Method and system for evaluating quality of micro-nano film on silicon substrate by using picosecond ultrasonic waves

Abstract

The invention provides a method and a system for evaluating the quality of a micro-nano film on a silicon substrate by using picosecond ultrasonic waves. The method comprises the following steps: S1, measuring the relative change of the reflectivity of a micro-nano film sample on the silicon substrate based on a detection system; s2, analyzing the oscillation signal of the micro-nano film sample on the silicon substrate based on the relative change of the reflectivity of the micro-nano film sample on the silicon substrate; s3, obtaining the sound velocity of the micro-nano film sample on the silicon substrate based on the oscillation signal of the micro-nano film sample on the silicon substrate; and S4, analyzing the acoustic performance and stress distribution of the film based on the sound velocity of the micro-nano film sample on the silicon substrate to obtain an analysis result, and evaluating the quality of the micro-nano film on the silicon substrate based on the analysis result.

Description

Technical Field

[0001] This invention relates to the field of silicon-based micro / nano thin film technology, and more specifically, to a method and system for quality assessment of silicon-based micro / nano thin films using picosecond ultrasound. Background Technology

[0002] Since the beginning of the 21st century, research interest in materials and structural systems has shifted from the macroscopic to the microscopic (micro-nano systems and micro-systems), and from monolithic materials with uniform composition to multilayer films and ultrathin films composed of different components. Compared with bulk materials, thin films have some fundamental differences, such as size effects, special microstructures, the need to be attached to a substrate, and the influence of surfaces and interfaces. These special characteristics in size, morphology, and structure cause thin films to exhibit many anomalous effects that are quite different from those of bulk materials.

[0003] In semiconductor manufacturing, accurate measurement of the acoustic properties of silicon-based thin-film materials is crucial. For example, to ensure the performance consistency of fabricated radio frequency filters, characterizing the quality differences between different wafers and the uniformity of the same wafer are critical steps before device fabrication. Commonly used quality assessment equipment includes instruments for thin-film thickness testing such as continuous spectrum analyzers, crystal molecular structure testing such as Fourier transform infrared spectroscopy (FT-IR) and XRD diffractometers, and micro-area structure observation such as SEM and TEM. However, acoustic performance assessment still has shortcomings. Traditional measurement techniques such as atomic force microscopy (AFM) and scanning electron microscopy (SEM), while providing surface morphology information, cannot directly measure the acoustic properties of thin films. Therefore, currently, the acoustic performance testing of thin films on silicon must wait until after device fabrication, judging it by observing device performance. Furthermore, these techniques are often destructive and unsuitable for online production processes.

[0004] Traditional acoustic performance characterization methods for bulk materials, namely ultrasonic testing, have limitations in the field of complex thin films due to the need for macroscopically sized transducers and pulsed laser frequencies in the kHz range. The thickness of some complex material films can even reach several nanometers to tens of nanometers, posing a challenge to the resolution of ultrasonic testing. With the development of femtosecond pulsed lasers (pulse width <200 fs) and advancements in laser mode-locking technology, researchers have been able to generate ultrasound in the ~100 GHz range, opening up a new research field—picosecond ultrasonics (PU). As a non-contact, non-destructive pump-probe laser acoustic technique, it can achieve a longitudinal spatial resolution within the sample accurate to 0.1 nm, and is now considered suitable for the non-destructive evaluation of micro and nano-thin films. The conventional method for measuring sound velocity using picosecond ultrasonics is the ultrasonic pulse-echo method, but the excitation and detection of PU rely on opaque materials (requiring the addition of a transducer layer, thus no longer a non-destructive evaluation). Furthermore, deriving the sound velocity requires prior knowledge of the thickness of each layer, and obtaining a distribution map is time-consuming. Therefore, there is an urgent need for a picosecond ultrasonic method for rapid online quality assessment of the acoustic properties of micro / nano films on silicon substrates without damage. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for quality assessment of micro / nano thin films on silicon substrates using picosecond ultrasound.

[0006] A method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound, provided by the present invention, includes:

[0007] Step S1: Measure the relative change in reflectivity of the micro / nano thin film sample on the silicon substrate using the detection system;

[0008] Step S2: Analyze the oscillation signal of the micro / nano thin film sample on the silicon substrate based on the relative change of reflectivity;

[0009] Step S3: Obtain the sound velocity of the micro-nano thin film sample on the silicon substrate based on the oscillation signal of the micro-nano thin film sample on the silicon substrate;

[0010] Step S4: Analyze the acoustic properties and stress distribution of the thin film based on the sound velocity of the micro / nano film sample on the silicon substrate to obtain the analysis results, and evaluate the quality of the micro / nano film on the silicon substrate based on the analysis results.

[0011] Preferably, the detection system includes: a femtosecond laser, a polarizing beam splitter, a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, an objective lens, an optical delay line, a depolarizing beam splitter, a non-polarizing beam splitter, an optical camera, a balanced photodetector, and a lock-in amplifier.

