Film layer measurement device and method

Abstract

The application relates to the field of semiconductor measurement technology, in particular to a film layer measurement device and method. The film layer measurement method comprises the following steps: irradiating a to-be-measured position on the upper surface of a to-be-measured film layer with pump light, the pump light causes the second-order nonlinear optical coefficient of the to-be-measured film layer to change and forms a local disturbance, the local disturbance propagates back and forth between the upper surface and the lower surface of the to-be-measured film layer; irradiating the to-be-measured position with probe light, the probe light generates frequency-doubled signal light based on the second-order nonlinear optical effect, when the local disturbance propagates to the upper surface, the light intensity of the frequency-doubled signal light reaches a peak value; receiving the frequency-doubled signal light, and obtaining the thickness of the to-be-measured film layer based on the time interval between adjacent two light intensity peaks of the frequency-doubled signal light. The application is at least beneficial to improving the film thickness measurement precision.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor measurement technology, and particularly relates to a film layer measurement device and method. Background Technology

[0002] As Moore's Law advances to 3nm and below, the miniaturization of semiconductor devices has profoundly reshaped the design and manufacturing logic of semiconductor structures, leading to a greater demand for precise measurement of film thickness in semiconductor structures. At the device architecture level, three-dimensional structures have become mainstream. From fin field-effect transistors and all-around gate nanosheets to three-dimensional NAND flash memory stacking technology, the complexity of metal interconnect systems has significantly increased. Taking metal interconnect vias as an example, their aspect ratio has exceeded 10:1. If the local film thickness deviation exceeds ±0.3nm, it will directly lead to abnormal contact resistance or uneven current density distribution. For example, if the film thickness of a metal interconnect via is too thin, the contact interface resistance will surge, causing signal transmission delay; while uneven film thickness of Cu interconnects will cause current to concentrate in local areas, accelerating electromigration effects and ultimately affecting the electrical performance and long-term reliability of the device. Therefore, traditional two-dimensional measurement methods are no longer suitable for more complex semiconductor structures.

[0003] Film thickness, as a geometric dimension benchmark, directly determines the topological consistency of the three-dimensional structure. For example, in a device with a fully surrounding gate structure, the film thickness deviation of the TiN gate will change the coupling strength between the gate and the channel, causing the device threshold voltage to drift and affecting the stability of the chip's logic function. As a carrier of physical performance, the film thickness uniformity is a prerequisite for ensuring the mechanical stability of the thin film. Film thickness fluctuations in the W and TiN stacked structure will exacerbate stress concentration and may induce interface delamination or wafer warping. Meanwhile, the film thickness deviation of Cu interconnects will change its thermal conductivity distribution and affect the heat dissipation efficiency of the device.

[0004] Film thickness measurement is not only a necessary step in closed-loop control of process parameters, but also a core metrological requirement to ensure the performance consistency and long-term reliability of advanced process devices. The following analyzes the current mainstream film thickness measurement technologies and their characteristics. The main mainstream film thickness measurement technologies include: transmission electron microscopy (TEM), ellipsometry, X-ray reflection, and laser ultrasound. TEM requires wafer cutting, grinding, and ion thinning using a focused ion beam, with a single sample preparation cycle typically exceeding 4 hours. Furthermore, the ion beam and mechanical stress introduced during sample preparation can distort the original stress field by more than 20%. While ellipsometry allows for non-contact measurement, it is less effective when the model is not fully calibrated. Under ideal conditions, the fitting error for film thickness, including multilayer heterostructures, can reach over 15%. This is due to the cross-coupling of optical constants and film thickness parameters. Furthermore, the micron-level spatial resolution (film thickness measurement range > 1 μm) cannot capture the submicron-level stress gradient at the edge of the fin field-effect transistor gate. While X-ray reflection has sub-nanometer thickness sensitivity, it is prone to mixing stress-induced density changes with the actual thickness. For example, when residual stress causes a 2.3% decrease in tungsten layer density, the fitting by X-ray reflection is easily misjudged as a 0.7 nm increase in film thickness. Although fully online laser ultrasound (LUS) supports high-speed scanning, its longitudinal resolution is limited by the acoustic wavelength, and the film thickness to be measured needs to be greater than 100 nm. For example, when measuring an 8 nm cobalt barrier layer, because the film thickness is far below the technical measurement threshold, it cannot effectively resolve the thickness or the error exceeds 100%. Moreover, it cannot penetrate the channel holes of three-dimensional NAND flash memory with an aspect ratio greater than 10:1 to resolve the sidewall thickness distribution.

