelength reflectance method according to the present invention is sufficiently high, the film formation control can be performed with the above accuracy.

Further, the film forming apparatus and the reflectance apparatus described above are installed in a clean room, and the transportation time of the calibration laminate is 1
Within 0 minutes. Since the transportation time is short, the sputtering apparatus can be quickly calibrated, and the influence of the oxide layer formed on the surface is small, and highly accurate analysis is possible.

Next, several magnetoresistive sensors were manufactured using the above-described calibrated film forming apparatus. Membrane composition is Ta (3)
/CrMnPt(30)/Co(3)/Cu(2.3)
/ Co (1) / NiFe (5) / Ta (5) and the value in parentheses is the design film thickness (nm).

After film formation, the film thickness of each sensor was measured and evaluated by an X-ray reflectometer, and the film thickness was 0.2 nm or less, which was in agreement with the designed film thickness. Since the film thickness of each sensor was controlled with sufficient accuracy, the sensitivities of the sensors were almost the same.

FIG. 12 shows a schematic view of a magnetic recording / reproducing apparatus incorporating the above sensor. FIG. 12A shows an arm in which a magnetic recording medium on which information is written, a magnetoresistive sensor for reading a magnetic signal, and a write head for writing a magnetic signal are mounted at the tip end. Information can be recorded and reproduced at all locations on the magnetic recording medium by a magnetic recording medium rotating around the center of the disc and an arm that is driven in the radial direction of the disc.

FIG. 12B shows the structure of the magnetoresistive sensor and the write head. The part surrounded by ∘ indicates the magnetoresistive sensor, and this surface is close to the surface of the magnetic recording medium. Read the recording signal.

FIG. 12C shows the film structure of the magnetoresistive sensor. This is a structure in which a magnetization signal is read by a change in electric resistance between electrodes due to a leakage magnetic field of magnetization written in a magnetic recording medium. In each magnetic recording / reproducing apparatus incorporating this sensor, since the sensitivities of the sensors are well matched, the fluctuation range of the output signal was small, and the yield was good as a recording apparatus.

According to the present invention, since the film thickness of the laminated body, which is the main element of the magnetoresistive sensor, can be controlled with high accuracy, the performance of each sensor can be suppressed within the allowable range, and the magnetic properties with stable performance can be suppressed. It is possible to produce recording / reproducing devices with steel yield.

[0074]

According to the thin film laminate analysis method of the present invention, the vibrating portion can be extracted by taking the ratio of the reflectances measured at two or more wavelengths, and the sum of the residual sum of squares at each wavelength can be obtained. Since the minimum algorithm is used, there is an excellent effect that the laminated structure inspection can be performed with higher accuracy than the conventional one.

Further, according to the control method of the present invention, since the film thickness of the laminated body, which is the main element of the magnetoresistive sensor, can be controlled with high accuracy, the performance of each sensor can be suppressed within the allowable range, A magnetic recording / reproducing device with stable performance can be produced with high yield.

[Brief description of drawings]

FIG. 1 is a diagram showing an analysis algorithm according to the present invention.

FIG. 2 is a diagram showing an analysis algorithm according to the present invention.

FIG. 3 shows the calculated reflectance (C) from the laminate shown in Table 1.
It is a figure which shows (u-Kb wavelength).

FIG. 4 shows the Co—Kb reflectance / in the laminate shown in Table 1.
It is the figure which calculated the ratio of Cu-Kb reflectance.

FIG. 5 is a diagram showing reflectance calculated from wavelengths of Cu-Kα and Cu-Kb from a Co / Cu / Co / Ta / Si substrate laminate.

FIG. 6 shows a reflectance ratio (Cu-Kb reflectance / C shown in FIG.
It is the figure which calculated u-K (alpha) reflectance.

FIG. 7 is a configuration diagram of an X-ray reflectance device according to an embodiment of the present invention.

FIG. 8 Ta / CrMnPt / Co / Cu / Co / Ni
Fe-Ta / Co-Kb and Cu-from the substrate stack
It is a figure which shows the reflectance measured at the wavelength of Kb.

9 is a C on the substrate side in the analysis of the laminated body shown in FIG.
FIG. 5 is a diagram showing the results of analysis of the distribution of the residual sum of squares χ ² with respect to the film thickness by the two-wavelength method and the conventional one-wavelength.

FIG. 10 is a configuration diagram of a conventional X-ray reflectance device.

FIG. 11 is a diagram illustrating a conventional analysis algorithm.

FIG. 12 is a schematic diagram showing a configuration of a magnetic recording / reproducing apparatus of the present invention.

[Explanation of symbols]

1 ... X-ray source, 2, 4, 7, 9 ... Slit, 3 ... Spectrometer,
5 ... rotary table, 6 ... sample, 8 ... solar slit,
10 ... Detector, 11 ... Control part, 12 ... Analysis device, 13 ...
Output device, 15 ... Spectrometer drive unit, 20 ... X-ray composite target, 21 ... Filament.

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takao Imagawa 2880, Kozu, Odawara-shi, Kanagawa Hitachi Ltd., Storage System Business Department (72) Inventor, Kazuhiro Ueda 7-1, 1-1 Mikamachi, Hitachi City, Ibaraki Prefecture Issue Hitachi, Ltd. Hitachi Research Laboratory (56) Reference JP 10-38821 (JP, A) JP 64-20405 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) ) G01N 23/207 G01B 15/02 G11B 5/39

Claims (2)

(57) [Claims] 1. An X-ray is incident on a laminated body in which two or more thin films are formed on a substrate at a low angle θ, the X-ray reflectance from the laminated body is measured, and based on an appropriate laminated structure model. In a laminated structure inspection method for inspecting the layer structure of the laminate by an optimization method that minimizes the residual sum of squares of the calculated reflectance and the measured reflectance, two or more wavelengths (λi = 1-n), the reflectance is measured, the residual sum of squares χ ² i is obtained for each wavelength, and i
, 1 to n, the refractive index that minimizes the sum Σχ ² i,
A laminated structure inspection method characterized by inspecting a layer structure of the laminated body by an optimization method for determining a film thickness and an interface width. 2. An X-ray source for generating two or more types of characteristic X-rays used for measurement, a spectroscope, a sample and a drive for a detector,
In an X-ray reflectance device equipped with a detector and a control unit, which irradiates a sample installed on the drive table with X-rays and measures reflected X-rays from the sample, only a specific characteristic X-ray is reflected twice. An X including a spectroscopic element to be taken out, a drive unit for rotating and translating the spectroscopic element, and a control unit for controlling the drive unit by the control unit.
A linear reflectance device, wherein the laminated structure inspection according to claim 1.
An X-ray reflectance device comprising an analyzing means for performing an analysis based on the method .