[0012] The femtosecond laser emits a laser beam, which passes through the polarization beam splitter to obtain a pump beam and a probe beam.

[0013] The pump beam passes sequentially through a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, and an objective lens to irradiate the sample, thereby exciting the sample to an excited state.

[0014] The probe beam passes sequentially through the optical delay line and the depolarizing beam splitter to obtain a probe beam and a reference beam. The reference beam enters the balanced photodetector to eliminate interference from energy fluctuations caused by environmental factors affecting the laser. The probe beam passes sequentially through the dichroic mirror and the objective lens to illuminate the same position on the sample, and a reflected probe beam is obtained. The reflected probe beam passes sequentially through the objective lens, the dichroic mirror, the depolarizing beam splitter, and the nonpolarizing beam splitter to enter the optical camera and the balanced photodetector, respectively. The optical camera is used to acquire the shape of the light spot on the sample. The balanced photodetector is used to obtain light intensity information, and the light intensity information is transmitted to a lock-in amplifier for demodulation to obtain the relative change in the sample reflectivity.

[0015] Preferably, the femtosecond laser includes: when the wavelength of the femtosecond laser meets a preset requirement, picosecond ultrasound is excited by absorption on a micro / nano thin film sample on a silicon substrate;

[0016]

[0017] Where n(λ) and k(λ) represent the real and imaginary parts of the complex refractive index, respectively; α represents the complex refractive index of the material; α(λ) represents the optical absorption coefficient; and λ represents the wavelength.

[0018] Preferably, step S2 includes:

[0019] ΔR(t)=ΔT(t)+B(t)

[0020] Where ΔR(t) represents the relative change in reflectivity of the silicon-based micro / nano thin film sample; ΔT(t) represents the relative change in temperature of the silicon-based micro / nano thin film sample; and B(t) represents the Brillouin oscillation signal of the silicon-based micro / nano thin film sample.

[0021] A temperature model was constructed based on two decay factors to fit the cooling process;

[0022]

[0023] Where k represents the heat absorption efficiency of the sample, e represents the natural constant, t0 represents the heat absorption of the two heat transfer media, A1 and A2 represent the heat absorption amplitude of the two heat transfer media, α1 and α2 represent the thermal conductivity coefficients of the two heat transfer media respectively, c1 and c2 represent the constants in the fitting process, and the sum of the two represents the temperature change caused by the sample being heated under laser irradiation for a long time.

[0024] The Brillouin oscillation signal B(t) of the micro-nano thin film sample on silicon is compressed to obtain the compressed B′(t) of the micro-nano thin film sample on silicon.

[0025]

[0026] Where Ψ represents the observation matrix, and Φ represents the sparse representation matrix. Represent the Fourier transform and frequency domain of B(t).

[0027] Preferably, step S3 includes:

[0028]

[0029] Where n represents the refractive index of the micro / nano thin film sample on a silicon substrate, v represents the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light; f BO This indicates the frequency of the Brillouin oscillation.

[0030] A system for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound, according to the present invention, comprises:

[0031] Module M1: Measure the relative change in reflectivity of micro / nano thin film samples on a silicon substrate using a detection system;

[0032] Module M2: Analysis of oscillation signals of micro / nano thin film samples on silicon based on the relative change of reflectivity;

[0033] Module M3: Obtains the sound velocity of a silicon-based micro / nano thin film sample based on the oscillation signal of the sample.

[0034] Module M4: Based on the sound velocity of micro / nano thin film samples on silicon, the acoustic properties and stress distribution of the thin film are analyzed to obtain the analysis results, and the quality of the micro / nano thin film on silicon is evaluated based on the analysis results.

[0035] Preferably, the detection system includes: a femtosecond laser, a polarizing beam splitter, a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, an objective lens, an optical delay line, a depolarizing beam splitter, a non-polarizing beam splitter, an optical camera, a balanced photodetector, and a lock-in amplifier.

[0036] The femtosecond laser emits a laser beam, which passes through the polarization beam splitter to obtain a pump beam and a probe beam.

[0037] The pump beam passes sequentially through a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, and an objective lens to irradiate the sample, thereby exciting the sample to an excited state.

[0038] The probe beam passes sequentially through the optical delay line and the depolarizing beam splitter to obtain a probe beam and a reference beam. The reference beam enters the balanced photodetector to eliminate interference from energy fluctuations caused by environmental factors affecting the laser. The probe beam passes sequentially through the dichroic mirror and the objective lens to illuminate the same position on the sample, and a reflected probe beam is obtained. The reflected probe beam passes sequentially through the objective lens, the dichroic mirror, the depolarizing beam splitter, and the nonpolarizing beam splitter to enter the optical camera and the balanced photodetector, respectively. The optical camera is used to acquire the shape of the light spot on the sample. The balanced photodetector is used to obtain light intensity information, and the light intensity information is transmitted to a lock-in amplifier for demodulation to obtain the relative change in the sample reflectivity.

