JP2017078632A - Observation method of tomographic structure, observation device, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To observe a tomographic structure having a light absorption layer by an optical coherence tomography.SOLUTION: A tomographic layer structure to be observed is observed by an optical coherence tomography. A tomographic structure which is an observation object 30 includes a light absorption layer made from a light absorption material that absorbs light having at least a portion of wavelength of the light source spectrum. The tomographic layer structure is irradiated with measurement light from a light source. A reflection light profile of the tomographic layer structure is generated based on interference light between reflection light reflected from the tomographic layer structure by the measurement light and reference light. A comparison is made between the reflection light profile and a plurality of light propagation simulation results in a plurality of simulated tomographic layer structures each having a simulated light absorption layer with a different layer thickness. A layer thickness of a simulated light absorption layer which becomes an object of the simulation for obtaining the light propagation simulation result matching the reflection light profile at the maximum is determined as the layer thickness of the light absorption layer owned by the tomographic layer structure which is the observation object 30.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a fault structure observation method, observation apparatus, and computer program.

  Non-Patent Document 1 discloses that optical coherence tomography (OCT) using a visible light source is used to obtain layer thickness measurement and tomographic imaging of a transparent resist thin film applied to the surface of a semiconductor substrate. Yes. By using OCT, the transparent resist thin film can be observed in a non-destructive and non-contact manner.

  In OCT, the layer thickness is obtained as the optical path length. The optical path length corresponds to the actual layer thickness multiplied by the refractive index of the material of the layer. Therefore, in order to obtain the actual layer thickness of the resist thin film, the optical path length between both interfaces of the transparent resist thin film obtained by OCT may be divided by the refractive index of the resist thin film.

Takeshi Nishi, Nobuhiko Ozaki, Hirotaka Osato, Eiichiro Watanabe, Naoki Ikeda, Yoshimasa Sugimoto, “Application of μm-OCT to Semiconductor Microfabrication Process Using Visible Light Broadband Light Source”, The 62nd JSAP Spring Meeting, Japan Society of Applied Physics, March 2015

  The semiconductor substrate under the transparent resist thin film described in Non-Patent Document 1 is opaque, and it has been considered that OCT is not suitable for observing the tomographic structure inside the semiconductor substrate. However, the present inventors have found that OCT can observe not only the transparent resist thin film coated on the surface of the semiconductor substrate but also an opaque semiconductor layer under the transparent resist thin film, thereby completing the present invention. did.

  From one viewpoint, the present invention is an observation method for observing a fault structure by optical coherence tomography (OCT). The tomographic structure to be observed has a light absorption layer made of a light absorbing material that absorbs light of at least a part of the wavelength of the light source spectrum. The tomographic structure is irradiated with measurement light from a light source. The light source spectrum is preferably broadband. In the light absorbing material, at least a part of the wavelength of the irradiated light is absorbed, so that the spectrum of the light changes as the irradiated light travels through the light absorbing material. When the spectrum of light changes, the refractive index for the light also changes due to the dispersion characteristics of the light absorbing material. Further, since the amount of light absorption increases with the traveling distance of light in the substance, the change in refractive index also varies with the traveling distance of light. As a result, the thickness of the light absorbing layer made of a light absorbing material is difficult to obtain by dividing the optical path length by a single refractive index, and the optical path in the light absorbing layer considering the refractive index change by some method. It is necessary to derive the relationship between the length and the actual film thickness.

  Accordingly, the observation method according to the present invention includes a step of generating a reflected light profile of the tomographic structure based on interference light between the reflected light reflected from the tomographic structure and the reference light, and the reflected light profile includes: Steps for comparing multiple light propagation simulation results for multiple simulated fault structures with simulated light absorption layers with different layer thicknesses, and for simulation to obtain light propagation simulation results that best match the reflected light profile Determining the layer thickness of the simulated light absorption layer as the layer thickness of the light absorption layer of the tomographic structure to be observed, thereby reflecting the reflected light profile of the fault structure and the light propagation simulation in the simulated fault structure By comparing with the result, the thickness of the light absorption layer can be easily determined. The above observation method can be applied whether or not the measurement light is absorbed in the light absorption layer.

  Note that the measurement light preferably includes light having a wavelength that is not absorbed by the light-absorbing substance, but when the reflected light is obtained without being absorbed even at the wavelength that is absorbed because the light absorption layer is thin, The measurement light may be only light having a wavelength that is absorbed by the light absorbing material. For example, when the light absorbing material absorbs visible light, the measurement light may be visible light if the light absorbing layer is sufficiently thin.

  From another viewpoint, the present invention is an observation apparatus that observes a tomographic structure to be observed by optical coherence tomography. The observation apparatus includes: a light source that irradiates light to the tomographic structure; a detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal; and a processing unit that processes the detection signal. Prepare. The tomographic structure has a light absorbing layer made of a light absorbing material that absorbs light having a wavelength of at least a part of the light source spectrum.

  The processing unit generates a reflected light profile of the tomographic structure based on the detection signal, and the plurality of light in the simulated tomographic structure having the reflected light profile and simulated light absorbing layers each having a different layer thickness. Comparing the simulation result with the propagation simulation result, and determining the layer thickness of the simulated light absorbing layer that is the object of simulation for obtaining the optical propagation simulation result that best matches the reflected light profile, the fault structure that is the observation target And determining the layer thickness of the light absorption layer.

  From another viewpoint, the present invention is a computer program for causing a computer to function as the processing unit.