[0005] In summary, current mainstream film thickness measurement technologies need improvement in both measurement efficiency and accuracy. Summary of the Invention

[0006] In view of this, the present invention aims to provide a film thickness measurement device and method, which at least helps to improve the accuracy of film thickness measurement.

[0007] To achieve the above objectives, the technical solution created by this invention is implemented as follows:

[0008] This invention provides a method for measuring a film layer, comprising: illuminating a test location on the upper surface of a film layer under test with a pump light, wherein the pump light induces a change in the second-order nonlinear optical coefficient of the film layer under test and forms a local disturbance, the local disturbance propagating back and forth between the upper and lower surfaces of the film layer under test; illuminating the test location with a probe light, wherein the probe light generates a frequency-doubled signal light based on a second-order nonlinear optical effect, and the intensity of the frequency-doubled signal light reaches a peak when the local disturbance propagates to the upper surface; receiving the frequency-doubled signal light, and obtaining the thickness of the film layer under test based on the time interval between two adjacent peak intensities of the frequency-doubled signal light.

[0009] Furthermore, the polarization direction of the pump light is parallel to that of the probe light, and there is a time delay between the pump light and the probe light.

[0010] Furthermore, after the probe light illuminates the position to be measured, part of the probe light is reflected by the position to be measured and transmitted together with the frequency-doubled signal light. The part of the probe light transmitted together with the frequency-doubled signal light is the fundamental frequency signal light. Before receiving the frequency-doubled signal light, the fundamental frequency signal light is also filtered out.

[0011] Furthermore, the probe light is phase-matched with the frequency-doubled signal light.

[0012] Furthermore, the wavelength of the pump light is matched with the strong absorption band of the film under test.

[0013] Furthermore, the power density of the pump light is greater than the lattice thermal distortion threshold of the film under test, and the power density of the pump light is less than the material damage threshold of the film under test.

[0014] Furthermore, the ratio of the pump light power density to the probe light power density is in the range of 10 to 100, with the pump light power density being in the range of 10. 4 W / cm 2 ~10 7 W / cm 2 Within the range.

[0015] Furthermore, the diameter of the spot formed by the pump light at the test location is in the range of 5μm to 10μm.

[0016] Furthermore, the polarization direction of the probe light is consistent with the direction of the non-zero component of the second-order nonlinear optical coefficient caused by the pump light.

[0017] Furthermore, obtaining the thickness of the film under test includes: obtaining the time interval Δt between two adjacent intensity peaks of the frequency-doubled signal light, and obtaining the propagation speed of local disturbances within the film under test. The thickness of the film to be measured .

[0018] Another aspect of this invention provides a film measurement device for implementing the aforementioned film measurement method. The film measurement device includes: a light source, which emits pump light and probe light to illuminate the test location on the upper surface of the film to be measured. The pump light induces a change in the second-order nonlinear optical coefficient of the film to be measured and forms a local disturbance. The local disturbance propagates back and forth between the upper and lower surfaces of the film to be measured. The probe light generates a frequency-doubled signal light based on the second-order nonlinear optical effect. A retroreflector is located in the transmission optical path of the probe light or in the transmission optical path of the pump light. The retroreflector moves back and forth to time-delay the probe light or the pump light. A signal receiver is used to receive the frequency-doubled signal light. A signal processing module is used to obtain the thickness of the film to be measured based on the frequency-doubled signal light.

