Silicon carbide wafer surface defect detection method based on white light interference microscopy

Abstract

The invention discloses a silicon carbide wafer defect detection method based on an LDLS white light interference technology, and the method comprises the steps: an unpolarized beam splitter prism is used for receiving white light emitted by an LDLS light source, and dividing the white light into detection light and reference light; the probe light is emitted to the surface of the silicon carbide wafer to be measured for measurement; the reference light is emitted to the reflector; the microscope objective is used for amplifying the surface of the tested sample and the surface of the reflector; the non-polarization splitting prism combines the detection light and the reference light to generate interference and transmits the light to the focusing lens; the focusing lens focuses the light to a photosensitive element of the CCD camera so as to form a clear image; the CCD camera detects and records an interference fringe image; the piezoelectric ceramic displacement device controls the reflecting mirror to generate a series of nanoscale micro displacement; and processing the detected interference data through a data processing process of surface reconstruction to complete morphology reconstruction. The method accurately measures the surface defects of the silicon carbide wafer through a white light interference microtechnique, and has the characteristics of high measurement precision, non-contact measurement and the like.

Description

Technical Field

[0001] The present invention relates to the field of surface defect detection, and in particular to a silicon carbide wafer defect detection method based on LDLS (laser driven white light source) white light interference technology. Background Art

[0002] Silicon carbide wafers have become an important material for manufacturing high-performance power devices and optoelectronic devices due to their excellent physical properties. However, since silicon carbide grows in a high-temperature environment and has high rigidity and chemical stability, it is easy to cause high-density defects on the wafer and surface. These surface defects will directly affect the performance and reliability of the device, and may cause current leakage, breakdown voltage drop, reduce the transmission efficiency and output power of the device, and even cause local overheating. Therefore, it is very important to accurately detect surface defects of silicon carbide wafers.

[0003] At present, the technology for detecting surface defects of silicon carbide wafers has made significant progress, but it still faces challenges. The special properties of silicon carbide wafers, such as high hardness, high chemical stability and high temperature environment during the growth process, require precise technology and algorithms for surface defect detection. Existing detection methods include optical microscopy, scanning electron microscopy (SEM), atomic force microscopy (AFM), etc., which can observe and identify defects of different scales and types on the surface of the wafer. However, these technologies still have certain limitations in measurement accuracy, detection speed and resolution. Therefore, it is necessary to develop a high-precision silicon carbide wafer surface defect detection method to improve the quality and reliability of the device manufacturing process. Summary of the invention

[0004] The present invention provides a method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology. The present invention makes up for the shortcomings of low measurement accuracy and low resolution of existing detection technologies, as described below:

[0005] A method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology, the method comprising:

[0006] A silicon carbide wafer surface defect detection device based on LDLS white light interferometry is built using a laser-driven white light source, a non-polarized beam splitter, a reflector, a silicon carbide wafer, a microscope objective, a piezoelectric ceramic shifter, a piezoelectric ceramic controller, a focusing lens, and a CCD camera.

[0007] With the help of piezoelectric ceramic displacer and CCD camera, the original data collection is realized;

[0008] Perform surface morphology reconstruction on the collected raw data to obtain position information of zero optical path difference;

[0009] Fitting the position information of zero optical path difference to obtain the profile information of the measured silicon carbide surface;

[0010] Based on the acquired contour information, the irregularity and height variation of the contour are analyzed, defects on the surface of the silicon carbide wafer are identified and located, and defect detection is achieved.

[0011] Among them, the non-polarizing beam splitter prism is used to split the light from the LDLS light source into two beams, one beam is used as the reference light, and the other beam is used to illuminate the sample to be tested; the reflector and the silicon carbide wafer act as the reference surface and the surface to be tested, respectively, for reflecting light; the microscope objective is used to focus and amplify the light reflected by the reflector and the surface of the silicon carbide wafer to ensure that the interference fringes are projected onto the detector.

[0012] The piezoelectric ceramic displacer is used to adjust the micro-displacement of the reflector to generate an optical path difference to form interference fringes; the piezoelectric ceramic controller is used to control the movement of the piezoelectric ceramic displacer to ensure the adjustment and stability of the position of the reflector.

[0013] The focusing lens is used to focus the light beam after interference; the CCD camera is used as a detector to obtain the image of the interference fringes to provide raw data for subsequent surface morphology analysis.