  From another viewpoint, the present invention irradiates the tomographic structure to be observed with measurement light that does not substantially include light having a wavelength that is absorbed by the light-absorbing substance. Even if the tomographic structure has a light absorption layer, the measurement light is not absorbed, so that a change in the light spectrum can be prevented. Therefore, a constant refractive index of the light absorption layer can be used for calculating the layer thickness of the light absorption layer. The light absorbing material here absorbs light of at least a part of wavelengths, for example, absorbs light of at least a part of wavelengths in the visible light spectrum. Measurement light that does not substantially contain light of a wavelength that is absorbed by the light absorbing material means that it does not include light of a wavelength that is absorbed by the light absorbing material at all, or substantially light of a wavelength that is absorbed by the light absorbing material. Means that it can only be ignored. For example, the light whose wavelength is absorbed by the light-absorbing substance is substantially negligible means, for example, that the intensity of the light absorbed by the light-absorbing substance is 10% or less of the peak intensity of the measurement light. . The constant refractive index means a constant refractive index that does not change depending on the distance through which light propagates. If absorption of measurement light occurs in the light absorption layer, the thickness of the light absorption layer cannot be calculated with a constant refractive index. However, if absorption of measurement light does not occur in the light absorption layer, the thickness depends on the material of the light absorption material. The layer thickness can be calculated using the refractive index of

  From another viewpoint, the present invention is a method for observing a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light of at least a part of the wavelength of a light source spectrum by optical coherence tomography. A step of irradiating the target tomographic structure with measurement light including light of a wavelength that is not absorbed by the light-absorbing substance from the light source, and light based on interference light between the reflected light reflected from the tomographic structure and the reference light. Obtaining a tomographic structure image of the absorption layer.

  From another viewpoint, the present invention is a method for observing a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light of at least a part of the wavelength of a light source spectrum by optical coherence tomography. Irradiating the target tomographic structure with measurement light from a light source; obtaining a tomographic structure image of the light absorption layer based on interference light between the reflected light reflected from the tomographic structure and the reference light; The step of acquiring the tomographic structure image of the light absorbing layer includes determining the layer thickness of the light absorbing layer based on the optical path length between both interfaces of the light absorbing layer, and calculating the film thickness of the light absorbing layer in the tomographic structure image. This is an observation method including making the obtained film thickness.

  From another viewpoint, the present invention is an observation apparatus for observing a tomographic structure to be observed by optical coherence tomography, comprising a light source that irradiates light to the tomographic structure, reflected light from the tomographic structure, and reference light. The tomographic structure includes a light absorption layer made of a light absorbing material that absorbs light of at least a part of the wavelength, and a tomographic structure includes a detection unit that detects interference light and outputs a detection signal; And the light source is configured to output measurement light including light of a wavelength that is not absorbed by the light absorbing material, and the processing unit acquires a tomographic structure image of the light absorption layer based on the detection signal. An observation device configured to execute processing including

  From another viewpoint, the present invention is an observation apparatus for observing a tomographic structure to be observed by optical coherence tomography, comprising a light source that irradiates light to the tomographic structure, reflected light from the tomographic structure, and reference light. The tomographic structure includes a light absorbing material that absorbs light having a wavelength of at least a part of the light source spectrum, and includes a detection unit that detects interference light and outputs a detection signal; An absorption layer, and the processing unit is configured to perform processing including acquiring a tomographic structure image of the light absorption layer based on the detection signal, and acquiring the tomographic structure image of the light absorption layer An observation apparatus comprising: obtaining a layer thickness of the light absorption layer based on an optical path length between both interfaces of the light absorption layer; and setting the film thickness of the light absorption layer in the tomographic structure image to the obtained film thickness It is.

  From another viewpoint, the present invention is a computer program for causing a computer to function as the processing unit.

It is a block diagram of an observation apparatus. It is explanatory drawing of the light which advances the tomographic structure of a to-be-observed object, and a tomographic structure. It is a flowchart which shows an observation procedure. It is a figure which shows the example of the observation pattern of a reflected light intensity distribution map. It is explanatory drawing of a simulation result table. It is a two-dimensional tomographic image to be observed. It is a light propagation simulation result of an observation object. It is a light propagation simulation result of an observation object. It is a light propagation simulation result of an observation object.

[1. Outline of Embodiment]

(1) The observation method according to the embodiment is a method of observing a tomographic structure to be observed by optical coherence tomography (OCT). The OCT may be, for example, any of Spectral Domain (SD) -OCT, Time Domain (TD) -OCT, and Swept Source (SS) -OCT.

  The tomographic structure to be observed is, for example, a tomographic structure of a semiconductor device. The semiconductor device to be observed may be a finished product or an incomplete product during the manufacturing process. The semiconductor device is, for example, a semiconductor device used as a light emitting device. The fault structure may be a single layer structure or a multilayer structure. The fault structure has a light absorption layer. There may be one or more light absorption layers included in the tomographic structure. The tomographic structure may have a layer that does not absorb light. The layer included in the tomographic structure (including the light absorption layer and / or the layer that does not absorb light) is preferably made of a material having a uniform material constant (dielectric constant / permeability) and a uniform refractive index. Those composed of a crystalline substance, an amorphous substance, or a polymer substance are preferable.

The light absorbing layer is made of a light absorbing material that absorbs light of at least a part of wavelengths. The light absorbing material is, for example, a semiconductor. The light here includes visible light as well as ultraviolet rays and infrared rays. When the light absorption layer is made of a light absorption material that absorbs light having at least a part of the wavelength of the light source spectrum, measurement light is absorbed in the light absorption layer. The light absorbing material may be, for example, a material that absorbs light having a wavelength of at least a part of the visible light spectrum. Further, the light absorbing material may be a material having band gap energy that absorbs light of at least a part of wavelengths of visible light, for example. Since semiconductors have a band gap, many semiconductors absorb at least some wavelengths of visible light.

  “Band gap energy” is an energy difference from a valence band to a conduction band in a crystal structure such as a semiconductor. “Bandgap energy that absorbs light” is bandgap energy equal to or lower than the energy of light. “Absorb light of at least some wavelengths of visible light” means to absorb all light in the wavelength range of visible light, or to absorb light of some wavelengths in the wavelength range of visible light It is.

  The energy of light corresponding to the wavelength range of visible light is about 1.7 [eV] to about 3 [eV]. Therefore, a substance having a band gap energy smaller than about 3 [eV] absorbs light having a wavelength of at least a part of visible light, and a substance having a band gap energy smaller than about 1.7 [eV] Absorbs all light in the wavelength range. Many semiconductors have a band gap energy less than about 3 [eV] and absorb at least a portion of visible light. For example, the band gap energy of gallium arsenide (GaAs) at room temperature is 1.42 [eV], and gallium arsenide (GaAs) at room temperature absorbs all light in the visible wavelength range and transmits visible light. do not do.