[0039] Preferably, the femtosecond laser includes: when the wavelength of the femtosecond laser meets a preset requirement, picosecond ultrasound is excited by absorption on a micro / nano thin film sample on a silicon substrate;

[0040]

[0041] Where n(λ) and k(λ) represent the real and imaginary parts of the complex refractive index, respectively; α represents the complex refractive index of the material; α(λ) represents the optical absorption coefficient; and λ represents the wavelength.

[0042] Preferably, the module M2 includes:

[0043] ΔR(t)=ΔT(t)+B(t)

[0044] Where ΔR(t) represents the relative change in reflectivity of the silicon-based micro / nano thin film sample; ΔT(t) represents the relative change in temperature of the silicon-based micro / nano thin film sample; and B(t) represents the Brillouin oscillation signal of the silicon-based micro / nano thin film sample.

[0045] A temperature model was constructed based on two decay factors to fit the cooling process;

[0046]

[0047] Where k represents the heat absorption efficiency of the sample, e represents the natural constant, t0 represents the heat absorption of the two heat transfer media, A1 and A2 represent the heat absorption amplitude of the two heat transfer media, α1 and α2 represent the thermal conductivity coefficients of the two heat transfer media respectively, c1 and c2 represent the constants in the fitting process, and the sum of the two represents the temperature change caused by the sample being heated under laser irradiation for a long time.

[0048] The Brillouin oscillation signal B(t) of the micro-nano thin film sample on silicon is compressed to obtain the compressed B′(t) of the micro-nano thin film sample on silicon.

[0049]

[0050] Where Ψ represents the observation matrix, and Φ represents the sparse representation matrix. Represent the Fourier transform and frequency domain of B(t).

[0051] Preferably, the module M3 includes:

[0052]

[0053] Where n represents the refractive index of the micro / nano thin film sample on a silicon substrate, v represents the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light; f BO This indicates the frequency of the Brillouin oscillation.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. This invention fully considers the characteristics of micro / nano thin films on silicon and designs a picosecond ultrasonic system for thin film quality assessment. Utilizing the characteristic that Brillouin oscillations (BO) contain only a few fixed oscillation frequencies, a CS super-resolution sampling scheme is incorporated into the device design. While maintaining the original resolution, the measurement time can be shortened by more than 10 times. By rationally selecting the laser wavelength, treating the thin film layer as a transparent medium and the substrate's single-crystal silicon as an opaque medium, an excitation and detection scheme is designed to directly excite ultrasound on the substrate without the need for an additional transducer layer growth step. Not only can the sound velocity be measured using the BO frequency induced by PU propagation, but the thickness information can also be deduced from the position of the BO frequency change and the sound velocity information, simplifying the testing process.

[0056] 2. This invention can obtain the sound velocity distribution map of micro-nano thin films on a complete silicon wafer at a relatively fast speed, which provides strong assistance for the acoustic property quality assessment of the thin film and further assists in the analysis of stress distribution, piezoelectric effect, etc. of the thin film.

[0057] 3. This invention focuses on the measurement of sound velocity in micro- and nano-thin films on silicon substrates. It employs compressed sensing technology to improve measurement efficiency and considers the influence of piezoelectric effect on sound velocity. Attached Figure Description

[0058] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0059] Figure 1 This is a schematic diagram of time-domain Brillouin scattering.

[0060] Figure 2 This is a schematic diagram of the optical path.

[0061] Figures 3a to 3b This is a schematic diagram of the sample and its corresponding data.

[0062] Figure 4a This is a schematic diagram of the temperature curve fitting.

[0063] Figure 4b This is a schematic diagram of the extracted BO signal.

[0064] Figure 5a The image shows the sound velocity distribution of the LN thin film in Z-cut LNOI.

[0065] Figure 5b The sound velocity distribution diagram of silicon oxide in Z-cut LNOI.

[0066] Figure 6a This is a FIB scan of LNOI.

[0067] Figure 6b This is a schematic diagram of the thickness obtained through time-frequency analysis. Detailed Implementation

[0068] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the scope of protection of the present invention.

[0069] Example 1

[0070] A method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound, provided by the present invention, includes:

[0071] Step S1: Measure the relative change in reflectivity of the micro / nano thin film sample on the silicon substrate using the detection system;

[0072] Step S2: Analyze the oscillation signal of the micro / nano thin film sample on the silicon substrate based on the relative change of reflectivity;

[0073] Step S3: Obtain the sound velocity of the micro-nano thin film sample on the silicon substrate based on the oscillation signal of the micro-nano thin film sample on the silicon substrate;

[0074] Step S4: Analyze the acoustic properties and stress distribution of the thin film based on the sound velocity of the micro / nano film sample on the silicon substrate to obtain the analysis results, and evaluate the quality of the micro / nano film on the silicon substrate based on the analysis results.