[0014] The collection of the original data is as follows:

[0015] The LDLS white light source emits white light; the non-polarized beam splitter prism receives the white light emitted by the LDLS light source, and divides the white light into a detection light and a reference light, and the detection light is emitted to the surface of the silicon carbide wafer to be measured through the microscope objective lens, and the reference light is emitted to the reflector through the microscope objective lens;

[0016] The reflector is moved on the piezoelectric ceramic displacer to adjust the optical path of the reference light and the optical path difference between the detection light and the reference light, so that the non-polarized beam splitter prism receives the reflected light from the surface of the silicon carbide wafer and the reflector for beam combining interference and transmits it to the focusing lens to focus on the photosensitive detection surface of the CCD camera, and the interference image is recorded by the CCD camera.

[0017] The method includes: obtaining defect morphology information on the surface of the silicon carbide wafer to be tested by reconstructing the surface morphology of the monitored interference image data.

[0018] The beneficial effects of the technical solution provided by the present invention are:

[0019] 1. The present invention adopts LDLS light source to provide high-brightness and high-stability white light for the shape measurement process, thereby providing high-quality interference signals and enhancing the accuracy and reliability of the measurement;

[0020] 2. The present invention can provide high-resolution silicon carbide wafer surface topography images and accurately detect tiny surface defects, such as scratches, cracks, particles, etc.;

[0021] 3. The present invention adopts a non-contact detection method, which does not require contact with the sample surface, avoids damage or contamination to the wafer, and ensures the integrity and purity of the surface;

[0022] 4. The present invention realizes rapid detection of silicon carbide wafer surface defects, improves production efficiency and detection speed; the present invention can also realize multifunctional detection mode. In addition to detecting surface defects, white light interferometry technology can also provide information such as surface height and morphology, providing more references for comprehensive analysis and evaluation of the quality of silicon carbide wafers;

[0023] 5. The present invention is based on white light interferometry technology, which can accurately identify and quantify surface defects of different types and sizes and provide accurate detection results.

[0024] In summary, the present invention has the advantages of high efficiency, accuracy, and non-destructiveness, and provides an important quality control method for the device manufacturing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the optical path for silicon carbide wafer defect detection based on LDLS white light interferometry technology;

[0026] Figure 2 It is a flow chart of a method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology;

[0027] Figure 3 Flowchart of data processing for surface topography reconstruction;

[0028] Figure 4 Interference signal and its envelope diagram generated for simulation;

[0029] Figure 5 Schematic diagram of surface defect detection results. DETAILED DESCRIPTION

[0030] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention are described in further detail below.

[0031] Example 1

[0032] A silicon carbide wafer defect detection method based on LDLS white light interferometry technology comprises the following steps: step 1, using a laser-driven white light source, a non-polarized beam splitter prism, a reflector, a silicon carbide wafer, a microscope objective, a piezoelectric ceramic shifter, a piezoelectric ceramic controller, a focusing lens, and a CCD camera to build a silicon carbide wafer surface defect detection device based on LDLS white light interferometry; step 2, using the piezoelectric ceramic shifter and the CCD camera to realize the collection of raw data; step 3, performing surface morphology reconstruction processing on the collected raw data to obtain position information of zero optical path difference; step 4, fitting the position information of zero optical path difference to obtain contour information of the measured silicon carbide surface; step 5, based on the acquired contour information, analyzing the irregularity and height change of the contour, identifying and locating the defects on the surface of the silicon carbide wafer, and realizing defect detection.

[0033] Further, in step 1, the laser driven white light source (LDLS) has a spectral range of 170nm-2100nm and a power of 0.5W, and is used to generate continuous white light to illuminate the sample to be tested. The non-polarizing beam splitter, reflector and focusing lens have a use band of 400nm-700nm, and the spectral range of the LDLS light source covers the above band.

[0034] The non-polarizing beam splitter is used to split the light from the LDLS light source into two beams, one as reference light and the other for irradiating the sample to be tested. The reflector and silicon carbide wafer serve as the reference surface and the surface to be tested, respectively, for reflecting light.

[0035] Among them, the microscope objective is used to focus and amplify the light reflected by the reflector and the surface of the silicon carbide wafer, ensuring that the interference fringes are clearly projected onto the detector and improving the detection resolution.

[0036] Among them, the piezoelectric ceramic displacer is used to adjust the micro-displacement of the reflector to produce an optical path difference to form interference fringes; the piezoelectric ceramic controller is used to control the movement of the piezoelectric ceramic displacer to ensure accurate and stable adjustment of the reflector position.

[0037] Among them, the focusing lens is used to focus the interfered light beam, reduce energy loss, and ensure the clarity of the interference fringes; the CCD camera, as a detector, obtains the image of the interference fringes and provides raw data for subsequent surface morphology analysis.