  The observation method according to the embodiment may include irradiating measurement light including light having a wavelength that is not absorbed by the light-absorbing substance to the tomographic structure to be observed. The measurement light may include light having a wavelength that is absorbed by the light absorbing material. If a substance that absorbs part of visible light is irradiated with broadband visible light, light having a part of visible light is absorbed, but light having the remaining wavelength is not absorbed. If a substance that absorbs all light in the visible light wavelength range but does not absorb infrared light is irradiated with visible light including infrared light, the visible light can be absorbed without being absorbed.

If the measurement light is broadband visible light, the optical axis resolution of OCT can be increased. When the optical axis direction resolution of OCT is Δz, it is preferable that the measurement light has a center wavelength λ ₀ and a bandwidth Δλ so that a high resolution in which Δz is smaller than 1 μm can be obtained. The optical axis direction resolution Δz of OCT is determined by the equation: Δz = 0.44λ ₀ ² / Δλ. In this equation, the bandwidth Δλ is the full width at half maximum (FWHM).

  Of the measurement light, light having a wavelength that is not absorbed by the light absorbing material is transmitted through the light absorption layer in the tomographic structure, reflected at the interface of the light absorption layer, and becomes reflected light. Based on the interference light between the reflected light and the reference light, a reflected light profile of the tomographic structure including the interface of the light absorption layer is generated. Of the measurement light, light having a wavelength that is absorbed by the light-absorbing substance is attenuated by light absorption. However, if the light absorption layer is thin, it is not completely absorbed and may become reflected light.

  The observation method according to the embodiment includes a step of comparing the reflected light profile with a plurality of light propagation simulation results in a plurality of simulated fault structures. Each of the plurality of simulated fault structures to be subjected to the light propagation simulation is a simulation of the tomographic structure to be observed, and each has a simulated light absorption layer having a different layer thickness.

  The simulated tomographic structure that is the object of simulation for obtaining the light propagation simulation result that best matches the reflected light profile most closely matches the tomographic structure to be observed. Therefore, the layer thickness of the simulated light absorption layer that is the object of simulation for obtaining the best matching simulation result can be determined as the layer thickness of the light absorption layer of the fault structure to be observed.

(2) Comparison between the reflected light profile and a plurality of light propagation simulation results in a plurality of simulated tomographic structures can be performed by, for example, pattern comparison. In this case, the reflected light profile includes, for example, an observation pattern indicating the reflected light intensity distribution in the depth direction of the tomographic structure to be observed, and the light propagation simulation result indicates, for example, the reflected light in the depth direction of the simulated tomographic structure. It is assumed that a prediction pattern indicating an intensity distribution is included.

  By comparing the observed pattern of the reflected light intensity distribution with the predicted pattern showing the reflected light intensity distribution predicted by the light propagation simulation using a method such as pattern matching, the best match with the observed pattern from the multiple predicted patterns The prediction pattern to be extracted can be extracted. The layer thickness of the simulated light absorption layer in the simulated fault structure that is the object of simulation for obtaining the predicted pattern that most closely matches the observation pattern is determined as the layer thickness of the light absorption layer of the fault structure that is the observation target.

(3) The comparison between the reflected light profile and a plurality of light propagation simulation results in the plurality of simulated tomographic structures may include a comparison of optical path lengths instead of or in addition to the pattern comparison. In this case, the reflected light profile includes the observed optical path length between both interfaces of the light absorption layer of the tomographic structure to be observed, and the light propagation simulation result includes the predicted optical path length between both interfaces of the simulated light absorption layer. Shall be.

  By comparing the observed optical path length with the predicted optical path length, the predicted optical path length that best matches the observed optical path length can be extracted from the plurality of predicted optical path lengths. The layer thickness of the simulated light absorption layer in the simulated fault structure that is the object of simulation to obtain the predicted optical path length that best matches the observed optical path length is determined as the layer thickness of the light absorption layer of the fault structure to be observed. The

(4) The observation apparatus according to the embodiment is for observing a tomographic structure to be observed by optical coherence tomography (OCT). The observation apparatus includes: a light source that irradiates light to the tomographic structure; a detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal; and a processing unit that processes the detection signal. Prepare.

  The tomographic structure to be observed has a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of the light source spectrum. The light source is preferably configured to output measurement light including light having a wavelength that is absorbed by the light absorbing material and light having a wavelength that is not absorbed by the light absorbing material. The light source includes, for example, a halogen lamp, an LED light source, or an SLD light source.

  The processing unit generates a reflected light profile of the tomographic structure based on the detection signal. The processing unit compares the reflected light profile with a plurality of light propagation simulation results in a plurality of simulated tomographic structures each having a simulated light absorption layer having a different layer thickness. The processing unit determines the thickness of the simulated light absorption layer of the simulated tomographic structure that is the object of simulation for obtaining the light propagation simulation result that best matches the reflected light profile, and the light absorption layer of the tomographic structure that is the observation target. The layer thickness is determined.

(5) The processing unit includes, for example, a computer including a CPU and a memory. The computer can exhibit a function as a processing unit by executing a computer program stored in the memory. A computer program for causing a computer to function as a processing unit can be stored in a computer-readable recording medium.

(6) An observation method according to another embodiment includes irradiating measurement light substantially free of light having a wavelength absorbed by the light-absorbing substance onto a tomographic structure to be observed. Based on the interference light between the reflected light reflected from the tomographic structure and the reference light, the optical path length of the light absorbing layer of the tomographic structure is obtained, and based on the optical path length and the refractive index of the light absorbing layer, The layer thickness of the absorption layer can be determined.