[0019] Furthermore, after the probe light illuminates the position to be measured, part of the probe light is reflected by the position to be measured and transmitted together with the frequency-doubled signal light. The part of the probe light transmitted together with the frequency-doubled signal light is the fundamental frequency signal light. The film measurement device also includes a dichroic mirror, which is used to reflect the frequency-doubled signal light and transmit the fundamental frequency signal light. The incident angle of the mixed beam of the frequency-doubled signal light and the fundamental frequency signal light into the dichroic mirror is 45°.

[0020] Furthermore, the film measurement device also includes a first optical lens, through which pump light is focused onto the position to be measured, and the numerical aperture of the first optical lens is in the range of 0.8 to 1.0.

[0021] Compared with existing technologies, the present invention can achieve the following beneficial effects: The film measurement method provided by the present invention is based on the second-order nonlinear optical effect, which refers to the process by which a material generates a frequency-doubled signal light under the action of a laser. The intensity of the frequency-doubled signal light is determined by the second-order nonlinear optical coefficient of the material. Decision, specific Where I represents the intensity of the frequency-doubled signal light, This represents the electric field amplitude of the probe light, which, when the pump light illuminates the test location, causes a second-order nonlinear optical coefficient in the test film. Changes in second-order nonlinear optical coefficients The local disturbance caused by the change propagates back and forth between the upper and lower surfaces of the film under test. When the local disturbance propagates to the upper surface, the second-order nonlinear optical coefficient at the test location illuminated by the probe light is... A peak is generated so that the intensity of the frequency-doubled signal light generated by the probe light illuminating the test location reaches the peak value. Thus, the thickness of the film to be tested can be obtained based on the time interval between two adjacent peak values ​​of the frequency-doubled signal light.

[0022] Compared to traditional ellipsometers (with a resolution of 1μm~5μm), the film measurement method provided by this invention, with the assistance of structured light illumination or near-field enhancement technologies, can achieve a spatial resolution of approximately 100nm, improving measurement accuracy and enabling the capture of submicron-level stress gradients at the gate edge of fin field-effect transistors. Furthermore, in a single measurement, in addition to providing film thickness information, it can also provide stress distribution characteristics. Combined with laser-ultrasound related modules, it can assist in acquiring material acoustic parameters (sound wave propagation speed, sound wave amplitude attenuation coefficient with propagation distance, and acoustic impedance and interface reflection coefficient). Employing a fully optical non-contact design, single-point measurement speed can reach the 10ms level. Combined with line scanning and compressed sensing algorithms, it can achieve rapid measurement of critical areas of the wafer, significantly improving measurement efficiency compared to traditional methods. Moreover, this invention relies on the ultra-high sensitivity of frequency-doubled signal light to lattice distortion to achieve thickness measurement. Compared to LUS, this invention does not depend on the sound wave wavelength, avoiding the limitation of the measurement range by the sound wave wavelength, and can resolve ultrathin film thicknesses far below the sound wave wavelength, expanding the film thickness measurement range. Attached Figure Description

[0023] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0024] Figure 1 A schematic diagram illustrating the principle of the film measurement method described in the embodiments of the present invention;

[0025] Figure 2 This is a schematic diagram of the structure of the film measurement device described in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0028] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0029] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0030] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] refer to Figure 1 This invention provides a method for measuring a film layer, comprising: illuminating the test location on the upper surface of a film layer 1 to be measured with a pump light 3, wherein the pump light 3 induces a change in the second-order nonlinear optical coefficient of the film layer 1 to be measured and forms a local disturbance, the local disturbance propagating back and forth between the upper and lower surfaces of the film layer 1 to be measured; illuminating the test location with a probe light 2, wherein the probe light 2 generates a frequency-doubled signal light 4 based on the second-order nonlinear optical effect, and the intensity of the frequency-doubled signal light 4 reaches a peak when the local disturbance propagates to the upper surface; receiving the frequency-doubled signal light 4, and obtaining the thickness of the film layer 1 to be measured based on the time interval between two adjacent peak intensities of the frequency-doubled signal light 4.

[0032] In some embodiments, the polarization direction of the pump light 3 is parallel to the polarization direction of the probe light 2, and there is a time delay between the pump light 3 and the probe light 2.