[0075] Specifically, the detection system includes: a femtosecond laser, a polarizing beam splitter PBS, a frequency doubling crystal BBO, an acousto-optic modulator AOM, a dichroic mirror, an objective lens, an optical delay line, a depolarizing beam splitter NPBS, a nonpolarizing beam splitter prism, an optical camera CCD, a balanced photodetector, and a lock-in amplifier.

[0076] The femtosecond laser emits a laser beam, which passes through the polarization beam splitter to obtain a pump beam and a probe beam. In this embodiment, to ensure that the probe beam does not affect the pump beam, the energy ratio of the probe beam to the pump beam is no greater than 1:10.

[0077] The pump beam passes sequentially through a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, and an objective lens to irradiate the sample, thereby exciting the sample to an excited state.

[0078] In this embodiment, the pump light is frequency doubled to a wavelength of 400nm by a frequency doubling crystal (BBO), and then its amplitude is sinusoidally modulated by an acousto-optic modulator (AOM) at a frequency of 100kHz.

[0079] The probe beam passes sequentially through the optical delay line and the depolarizing beam splitter to obtain a probe beam and a reference beam. The reference beam enters the balanced photodetector to eliminate interference from energy fluctuations caused by environmental factors affecting the laser. The probe beam passes sequentially through the dichroic mirror and the objective lens to illuminate the same position on the sample, and a reflected probe beam is obtained. The reflected probe beam passes sequentially through the objective lens, the dichroic mirror, the depolarizing beam splitter, and the nonpolarizing beam splitter to enter the optical camera and the balanced photodetector, respectively. The optical camera is used to acquire the shape of the light spot on the sample. The balanced photodetector is used to obtain light intensity information, and the light intensity information is transmitted to a lock-in amplifier for demodulation to obtain the relative change in the sample reflectivity.

[0080] The frequency doubling crystal is used to double the frequency of the pump light and adjust the wavelength of the pump light.

[0081] The acousto-optic modulator is used to modulate the amplitude of the pump light;

[0082] In this embodiment, the probe light passes through an optical delay line composed of an electric displacement stage and a hollow retroreflector, and its optical path difference with the pump light can be controlled within the range of -10 mm to 190 mm. The probe light is then split again into a reference portion and a probe portion by a non-polarizing beam splitter (NPBS). The reference portion is directly received by a balanced photodetector (BPD), while the probe portion and the pump light are combined by a cold light mirror and then focused onto the sample surface by a 40x objective lens, resulting in a focused spot radius of 10–20 micrometers. The probe light returns via the primary optical path and is finally received by the balanced photodetector. The final amplitude value of the probe light is transmitted to a lock-in amplifier in the form of an electrical signal, the amplitude change is calculated, and the result is recorded by a data acquisition card.

[0083] Specifically, the femtosecond laser includes: when the wavelength of the femtosecond laser meets the preset requirements, picosecond ultrasound is absorbed and excited on a micro / nano thin film sample on a silicon substrate;

[0084]

[0085] Where n(λ) and k(λ) represent the real and imaginary parts of the complex refractive index, respectively; α represents the complex refractive index of the material; α(λ) represents the optical absorption coefficient; and λ represents the wavelength.

[0086] In this embodiment, when a femtosecond laser irradiates an opaque solid surface, the energy is absorbed by the solid surface, causing the region to rapidly heat up and expand, thereby generating picosecond ultrasound. Micro- and nano-thin films on silicon are transparent in the 400-800nm ​​wavelength range and cannot absorb laser light; however, single-crystal silicon can absorb pulsed laser light with wavelengths of 400nm and below to excite ultrasound, while beams of 400nm and above will either be transmitted or incompletely absorbed and unable to excite ultrasound. Therefore, this invention employs a method of 400nm excitation for 800nm ​​detection.

[0087] Specifically, step S2 includes:

[0088] ΔR(t)=ΔT(t)+B(t)

[0089] Where ΔR(t) represents the relative change in reflectivity of the silicon-based micro / nano thin film sample; ΔT(t) represents the relative change in temperature of the silicon-based micro / nano thin film sample; and B(t) represents the Brillouin oscillation signal of the silicon-based micro / nano thin film sample.

[0090] A temperature model was constructed based on two decay factors to fit the cooling process;

[0091]

[0092] Where k represents the heat absorption efficiency of the sample, e represents the natural constant, t0 represents the heat absorption of the two heat transfer media, A1 and A2 represent the heat absorption amplitude of the two heat transfer media, α1 and α2 represent the thermal conductivity coefficients of the two heat transfer media respectively, c1 and c2 represent the constants in the fitting process, and the sum of the two represents the temperature change caused by the sample being heated under laser irradiation for a long time.