(7) A light source of an observation apparatus according to another embodiment is configured to output measurement light that substantially does not include light having a wavelength that is absorbed by the light absorbing material. The processing unit executes processing for obtaining the optical path length of the light absorption layer. The processing unit performs a process of determining the layer thickness of the light absorption layer based on the obtained optical path length and the refractive index of the light absorption layer. The light source may be constituted by a light emitting device that outputs only light that does not substantially include light having a wavelength that is absorbed by the light absorbing material, or outputs light that includes light having a wavelength that is absorbed by the light absorbing material. And a filter that cuts light having a wavelength that is absorbed by the light-absorbing substance out of the light output from the light-emitting device.

(8) The observation method in the embodiment is a method of observing a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light having at least a part of the light source spectrum by optical coherence tomography. The step of irradiating the tomographic structure with measurement light including light of a wavelength that is not absorbed by the light-absorbing substance from the light source, and light absorption based on the interference light between the reflected light reflected from the tomographic structure and the reference light And a step of acquiring a tomographic structure image of the layer. The tomographic structure image may be a two-dimensional image or a three-dimensional image.

(9) The observation method according to the embodiment is a method of observing, by optical coherence tomography, a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light of at least a part of the wavelength of the light source spectrum. Irradiating measurement light from a light source to a tomographic structure to be observed, and acquiring a tomographic structure image of the light absorption layer based on interference light between reflected light reflected from the tomographic structure and the reference light. The step of acquiring a tomographic structure image of the light absorbing layer includes obtaining a layer thickness of the light absorbing layer based on an optical path length between both interfaces of the light absorbing layer, and obtaining a film thickness of the light absorbing layer in the tomographic structure image May be an observation method including making the obtained film thickness. In this case, a tomographic structure image showing the film thickness more accurately is obtained.

(10) An observation apparatus according to an embodiment is an observation apparatus that observes a tomographic structure to be observed by optical coherence tomography, and includes a light source that irradiates light to the tomographic structure, reflected light from the tomographic structure, and reference light. And a processing unit for processing the detection signal, and the tomographic structure is made of a light-absorbing material that absorbs light of at least some wavelengths of the light source spectrum. The light source includes a light absorption layer, the light source is configured to output measurement light including light having a wavelength that is not absorbed by the light absorption material, and the processing unit acquires a tomographic structure image of the light absorption layer based on the detection signal. May be an observation device configured to execute processing including the above.

(11) An observation apparatus according to an embodiment is an observation apparatus that observes a tomographic structure to be observed by optical coherence tomography, and includes a light source that irradiates light to the tomographic structure, reflected light from the tomographic structure, and reference light. And a processing unit for processing the detection signal, and the tomographic structure is made of a light-absorbing material that absorbs light of at least some wavelengths of the light source spectrum. A light-absorbing layer, and the processing unit is configured to perform processing including acquiring a tomographic structure image of the light-absorbing layer based on the detection signal, and acquiring the tomographic structure image of the light-absorbing layer Is an observation apparatus that includes determining the thickness of the light absorption layer based on the optical path length between both interfaces of the light absorption layer, and setting the film thickness of the light absorption layer in the tomographic structure image to the determined film thickness. May be.

(12) As described above, the processing unit includes, for example, a computer including a CPU and a memory. The computer can exhibit a function as a processing unit by executing a computer program stored in the memory. A computer program for causing a computer to function as a processing unit can be stored in a computer-readable recording medium.

[2. Details of Embodiment]
Hereinafter, embodiments will be described in detail with reference to the drawings.

[2.1 First Embodiment]
FIG. 1 shows an observation apparatus 100 according to the first embodiment. The observation apparatus 100 is used, for example, to observe the tomographic structure of the semiconductor device that is the observation target 30 in a non-destructive / non-contact manner during or after the semiconductor device manufacturing process. The semiconductor device to be observed 30 is an optical device such as an optical integrated circuit, for example.

FIG. 2A shows a cross-sectional structure of the optical device semiconductor substrate as the observation target 30. This observation object 30 is an Al _0.35 Ga _0.65 AS / GaAs substrate coated with a photoresist. The Al _0.35 Ga _0.65 AS / GaAs substrate has an Al _0.35 Ga _0.65 AS layer 32 as a layer immediately below the photoresist layer 31, and further has a GaAs layer 33 as a layer therebelow. In the tomographic structure of the observation object 30, the layer thickness of the photoresist layer 31 is about 2 μm, and the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 is about 0.5 μm. In the observation apparatus 30, the layer thickness measurement of the photoresist layer 31 and the Al _0.35 Ga _0.65 AS layer 33 and the tomographic structure imaging of the observation target 30 are performed.

The observation apparatus 100 is configured as SD-OCT, and includes an optical interference measurement unit 10 and a processing unit 20 as shown in FIG. The optical interference measurement unit 10 includes a light source 11, lenses 12 and 13, a splitter 14, a reference mirror 15, a lens 16, and a detection unit 17. The light source 11 is composed of a halogen lamp. The halogen lamp can output visible light broadband. As the halogen lamp, a lamp that outputs visible light in a wide band with a center wavelength λ ₀ : 625 nm and a bandwidth Δλ (FWHM): 260 nm was used. Since the light source 11 outputs broadband visible light, the observation object 30 can be observed with high resolution.

  In the optical interference measurement unit 10, the light output from the light source 11 is divided into measurement light and reference light by the splitter 14. The measurement light is irradiated onto the observation target 30 and becomes reflected light from the observation target 30. The reference light is reflected by the reference mirror 15 and merges with the reflected light. Interference light is generated by the merge of the reflected light and the reference light. The interference light (interference signal) is detected by the detection unit 17. The detection unit 17 outputs an electrical signal corresponding to the interference light as a detection signal. The detection signal is given to the processing unit 20. The detection unit 17 of the embodiment is configured as a spectrometer, and outputs an interference light spectrum as a detection signal.

  The processing unit 20 includes a computer, and the computer exhibits the function as the processing unit 20 by executing a computer program for causing the computer to exhibit the function as the processing unit 20. The processing unit 20 performs layer thickness measurement and tomographic imaging processing of the observation target based on the detection signal.