[0038] In step 2, the piezoelectric ceramic displacer controls the reflector to produce a slight displacement, and the optical path difference between the detection light and the reference light is within the coherence length, and the CCD camera can detect the interference fringes.

[0039] Furthermore, the raw data collection process in step 2 includes the following steps:

[0040] White light is emitted by the LDLS white light source;

[0041] The white light emitted by the LDLS light source is received through a non-polarized beam splitter prism, and the white light is divided into a detection light and a reference light. The detection light is emitted to the surface of the silicon carbide wafer to be tested through a microscope objective lens, and the reference light is emitted to a reflector through a microscope objective lens;

[0042] The reflector is moved on the piezoelectric ceramic displacer to adjust the optical path of the reference light and the optical path difference between the detection light and the reference light, so that the non-polarized beam splitter prism receives the reflected light from the surface of the silicon carbide wafer and the reflector for beam combining interference and transmits it to the focusing lens to focus on the photosensitive detection surface of the CCD camera. The interference signal is monitored by the CCD camera and the interference image is recorded.

[0043] Furthermore, in step 3 and step 4, the defect morphology information of the surface of the silicon carbide wafer to be tested can be obtained by reconstructing the surface morphology of the monitored interference image data.

[0044] Example 2

[0045] like Figure 1 The figure shows the principle structure diagram of measuring defects of silicon carbide wafers based on LDLS white light interferometry. The LDLS light source emits white light, which is divided into two beams by a non-polarized beam splitter prism. One beam is used as a reference arm and hits the reflector through a microscope objective lens; the other beam is used as a measuring arm and hits the surface of the silicon carbide wafer to be measured through a microscope objective lens. The light reflected by the reflector and the sample to be measured converges to the non-polarized beam splitter prism through the microscope objective lens to interfere, and is received by the CCD camera through the focusing lens. By adjusting the optical path of the reference arm and the measuring arm, the position of zero optical path difference is found to generate interference fringes. The scanning step length is set to s and the scanning range is set to d through the piezoelectric ceramic controller. The piezoelectric ceramic shifter controls the reflector to move in a single direction with a step size of s, generates an optical path difference through a micro-displacement, and generates a series of interference fringes. At the same time, the CCD camera detects and records the interference fringe pattern generated by each step displacement as the initial data of the white light interference signal in the subsequent image processing part.

[0046] After obtaining the white light interference signal, the signal needs to be processed using the surface topography reconstruction algorithm to calculate the height value of each point, thereby completing the reconstruction of the entire surface and realizing the detection of surface defects of silicon carbide wafers. The specific solution process of the surface topography reconstruction method is as follows:

[0047] (1) Data preprocessing

[0048] First, each pixel of the detected interference image is read in order to process the interference signal.

[0049] (2) Fourier transform

[0050] The fringe intensity distribution produced by white light interference is spatially incoherent and can be expressed as:

[0051] I(x,y,z)=I0(x,y)+I s (x,y,z) (1)

[0052] Where x and y are the coordinate values ​​of the pixel, and z is the scanning height of the measured surface. I0 is the background intensity, which is a constant at a certain point on the measured surface. s (x, y, z) is the signal intensity, which is determined by the relative reflectivity of the reference mirror and the measured surface, the interference fringe envelope v(x, y, z), the central wavelength λ0 of white light, and the phase change α(x, y) caused by the reflection of the measured surface. The peak of v(x, y, z) appears at the zero optical path difference position. By finding these peak points in the xy plane, the profile of the measured surface can be obtained.

[0053] Performing Fourier transform on I(x,y,z) yields:

[0054]

[0055] in, is the wave number corresponding to the central wavelength, Γ(k) is the Fourier transform of v(x, y, z), δ is the unit pulse function, λ0 is the central wavelength of the light source, k is the wave number, c(x, y) is the irradiance change caused by the uneven reflectivity of the surface to be measured, and h is the surface height of the object to be measured.

[0056] (3) Extract the positive frequency component and move it to the frequency center

[0057] Remove zero-frequency and negative frequency components, and move the positive frequency components to the center of the frequency domain. Rearrange equation (2) to obtain:

[0058]

[0059] Equation (3) contains the zero-frequency component and the positive and negative first-order side lobes. After removing the zero-frequency and negative first-order side lobes and shifting the positive first-order side lobes to the center of the frequency domain, we get:

[0060] I′(x,y,k)=Γ(x,y,k) (4)

[0061] Among them, I′(x, y, k) is the positive first-order sidelobe component after frequency shift.