[0093] The Brillouin oscillation signal B(t) of the micro-nano thin film sample on silicon is compressed to obtain the compressed B′(t) of the micro-nano thin film sample on silicon.

[0094]

[0095] Where Ψ represents the observation matrix, and Φ represents the sparse representation matrix. Represent the Fourier transform and frequency domain of B(t).

[0096] Specifically, step S3 includes:

[0097]

[0098] Where n represents the refractive index of the micro / nano thin film sample on a silicon substrate, v represents the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light; f BO f BO It is the frequency corresponding to the peak value of B′(t) in the frequency domain, i.e., the Brillouin oscillation frequency.

[0099] Specifically, step S4 includes:

[0100] After steps S1, S2, and S3, the actual sound velocity of the thin film material can be measured. The theoretical sound velocity required for the silicon-based micro / nano thin film sample is calculated during the device design phase, based on the material's modulus (or elastic constant) and density. The measured value needs to be close to the theoretical value. If the sound velocity is 5% or more lower than the theoretical value, it indicates damage during fabrication and a degradation in acoustic performance. If the sound velocity is 5% or more higher than the theoretical value, it indicates high stress in the sample; the magnitude of this stress can be determined by the ratio of the two values.

[0101] This invention also provides a system for quality assessment of micro / nano films on a silicon substrate using picosecond ultrasound. The system for quality assessment of micro / nano films on a silicon substrate using picosecond ultrasound can be implemented by executing the process steps of the method for quality assessment of micro / nano films on a silicon substrate using picosecond ultrasound. That is, those skilled in the art can understand the method for quality assessment of micro / nano films on a silicon substrate using picosecond ultrasound as a preferred embodiment of the system for quality assessment of micro / nano films on a silicon substrate using picosecond ultrasound.

[0102] Example 2

[0103] Example 2 is a preferred example of Example 1.

[0104] This invention discloses a method for quality assessment of silicon-based micro / nano thin films using picosecond ultrasound. The method includes: a femtosecond laser emitted from a femtosecond laser is divided into a pump light and a probe light. The pump light excites picosecond ultrasound on the silicon substrate via a thermoelastic mechanism, and the probe light monitors the sample changes caused by the pump light. This method can simultaneously plot the sound velocity distribution of the thin film on the silicon substrate without external transducer materials and with unknown thickness, allowing for analysis of the film's acoustic properties and stress distribution. Furthermore, compressed sensing (CS) reduces the number of sampling points, shortening the overall testing time by more than tenfold. This invention provides a non-contact, non-destructive characterization method for silicon-based micro / nano thin films, enabling more accurate quality assessment.

[0105] like Figure 1 As shown, when picosecond ultrasound propagates in an optically transparent material, the laser pulse preferentially interacts with phonons in the picosecond ultrasound acoustic pulse spectrum that satisfy the Brillouin scattering (BS) condition. The weak light pulse scattered by the coherent phonons interferes with the photodetector with the light pulses reflected from various interfaces of the sample, which have significantly higher amplitudes. The recorded transient light reflectivity signal is proportional to the first-order electric field product of the two light fields, realizing the heterodyne of the weak field to the strong field. Due to the change in the spatial position of the picosecond ultrasound, the relative phase of the light scattered by the acoustic pulse and reflected from the fixed interface changes continuously with time, thus the measured signal changes with time. When the picosecond ultrasound propagates at a constant speed in a spatially homogeneous medium, the phase difference between the interfering light fields changes linearly with time, causing the signal amplitude to vary sinusoidally at a frequency exactly equal to the Brillouin frequency. The contribution of acoustically induced oscillations to the reflectivity signal is usually called Brillouin oscillation. This application is called time-domain Brillouin scattering (TDBS), a photoacoustic interferometry method. When the probe light is perpendicular to the sample surface, the frequency f of B0... BO satisfy:

[0106]

[0107] Where n is the refractive index of the material, v is the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light. Therefore, TDBS can directly measure the sound velocity of a material when the thickness of the medium is unknown, and can be used in fields such as elastic constant testing of transparent materials, sound attenuation analysis, and depth-direction stress distribution testing.

[0108] The platinum absorption (PU) that causes the BO phenomenon arises from the thermal expansion triggered by the transient absorption of femtosecond laser light in the shallow region of the sample surface. The ability to generate picosecond-level pulses can be determined by the optical absorption depth (OAD) of the laser within the corresponding material. At a wavelength of λ, OAD can be measured by the material's complex refractive index. Calculations are performed, and the following conditions are met:

[0109]

[0110] Where n(λ) and k(λ) are the real and imaginary parts of the complex refractive index, and α(λ) is the optical absorption coefficient, the reciprocal of which is the OAD. The film thickness on the silicon substrate is typically on the submicron scale, and characterizing this structure requires ultrasound with wavelengths on the order of hundreds of nanometers. The single-crystal silicon OAD on the substrate shrinks to below 100 nm at λ = 400 nm, satisfying the requirement for long ultrasound waves. Furthermore, the piezoelectric films and silicon oxide commonly used on silicon substrates are usually transparent at λ = 400 nm, also meeting the conditions for observing the OAD.