  FIG. 3 shows an observation procedure using the observation apparatus 100. First, in step S1, the optical interference measurement unit 10 performs optical interference measurement with visible broadband light to obtain an interference light spectrum. In step S1, the light source 11 irradiates the tomographic structure of the semiconductor device to be observed with measurement light. The detection unit 17 outputs a detection signal (interference light spectrum) based on the interference light and supplies the detection signal to the processing unit 20.

  In step S <b> 2, the processing unit 20 converts the interference light spectrum indicated by the detection signal into a wave number region and performs a Fourier transform to generate a reflected light profile of the tomographic structure that is the observation target 30. The reflected light profile is, for example, a reflected light intensity profile in the optical axis depth direction. FIG. 4 shows a reflected light intensity profile of the observation target 30 in the optical axis depth direction. This profile is a reflected light intensity distribution pattern in which the horizontal axis indicates the depth and the vertical axis indicates the intensity of the reflected light. Hereinafter, the pattern of FIG. 4 is referred to as an “observation pattern”. Note that it is sufficient that the reflected light profile is an analysis result regarding the reflected light.

As shown in FIG. 2A, the reflected light from the observation object 30 is generated at the first interface 30a, the second interface 30b, and the third interface 30c of the observation object 30. That is, each interface 30a, 30b, 30c is a reflective surface. The first interface 30a is the surface of the photoresist layer 33, the second interface 30b is the interface between the photoresist layer 31 and the Al _0.35 Ga _0.65 AS layer 32, and the third interface 30c is , An interface between the Al _0.35 Ga _0.65 AS layer 32 and the GaAs layer 33. Hereinafter, the reflected light from the first interface 30a is referred to as first reflected light, the reflected light from the second interface 30b is referred to as second reflected light, and the reflected light from the third interface 30b is referred to as third reflected light. That's it.

  In the observation pattern shown in FIG. 4, a peak of reflected light intensity appears at the depth position corresponding to each interface 30a, 30b, 30c. The processing unit 20 can detect the positions of the interfaces 30a, 30b, and 30c by detecting the peak of the reflected light intensity in the observation pattern.

  In the observation pattern of FIG. 4, a first peak occurs at a depth of about 19 μm, which corresponds to the first reflected light from the first interface 30a. Similarly, a second peak occurs at a depth of about 23 μm, which corresponds to the second reflected light from the second interface 30b. Similarly, a third peak occurs at a position between the depths of 25 μm and 26 μm, which corresponds to the third reflected light from the third interface 30c.

  The processing unit 20 obtains the interval between the first peak and the second peak based on the observation pattern of FIG. 4 in order to obtain the layer thickness of the photoresist layer 31. In the observation pattern of FIG. 4, the distance between the first peak and the second peak is 3.77 μm. The distance between the first peak and the second peak obtained from the observation pattern in FIG. 4 is the optical path length of the photoresist layer 31 (the optical path length from the first interface 30a to the second interface 30b). It is necessary to convert to the actual layer thickness of the resist layer 31.

Since the photoresist layer 31 is transparent and does not absorb light, the refractive index of the photoresist layer 31 may be used to obtain the actual layer thickness from the optical path length. Specifically, the processing unit 20 divides the optical path length of 3.77 μm of the photoresist layer 31 by the refractive index of 1.62 of the photoresist layer 31, and the layer thickness of the photoresist layer 31 is about 2.33 μm. And decide. The refractive index 1.62 of the photoresist layer 31 is a value in the case of light having a center wavelength λ ₀ = 625 nm.

The processing unit 20 obtains the optical path length of the Al _0.35 Ga _0.65 AS layer 32 (the optical path length from the second interface 30b to the third interface 30c) from the distance between the second peak and the third peak. it can. In the observation pattern of FIG. 4, the optical path length of the Al _0.35 Ga _0.65 AS layer 32 is 2.54 μm. However, since the Al _0.35 Ga _0.65 AS layer 32 is a light absorbing layer that absorbs most of the visible light, the Al _0.35 Ga _0.65 AS layer can be used simply by using the refractive index. An accurate layer thickness of 32 cannot be obtained.

Here, the band gap energy of Al _0.35 Ga _0.65 AS is 1.86 [eV], and the wavelength (optical absorption edge) corresponding to this band gap energy = 1.86 [eV] is 667 nm. is there. Al _0.35 Ga _0.65 AS absorbs light having a wavelength of 667 nm or less and transmits more light without absorbing it. Thus, Al _0.35 Ga _0.65 AS is an opaque material that absorbs most light in the visible light wavelength range, and the Al _0.35 Ga _0.65 AS layer 32 is a light absorption layer. is there.

Al _0.35 Ga _0.65 AS absorbs most of the visible light, but the measurement light output from the light source 11 according to the embodiment is broadband light, and as shown in FIG. In addition, light having a wavelength larger than 667 nm which is an optical absorption edge of Al _0.35 Ga _0.65 AS is also included. Measurement light having a wavelength greater than 667 nm is not absorbed by the Al _0.35 Ga _0.65 AS layer 32, but is reflected by the third interface 30 c to become third reflected light.

On the other hand, measurement light having a wavelength smaller than 667 nm is absorbed by the Al _0.35 Ga _0.65 AS layer 32. Therefore, as shown in FIG. 2 (c), the spectrum of the measurement light traveling in the Al _0.35 Ga _0.65 AS layer 32 has a light intensity attenuated in a wavelength region smaller than 667 nm. Since the absorption of the measurement light gradually occurs while passing through the Al _0.35 Ga _0.65 AS layer 32, the attenuation of the light intensity as shown in FIG. 2C increases with the progress of the light, The spectrum of the measurement light (particularly, the peak wavelength of the light) changes during passage through the Al _0.35 Ga _0.65 AS layer 32. As a result, it is difficult to obtain the refractive index of the Al _0.35 Ga _0.65 AS layer 32 in a state where the layer thickness is unknown.

The optical path length T from the position _{x 1} to the position _{x 2} of the third interface 30c of the second interface 30b can be expressed by the following equation.
Here, n g _(x) is the group refractive index at any position from position x ₁ to the position x _2, x is the position of the optical axis depth direction. It is also clear from the above formula that it is difficult to obtain the refractive index of the Al _0.35 Ga _0.65 AS layer 32 in a state where the layer thickness is unknown.