[0062] (4) Obtaining the zero optical path difference position using the center of gravity method

[0063] Perform inverse Fourier transform on I′(x,y,k) and numerically modulate it to obtain the envelope curve v(x,y,z) of the interference signal:

[0064] v(x,y,z)= |F-1 {Γ(x,y,k)}| (5)

[0065] Among them, F -1 is the inverse Fourier transform.

[0066] The peak value of the envelope curve v(x,y,z) corresponds to the position of zero optical path difference. The position of zero optical path difference can be obtained by extracting the envelope peak value using the centroid method:

[0067]

[0068] In the formula, n represents the number of scanning steps, i represents the scanning sampling position, and v i represents the envelope signal at scanning position i.

[0069] (5) 3D shape reconstruction

[0070] The relative height of a single point of the sample under test is obtained by multiplying the zero optical path difference and the scanning step length:

[0071] h= s×C (7)

[0072] Where s is the scanning step size of the piezoelectric ceramic displacer.

[0073] The morphology of the measured surface can be obtained, thereby completing the measurement of surface defects of the silicon carbide wafer.

[0074] The method of the present invention can also be applied to the surface morphology measurement of various samples, and can realize the morphology measurement at the nanometer level.

[0075] The implementation cases of the present invention are only examples for explaining the technical solutions and are not to be interpreted restrictively. For those skilled in the art, even if they do not deviate from the technical essence and creative features of the present invention, any modification, equivalent replacement or equivalent transformation of the embodiments of the present invention shall be regarded as an equivalent embodiment of the present invention and shall be within the scope of protection of the present invention.

[0076] Those skilled in the art will appreciate that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0077] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology, characterized in that: The method comprises: A silicon carbide wafer surface defect detection device based on LDLS white light interferometry is built using a laser-driven white light source, a non-polarized beam splitter, a reflector, a silicon carbide wafer, a microscope objective, a piezoelectric ceramic shifter, a piezoelectric ceramic controller, a focusing lens, and a CCD camera. With the help of piezoelectric ceramic displacer and CCD camera, the original data collection is realized; Perform surface morphology reconstruction on the collected raw data to obtain position information of zero optical path difference; Fitting the position information of zero optical path difference to obtain the profile information of the measured silicon carbide surface; Based on the acquired contour information, the irregularity and height variation of the contour are analyzed, defects on the surface of the silicon carbide wafer are identified and located, and defect detection is achieved.

2. The method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology according to claim 1, characterized in that: The non-polarizing beam splitter prism is used to split the light from the LDLS light source into two beams, one beam is used as the reference light, and the other beam is used to illuminate the sample to be tested; the reflector and the silicon carbide wafer serve as the reference surface and the surface to be tested, respectively, for reflecting the light; the microscope objective is used to focus and amplify the light reflected by the reflector and the surface of the silicon carbide wafer to ensure that the interference fringes are projected onto the detector.

3. The method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology according to claim 1, characterized in that: The piezoelectric ceramic displacer is used to adjust the micro-displacement of the reflector to generate an optical path difference to form interference fringes; the piezoelectric ceramic controller is used to control the movement of the piezoelectric ceramic displacer to ensure the adjustment and stability of the position of the reflector.

4. The method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology according to claim 1, characterized in that: The focusing lens is used to focus the light beam after interference; the CCD camera is used as a detector to obtain the image of the interference fringes to provide raw data for subsequent surface morphology analysis.

5. The method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology according to claim 1, characterized in that: The collection of the original data is as follows: The LDLS white light source emits white light; the non-polarized beam splitter prism receives the white light emitted by the LDLS light source, and divides the white light into a detection light and a reference light, and the detection light is emitted to the surface of the silicon carbide wafer to be measured through the microscope objective lens, and the reference light is emitted to the reflector through the microscope objective lens; The reflector is moved on the piezoelectric ceramic displacer to adjust the optical path of the reference light and the optical path difference between the detection light and the reference light, so that the non-polarized beam splitter prism receives the reflected light from the surface of the silicon carbide wafer and the reflector for beam combining interference and transmits it to the focusing lens to focus on the photosensitive detection surface of the CCD camera, and the interference image is recorded by the CCD camera.

6. The method for detecting defects in silicon carbide wafers based on LDLS white light interferometry technology according to claim 2, characterized in that: The method comprises: obtaining defect morphology information of the surface of the silicon carbide wafer to be tested by reconstructing the surface morphology of the monitored interference image data.

Images