[0111] Because picosecond ultrasound can reach frequencies of nearly 100 GHz with pulse widths of less than 50 ps, ​​a detector operating within the required frequency range is needed to successfully detect and extract picosecond ultrasound information. Furthermore, according to the Nyquist sampling theorem, to avoid signal distortion, the detector's sampling rate needs to reach ~200 GHz. Therefore, pump-probe, as an optical detection technique that trades time for a higher sampling rate, is well-suited for studying ultrafast processes and plays a crucial role in picosecond ultrasound signal detection. However, fully representing the sequential propagation and reflection of PU within multiple thin films on a silicon substrate requires acquiring at least approximately 1 ns of effective signal. This necessitates continuous sampling by the pump system for over 2 hours, while generating a velocity distribution map requires scanning at least 20 points on the wafer, totaling over 40 hours.

[0112] To achieve the function of quickly drawing sound velocity distribution maps, a super-resolution sampling scheme based on compressed sensing (CS) was adopted. By reducing the number of sampling points, the acquisition speed is accelerated, and the test time is shortened by more than ten times while maintaining the original resolution.

[0113] The ~800nm ​​pulsed laser emitted by the femtosecond laser is split into a higher-energy pump beam and a lower-energy probe beam by a PBS. Before reaching the sample, the probe beam energy is controlled to be no more than 1 / 10 of the pump beam energy to avoid secondary excitation of the PU by the probe beam, which would introduce noise. The pump beam is frequency-doubled to 400nm by a BBO crystal, meeting the experimental requirements for POI materials. Since the amplitude of the probe beam is typically affected by BO within a range of 10... -6The pump beam, after frequency doubling, needs to be amplitude modulated by an acousto-optic modulator (AOM) to facilitate the subsequent extraction of the weak boron (BO) signal. The probe beam, after passing through a delay line, is split again into a probe portion and a reference portion, which are received separately by a balanced photodetector (BPD) to eliminate errors introduced by laser power fluctuations. The pump and probe beams are combined by a cold-light mirror and focused onto the sample surface through a 40X objective lens. After focusing, the spot radius of the pump beam is 10–20 micrometers, while the probe beam is slightly larger to ensure complete coverage of the pump beam even with slight beam jitter. An optical interference module is introduced to measure the contribution of sample surface displacement and detect transient changes in optical complex reflectivity, significantly improving measurement sensitivity and signal-to-noise ratio. To meet the requirements for two-dimensional scanning of the wafer surface, the experimental sample is fixed on a five-dimensional displacement sample stage. By moving the sample stage to change the laser focus on different positions on the wafer, the pitch and yaw angles of the sample can also be adjusted to ensure that the laser beam is perpendicularly irradiated onto the wafer surface. The probe light reflected from the sample surface is filtered and then acquired by the BPD. Its amplitude is transmitted to the LIA as an electrical signal. Finally, according to the CS sampling requirements, the PC controls the position of the delay line and simultaneously collects the measurement results provided by the LIA. The complete system is as follows: Figure 2 As shown.

[0114] Figure 3b ) is in Figure 3a The results of a single-point experiment on the sample are shown in the figure. Figure 3a In this context, LN stands for lithium niobate; SiO2 stands for silicon oxide. Figure 3b The vertical axis represents the change in reflectivity, and the horizontal axis represents time. The complete signal consists of a rapid heating process of approximately 10 ps followed by an exponentially decaying slow cooling process; the signal itself does not directly satisfy the characteristics of frequency domain sparsity. This invention employs a segmented sampling strategy, preprocessing the data after sampling before performing sensor reconstruction.

[0115] The relative change ΔR of the sample reflectance during measurement can be decomposed into two parts: temperature change and BO.

[0116] ΔR(t)=ΔT(t)+B(t)

[0117] ΔT(t) consists of two parts: rapid heating and slow cooling. The heating part reflects the electronic heat capacity and lattice heat capacity of the material, and the heating rate can be numerically fitted by introducing the Sigmoid function into the two-temperature equation. The cooling part conforms to the Fourier heat conduction model, and the cooling rate is related to the thermal conductivity of the material. Considering that the excitation point of PU is at the interface between silicon and the thin film, the heat dissipation process occurs simultaneously in both media. Therefore, a temperature model composed of two attenuation factors is used to fit the cooling process.

[0118]

[0119] The extracted ΔT(t) process is complete and has a smaller residual value, such as Figure 4a As shown in the figure.