As described above, the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 cannot be obtained in the same manner as the layer thickness of the photoresist layer 31. Therefore, the processing unit 20 of the present embodiment compares the reflected light profile with the light propagation simulation result in step S3 in order to obtain the layer thickness of the Al _0.35 Ga _0.65 AS layer 32.

The light propagation simulation predicts light propagation in a simulated fault structure that simulates the fault structure of the observation target 30. In this light propagation simulation, the simulated tomographic structure is irradiated with the same light as the measurement light output from the light source 11 (center wavelength λ ₀ : 625 nm, bandwidth Δλ (FWHM): 260 nm). The simulated fault structure includes a layer simulating the photoresist layer 31, a layer simulating the Al _0.35 Ga _0.65 AS layer 32, and a layer simulating the GaAs layer 33, similarly to the fault structure of the observation object 30. Have. A layer simulating the Al _0.35 Ga _0.65 AS layer 32 which is a light absorption layer is a simulated light absorption layer.

In the light propagation simulation, the layer thickness of each simulated layer is set to a predetermined value. For example, the layer thickness of the layer simulating the photoresist layer 31 is set to 2 μm, and the layer thickness of the layer simulating the Al _0.35 Ga _0.65 AS layer 32 is set to 500 nm. The light propagation simulation is performed a plurality of times by changing the layer thickness of the simulated light absorption layer simulating the Al _0.35 Ga _0.65 AS layer 32. That is, there are a plurality of simulated tomographic structures to be subjected to the light propagation simulation, and each of the plurality of simulated tomographic structures has simulated light absorption layers having different layer thicknesses. In the light propagation simulation for a plurality of times, the layer thickness of the simulated light absorption layer is changed, for example, by 10 nm.

  The light propagation simulation is performed, for example, by electromagnetic field analysis using a time domain difference method (FDTD method). By the light propagation simulation for each of the plurality of simulated fault structures, a plurality of light propagation simulation results for each of the plurality of simulated fault structures are obtained.

  By the light propagation simulation, a prediction pattern indicating the reflected light intensity distribution in the depth direction of the simulated fault structure can be obtained as a light propagation simulation result. The prediction pattern is a prediction of a reflected light intensity distribution (observation pattern) obtained when a tomographic structure having a light absorption layer with a known layer thickness is observed by OCT. Since the reflected light intensity from each interface 30a, 30b, 30c is determined by the material and layer thickness of each layer 31, 32, 33, the reflected light intensity from each interface 30a, 30b, 20c is different for each interface 30a, 30b, 30c. Different. For this reason, the reflected light intensity distribution has a unique pattern according to the layer thickness of each layer. Therefore, if a predicted pattern of the reflected light intensity distribution in the depth direction of the tomographic structure is obtained for a tomographic structure having a light absorption layer with a known layer thickness, a predicted pattern of the reflected light intensity distribution in the depth direction of the tomographic structure is obtained. It can be predicted how much intensity the reflected light is generated at which depth position of the reflected light intensity distribution. Therefore, by comparing the actually observed reflected light intensity distribution (observation pattern) with the predicted pattern, it becomes easy to identify the intensity peak due to the reflected light from each interface (particularly the third interface 30c).

The processing unit 20 stores a simulation result table 21 illustrated in FIG. 5 in a memory included in the processing unit 20. In the table 21 of FIG. 5, a plurality of prediction patterns as a plurality of light propagation simulation results and an Al _0.35 Ga _0.65 AS layer in the simulated fault structure that is the target of the light propagation simulation for obtaining each prediction pattern. The layer thickness of the (simulated light absorption layer) is associated.

  The processing unit 20 compares each of the plurality of prediction patterns stored in the table 21 with the observation pattern, for example, by pattern matching, and extracts a prediction pattern that most closely matches the observation pattern from the plurality of prediction patterns. (Step S3 in FIG. 3). The prediction pattern stored in the table 21 only needs to have information necessary for comparison with the observation pattern, and may be image data of the reflected light intensity distribution, or the feature amount of the reflected light intensity distribution. It may be information indicating.

The processing unit 20 uses the table 21 to determine the layer thickness of the Al _0.35 Ga _0.65 AS layer (simulated light absorption layer) in the simulated fault structure that is the object of simulation for obtaining a predicted pattern that most closely matches the observation pattern. Get by referring to. The processing unit 20 determines the layer thickness obtained by referring to the table 21 as the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 included in the tomographic structure that is the observation target 30 (step S4 in FIG. 3). .

The processing unit 20 outputs the layer thicknesses of the Al _0.35 Ga _0.65 AS layer 32 and the photoresist layer 31 determined as described above. The layer thickness may be output as a numerical value, or a two-dimensional or three-dimensional tomographic structure image having the determined layer thickness may be output. FIG. 6 shows a two-dimensional tomographic structure image of the observation target 30 obtained by scanning the measurement light in the plane direction of the observation target 30. In this two-dimensional tomographic image, the three interfaces 30a, 30b, and 30c are imaged with three white lines, and the space between the white lines indicates the layer thickness of each layer. The two-dimensional image of the tomographic structure in FIG. 6 is different from the actual layer thickness because the depth direction is indicated by the optical path length, but the tomographic structure image is determined using the layer thickness determined as described above. By correcting the thickness of each layer (the position in the depth direction of the white line indicating the interfaces 30a, 30b, and 30c), a tomographic structure showing the actual layer thickness can be imaged. Even with a tomographic structure image whose depth direction is indicated by the optical path length, qualitative tomographic structure verification such as confirmation of the uniformity of each layer can be performed, but if the tomographic image shows the actual layer thickness, a quantitative tomographic structure Verification is also possible.

[2.2 Modification of First Embodiment]
The simulation result table 21 in FIG. 5 has the predicted pattern of the reflected light intensity distribution as the light propagation simulation result, but instead of or in addition to the predicted pattern, the predicted value of the optical path length of the simulated light absorbing layer (predicted) (Optical path length) may be included as a light propagation simulation result. In this case, in the table 21, the predicted optical path length as a plurality of light propagation simulation results and the layer thickness of the simulated light absorption layer that is the target of the light propagation simulation for obtaining the predicted optical path length are associated with each other. It will be. Note that the optical path length of the simulated light absorption layer can be predicted by calculation according to the above-described formula of the optical path length T.

Each of the processing units 20 includes a plurality of predicted optical path lengths stored in the table 21 and an optical path length of the light absorption layer (Al _0.35 Ga _0.65 AS layer 32) obtained from the observation pattern (observed optical path length = 2. And a predicted optical path length that best matches the observed optical path length is extracted from a plurality of predicted optical path lengths.

The processing unit 20 refers to the table 21 for the layer thickness of the Al _0.35 Ga _0.65 AS layer (simulated light absorption layer) subjected to simulation for obtaining the predicted optical path length that most closely matches the observed optical path length. To get it. The processing unit 20 determines the layer thickness obtained by referring to the table 21 as the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 included in the tomographic structure that is the observation target 30.

[2.3 Consideration of multiple reflected light in photoresist layer]
In the observation pattern of FIG. 4, a peak (fourth peak) is recognized at a position 0.69 μm away from the third peak toward the longer wavelength side. The fourth peak is a peak due to double reflection of the second reflected light in the photoresist layer 31 as shown in FIG. The distance from the second peak to the fourth peak is 3.23 μm, which is almost the same as the distance from the first peak to the second peak of 3.77 μm. Therefore, the fourth peak is a double reflection of the second reflected light. It turns out that it is a peak by.

  The peak due to the double reflection of the second reflected light in the photoresist layer 31 is observed because the double reflected light in the photoresist layer is generated in the light propagation simulation results shown in FIGS. It is clear from

7 to 9 show simulation results for the simulated fault structure. As described above, the simulated fault structure is a layer that simulates the photoresist layer 31 and a layer that simulates the Al _0.35 Ga _0.65 AS layer 32. And a layer simulating the GaAs layer 33. FIG. 7 shows the case where the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 is 500 nm, and FIG. 8 shows the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 is 1000 nm. 9 shows a simulation result when the layer thickness of the layer simulating the Al _0.35 Ga _0.65 AS layer 32 is 1500 nm. In any simulated fault structure, the layer thickness of the layer simulating the photoresist layer 31 was unified at 2 μm.

7 to 9, the horizontal axis represents the light propagation length, and the vertical axis represents the optical signal intensity. 7 to 9 show not the reflected light from the interface but the optical signal intensity (the intensity of the attenuated light during propagation) at each position of the optical propagation path. As shown in FIGS. 7 to 9, the distance between the light at the second interface and the light at the third interface changes according to the layer thickness difference of the layer simulating the Al _0.35 Ga _0.65 AS layer 32. On the other hand, the double reflected light in the photoresist layer always occurs at the same position.

  As shown in FIG. 9, when the propagation length of the light at the third interface and the propagation length of the double reflected light in the photoresist layer are greatly different, the reflected light is observed from the third interface 30c. The three reflected light and the double reflected light in the photoresist layer are observed at different depth positions. On the other hand, as shown in FIG. 8, when the propagation length of the light at the third interface and the propagation length of the double reflected light in the photoresist layer are substantially the same, or as shown in FIG. When the difference between the propagation length of light and the propagation length of double reflected light in the photoresist layer is small, as shown in FIG. 4, when the reflected light is observed, the second length from the third interface 30c is The three reflected light (third peak) and double reflected light in the photoresist layer (peak due to double reflection of the second reflected light) are observed overlapping at different depth positions. In this case, it may be difficult to accurately obtain the depth position of the third peak due to the influence of the peak due to double reflection of the second reflected light.

If it is difficult to obtain the depth position of the third peak with high accuracy, the observation optical path length from the second peak to the third peak cannot be obtained with high accuracy, and "2.2 Modification of First Embodiment" In the case of the method of _{obtaining the} observation optical path length from the observation pattern as in the case of "", the accuracy of the layer thickness of the Al _0.35 Ga _0.65 AS layer 32 is lowered.

  On the other hand, as shown in FIG. 5, in the case of the method of comparing the observed pattern with the predicted pattern, the predicted pattern is generated by a light propagation simulation considering the multiple reflected light. By comparing an observation pattern including a predicted pattern including multiple reflected light, the film thickness can be accurately determined even when there is multiple reflected light, without obtaining the optical path length. When the influence of the multiple reflected light is small, the film thickness can be determined with high accuracy even from the optical path length as in “2.2 Modification of First Embodiment”. In the observation pattern, not only the peak due to reflected light from each interface but also the peak due to multiple reflected light and the peak due to noise as described above may occur. Thus, it may be necessary to determine the peak due to the reflected light from each interface. In this regard, in the prediction pattern, since the depth position and intensity of the reflected light from each interface 30a, 30b, 30c (particularly the third interface 30c) are obtained, the reflected light from each interface in the observation pattern is obtained. The peak closest to the predicted position (predicted depth position and predicted intensity) of the peak can be easily determined as the peak due to the reflected light from each interface.

[2.4 Second Embodiment]
The observation apparatus 100 according to the second embodiment is the same as the observation apparatus 100 of the first embodiment except for the processing contents of the light source 11 and the processing unit 20. The measurement light output from the light source 11 does not substantially include light having a wavelength absorbed by the Al _0.35 Ga _0.65 AS layer 32. Since the optical absorption edge of Al _0.35 Ga _0.65 AS is 667 nm, the measurement light does not substantially include light having a wavelength of 667 nm or less, and is light having a wavelength larger than 667 nm. The light source 11 outputs, for example, infrared rays, and more specifically, the center wavelength λ ₀ = 900 nm and the bandwidth Δλ = 500 nm. In this case, the irradiation light output from the light source 11 includes light having a wavelength smaller than 667 nm, but is very small. Even if the light is absorbed by the Al _0.35 Ga _0.65 AS layer 32, the irradiation light is emitted. The spectrum of light hardly changes.