[0120] B(t) after removing the temperature curve is as follows Figure 4b As shown in the figure, it consists of two oscillation signals of specific frequencies, corresponding sequentially to the BO caused by the propagation of PU within silicon oxide and the piezoelectric film:

[0121]

[0122] B(t) exhibits sparseness in the frequency domain, with the peak values ​​of the spectrum including the BO frequencies corresponding to both media. Therefore, the prerequisite for performing CS on ΔR(t) is to extract the temperature from the signal. Since the arrival time of the excitation light on the sample is known, a high-resolution method can be used to acquire the heating process to ensure the accuracy of the fitting, while CS reduces the number of sampling points during the cooling process. One-dimensional signals can be directly sampled from the original experiment using a Gaussian random sequence, achieving compressed signal sampling.

[0123] The compressed signal B′(t) and the original signal B(t) satisfy the following correspondence:

[0124]

[0125] As known quantities, the product of Ψ and Φ forms the restored matrix, from which the BO frequencies can be solved in sparse form.

[0126] After the acquisition and restoration process by CS, the time for a complete sampling is reduced to approximately 600 seconds. This greatly facilitates multi-point sampling on a complete wafer, and it is now possible to acquire sound velocity information from more than twenty different locations on a single wafer plane within 4 hours.

[0127] Finally, 40 sampling points can be selected evenly distributed on the wafer surface, and the sound velocities of LN and silicon oxide at each point can be calculated sequentially to plot the sound velocity distribution map of the wafer, as shown below. Figure 5a , Figure 5b As shown.

[0128] Furthermore, time-frequency analysis can be applied to B(t). By analyzing the time points of frequency changes and combining this with the derived sound velocity value, the thickness information of the thin film can be inferred. For example... Figure 6a , Figure 6b As shown, the thickness of each layer profile of the LNOI sample measured by FIB is basically consistent with the calculated values. Furthermore, information such as the stress distribution and elastic constants of the thin film on the wafer can also be provided.

[0129] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0130] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound, characterized in that, include: Step S1: Measure the relative change in reflectivity of the micro / nano thin film sample on the silicon substrate using the detection system; Step S2: Analyze the oscillation signal of the micro / nano thin film sample on the silicon substrate based on the relative change of reflectivity; Step S3: Obtain the sound velocity of the micro-nano thin film sample on the silicon substrate based on the oscillation signal of the micro-nano thin film sample on the silicon substrate; Step S4: Analyze the acoustic properties and stress distribution of the thin film based on the sound velocity of the micro / nano film sample on the silicon substrate to obtain the analysis results, and evaluate the quality of the micro / nano film on the silicon substrate based on the analysis results.

2. The method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 1, characterized in that, The detection system includes: a femtosecond laser, a polarizing beam splitter, a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, an objective lens, an optical delay line, a depolarizing beam splitter, a nonpolarizing beam splitter, an optical camera, a balanced photodetector, and a lock-in amplifier. The femtosecond laser emits a laser beam, which passes through the polarization beam splitter to obtain a pump beam and a probe beam. The pump beam passes sequentially through a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, and an objective lens to irradiate the sample, thereby exciting the sample to an excited state. The probe beam passes sequentially through the optical delay line and the depolarizing beam splitter to obtain a probe beam and a reference beam. The reference beam enters the balanced photodetector to eliminate interference from energy fluctuations caused by environmental factors affecting the laser. The probe beam passes sequentially through the dichroic mirror and the objective lens to illuminate the same position on the sample, and a reflected probe beam is obtained. The reflected probe beam passes sequentially through the objective lens, the dichroic mirror, the depolarizing beam splitter, and the nonpolarizing beam splitter to enter the optical camera and the balanced photodetector, respectively. The optical camera is used to acquire the shape of the light spot on the sample. The balanced photodetector is used to obtain light intensity information, and the light intensity information is transmitted to a lock-in amplifier for demodulation to obtain the relative change in the sample reflectivity.

3. The method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 2, characterized in that, The femtosecond laser includes: when the wavelength of the femtosecond laser meets the preset requirements, picosecond ultrasound is excited by absorption on a micro / nano thin film sample on a silicon substrate; Where n(λ) and k(λ) represent the real and imaginary parts of the complex refractive index, respectively; α represents the complex refractive index of the material; α(λ) represents the optical absorption coefficient; and λ represents the wavelength.

4. The method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 1, characterized in that, Step S2 includes: ΔR(t)=ΔT(t)+B(t) Where ΔR(t) represents the relative change in reflectivity of the silicon-based micro / nano thin film sample; ΔT(t) represents the relative change in temperature of the silicon-based micro / nano thin film sample; and B(t) represents the Brillouin oscillation signal of the silicon-based micro / nano thin film sample. A temperature model was constructed based on two decay factors to fit the cooling process; Where k represents the heat absorption efficiency of the sample, e represents the natural constant, t0 represents the heat absorption of the two heat transfer media, A1 and A2 represent the heat absorption amplitude of the two heat transfer media, α1 and α2 represent the thermal conductivity coefficients of the two heat transfer media respectively, c1 and c2 represent the constants in the fitting process, and the sum of the two represents the temperature change caused by the sample being heated under laser irradiation for a long time. The Brillouin oscillation signal B(t) of the micro-nano thin film sample on silicon is compressed to obtain the compressed B′(t) of the micro-nano thin film sample on silicon. Where Ψ represents the observation matrix, and Φ represents the sparse representation matrix. Represent the Fourier transform and frequency domain of B(t).