Therefore, the processing unit 20 _obtains the optical path length of the Al _0.35 Ga _0.65 AS layer 32 from the observation pattern, and calculates the obtained optical path length by the refractive index in the Al _0.35 Ga _0.65 AS layer 32 of the measurement light. By dividing, the thickness of the Al _0.35 Ga _0.65 AS layer 32 can be determined.

Even when the light source 11 that outputs the measurement light that does not substantially include the light having the wavelength absorbed by the Al _0.35 Ga _0.65 AS layer 32 is used, the film thickness is compared with the observation pattern and the prediction pattern. May be determined.

[3. Addendum]
The present invention is not limited to the above-described embodiments, and various modifications are possible.

DESCRIPTION OF SYMBOLS 10 Optical interference measurement part 11 Light source 17 Detection part 20 Processing part 30 Observation object 100 Observation apparatus

Claims (12)

A method for observing a fault structure to be observed by optical coherence tomography,
Irradiating measurement light from a light source to the tomographic structure having a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum;

Generating a reflected light profile of the tomographic structure based on interference light between reflected light and reference light reflected from the tomographic structure by the measurement light;
Comparing the reflected light profile with a plurality of light propagation simulation results in a plurality of simulated tomographic structures each having a simulated light absorbing layer of a different layer thickness;
The layer thickness of the simulated light absorption layer that is the object of simulation for obtaining the light propagation simulation result that best matches the reflected light profile is defined as the layer thickness of the light absorption layer of the tomographic structure that is the observation target. A step to determine;
Observation method including The reflected light profile includes an observation pattern indicating a reflected light intensity distribution in the depth direction of the tomographic structure to be observed;
The light propagation simulation result includes a prediction pattern indicating a reflected light intensity distribution in the depth direction of the simulated fault structure,
The observation method according to claim 1, wherein the comparing includes comparing the observation pattern with the prediction pattern. The reflected light profile includes an observation optical path length between both interfaces of the light absorption layer of the tomographic structure to be observed;
The light propagation simulation result includes a predicted optical path length between both interfaces of the simulated light absorption layer,
The observation method according to claim 1, wherein the comparing includes comparing the observed optical path length with the predicted optical path length. An observation device for observing a tomographic structure to be observed by optical coherence tomography,
A light source for irradiating the tomographic structure with light;
A detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal;
A processing unit for processing the detection signal;
With
The tomographic structure has a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum;

The processor is
Generating a reflected light profile of the tomographic structure based on the detection signal;
Comparing the reflected light profile with a plurality of light propagation simulation results in a plurality of simulated tomographic structures each having a simulated light absorbing layer of a different layer thickness;
The layer thickness of the simulated light absorption layer that is the object of simulation for obtaining the light propagation simulation result that best matches the reflected light profile is defined as the layer thickness of the light absorption layer of the tomographic structure that is the observation target. To decide,
An observation device configured to execute a process including:   The computer program for functioning a computer as a process part of Claim 4. A method of observing, by optical coherence tomography, a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light of at least some wavelengths,
Irradiating the tomographic structure to be observed with measurement light substantially free of light having a wavelength absorbed by the light absorbing material;
Obtaining an optical path length of the light absorption layer of the tomographic structure based on interference light between the reflected light and the reference light reflected from the tomographic structure of the measurement light;
Determining a layer thickness of the light absorbing layer based on the optical path length and a refractive index of the light absorbing layer;
Observation method including An observation device for observing a tomographic structure to be observed by optical coherence tomography,
A light source for irradiating the tomographic structure with light;
A detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal;
A processing unit for processing the detection signal;
With
The tomographic structure has a light absorption layer made of a light absorbing material that absorbs light of at least some wavelengths,
The light source is configured to output measurement light substantially free of light having a wavelength absorbed by the light absorbing material;
The processor is
Obtaining an optical path length of the light absorption layer of the tomographic structure based on the detection signal;
Determining the layer thickness of the light absorption layer based on the optical path length and the refractive index of the light absorption layer;
An observation device configured to execute a process including: A method for observing, by optical coherence tomography, a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum,
Irradiating the tomographic structure to be observed from a light source with measurement light including light of a wavelength that is not absorbed by the light absorbing material;
Obtaining a tomographic structure image of the light absorption layer based on interference light between reflected light and reference light reflected from the tomographic structure by the measurement light;
Observation method including A method for observing, by optical coherence tomography, a tomographic structure having a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum,
Irradiating measurement light from a light source to the tomographic structure to be observed;
Obtaining a tomographic structure image of the light absorption layer based on interference light between reflected light and reference light reflected from the tomographic structure by the measurement light;
Including
The step of obtaining a tomographic structure image of the light absorbing layer is to obtain a layer thickness of the light absorbing layer based on an optical path length between both interfaces of the light absorbing layer, and to obtain a film thickness of the light absorbing layer in the tomographic structure image An observation method including making the obtained film thickness. An observation device for observing a tomographic structure to be observed by optical coherence tomography,
A light source for irradiating the tomographic structure with light;
A detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal;
A processing unit for processing the detection signal;
With
The tomographic structure has a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum;
The light source is configured to output measurement light including light of a wavelength that is not absorbed by the light absorbing material;
The processing unit is configured to execute a process including acquiring a tomographic structure image of the light absorption layer based on the detection signal. An observation device for observing a tomographic structure to be observed by optical coherence tomography,
A light source for irradiating the tomographic structure with light;
A detection unit that detects interference light between the reflected light from the tomographic structure and the reference light and outputs a detection signal;
A processing unit for processing the detection signal;
With
The tomographic structure has a light absorption layer made of a light absorbing material that absorbs light having a wavelength of at least a part of a light source spectrum;
The processing unit is configured to perform processing including acquiring a tomographic structure image of the light absorption layer based on the detection signal;
Acquiring a tomographic structure image of the light absorbing layer is to obtain a layer thickness of the light absorbing layer based on an optical path length between both interfaces of the light absorbing layer, and to obtain a film thickness of the light absorbing layer in the tomographic structure image An observation device that includes making the desired film thickness.   The computer program for functioning a computer as a process part of Claim 10 or Claim 11.