5. The method for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 1, characterized in that, Step S3 includes: Where n represents the refractive index of the micro / nano thin film sample on a silicon substrate, v represents the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light; f BO This represents the frequency of Brillouin oscillations.

6. A system for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound, characterized in that, include: Module M1: Measure the relative change in reflectivity of micro / nano thin film samples on a silicon substrate using a detection system; Module M2: Analysis of oscillation signals of micro / nano thin film samples on silicon based on the relative change of reflectivity; Module M3: Obtains the sound velocity of a silicon-based micro / nano thin film sample based on the oscillation signal of the sample. Module M4: Based on the sound velocity of micro / nano thin film samples on silicon, the acoustic properties and stress distribution of the thin film are analyzed to obtain the analysis results, and the quality of the micro / nano thin film on silicon is evaluated based on the analysis results.

7. The system for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 6, characterized in that, The detection system includes: a femtosecond laser, a polarizing beam splitter, a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, an objective lens, an optical delay line, a depolarizing beam splitter, a nonpolarizing beam splitter, an optical camera, a balanced photodetector, and a lock-in amplifier. The femtosecond laser emits a laser beam, which passes through the polarization beam splitter to obtain a pump beam and a probe beam. The pump beam passes sequentially through a frequency doubling crystal, an acousto-optic modulator, a dichroic mirror, and an objective lens to irradiate the sample, thereby exciting the sample to an excited state. The probe beam passes sequentially through the optical delay line and the depolarizing beam splitter to obtain a probe beam and a reference beam. The reference beam enters the balanced photodetector to eliminate interference from energy fluctuations caused by environmental factors affecting the laser. The probe beam passes sequentially through the dichroic mirror and the objective lens to illuminate the same position on the sample, obtaining reflected probe light. The reflected probe light passes sequentially through the objective lens, the dichroic mirror, the depolarizing beam splitter, and the nonpolarizing beam splitter to enter the optical camera and the balanced photodetector, respectively. The optical camera acquires the shape of the light spot on the sample. The balanced photodetector obtains the light intensity information, which is then transmitted to a lock-in amplifier for demodulation to obtain the relative change in the sample's reflectivity. The probe beam passes through the optical delay line, creating an optical path difference between the probe beam and the pump beam, so that the probe beam and the pump beam illuminate the same position on the sample through the objective lens with a preset time difference.

8. The system for quality assessment of micro / nano thin films on silicon substrates using picosecond ultrasound according to claim 7, characterized in that, The femtosecond laser includes: when the wavelength of the femtosecond laser meets the preset requirements, picosecond ultrasound is excited by absorption on a micro / nano thin film sample on a silicon substrate; Where n(λ) and k(λ) represent the real and imaginary parts of the complex refractive index, respectively; α represents the complex refractive index of the material; α(λ) represents the optical absorption coefficient; and λ represents the wavelength.

9. The system for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 6, characterized in that, The module M2 includes: ΔR(t)=ΔT(t)+B(t) Where ΔR(t) represents the relative change in reflectivity of the silicon-based micro / nano thin film sample; ΔT(t) represents the relative change in temperature of the silicon-based micro / nano thin film sample; and B(t) represents the Brillouin oscillation signal of the silicon-based micro / nano thin film sample. A temperature model was constructed based on two decay factors to fit the cooling process; Where k represents the heat absorption efficiency of the sample, e represents the natural constant, t0 represents the heat absorption of the two heat transfer media, A1 and A2 represent the heat absorption amplitude of the two heat transfer media, α1 and α2 represent the thermal conductivity coefficients of the two heat transfer media respectively, c1 and c2 represent the constants in the fitting process, and the sum of the two represents the temperature change caused by the sample being heated under laser irradiation for a long time. The Brillouin oscillation signal B(t) of the micro-nano thin film sample on silicon is compressed to obtain the compressed B′(t) of the micro-nano thin film sample on silicon. Where Ψ represents the observation matrix, and Φ represents the sparse representation matrix. Represent the Fourier transform and frequency domain of B(t).

10. The system for quality assessment of micro / nano thin films on a silicon substrate using picosecond ultrasound according to claim 6, characterized in that, The module M3 includes: Where n represents the refractive index of the micro / nano thin film sample on a silicon substrate, v represents the longitudinal wave velocity of the pulsed ultrasound, and λ is the wavelength of the probe light; f BO This represents the frequency of Brillouin oscillations.

Images