[0033] Specifically, after the pump light 3 irradiates the test film 1, it generates ultrasonic waves by inducing transient thermal expansion. The ultrasonic waves propagate into the interior of the test film 1. When the ultrasonic waves reach the interface between the test film 1 and the substrate (the lower surface of the test film 1), they are reflected and return to the upper surface of the test film 1. The local lattice distortion caused by the ultrasonic waves modulates the second-order nonlinear optical coefficient of the material. When probe light 2 illuminates the position to be measured, the second-order nonlinear optical coefficient... When the disturbance propagates to the upper surface, the intensity of the generated frequency-doubled signal light 4 will change. By precisely controlling the time delay τ of the probe light 2 relative to the pump light 3 and measuring the evolution curve of the frequency-doubled signal light 4 with time delay τ, the film thickness can be demodulated and calculated.

[0034] In some embodiments, the wavelength of pump light 3 matches the strong absorption band of the film layer 1 under test. This is because: the core function of pump light 3 is to actively control the microstructure of the material, inducing symmetry breaking and optical constant changes at the center of the heated lattice, thus creating a prerequisite for second-order nonlinear optical effects. The parameter design of pump light 3 aims for "efficient and localized heating." Therefore, the wavelength of pump light 3 needs to be located in the strong absorption band of the film layer 1 under test. For example, for semiconductor thin films such as Si and TiN, or metal thin films, light with a wavelength of 808 nm or 1064 nm in the near-infrared band can be used as pump light 3 to ensure that the energy of pump light 3 is absorbed by the lattice rather than penetrated. The strong absorption band of the film layer under test corresponds to the intraband transitions of electrons within the metal and the interband absorption of photons within the semiconductor material, and it is located in the film layer under test where the absorption coefficient α > 10. 4 cm -1 The corresponding spectral range.

[0035] In some embodiments, the power density of the pump light 3 is greater than the lattice thermal distortion threshold of the film layer 1 under test, and less than the material damage threshold of the film layer 1 under test. The power density of the pump light 3 must reach the lattice thermal distortion threshold of the material of the film layer 1 under test to ensure that the second-order nonlinear optical coefficient of the material of the film layer 1 under test is induced. The power density of the pump light 3 needs to be lower than the damage threshold of the test film 1 material in order to avoid melting or irreversible lattice damage to the test film 1 material.

[0036] In some cases, the lattice thermal distortion threshold of the material of the test film layer 1 is typically around 10. 4 W / cm²~10 7 Within the range of W / cm², the lattice thermal distortion threshold can be adjusted according to the thermal conductivity of the material.

[0037] In some embodiments, the pulse width of the pump light 3 is on the order of femtoseconds (fs), picoseconds (ps), or nanoseconds (ns), and further on the order of picoseconds or femtoseconds.

[0038] In some embodiments, the ratio of the power density of the pump light 3 to the power density of the probe light 2 is in the range of 10 to 100, and the power density of the pump light 3 is in the range of 10. 4 W / cm 2 ~10 7 W / cm 2 Within the range.

[0039] In some embodiments, the test film 1 is a semiconductor thin film, such as Si or GaAs, and the power density of the pump light 3 is typically controlled at 10. 4 W / cm 2 ~10 6 W / cm 2 Within this range, semiconductor thin films have moderate thermal conductivity and low melting points; for example, the thermal conductivity of Si is approximately 150 W / (m²). The melting point of Si is about 1414℃. Too low a power density cannot induce significant lattice distortion, while too high a power density can easily lead to localized melting.

[0040] In some embodiments, the film layer 1 under test is a metal thin film, such as TiN or W, and the power density of the pump light 3 is typically controlled at 10. 4 W / cm 2 ~10 6 W / cm 2 Within this range, metals have high thermal conductivity and high melting point; for example, the thermal conductivity of W is approximately 174 W / (m²). K), with a melting point of approximately 3422℃, allows the crystal lattice to rapidly accumulate heat, forming directional stretching or compression distortion, and is less likely to exceed the damage threshold.

[0041] In some embodiments, the test film 1 is a multilayer heterostructure, for example, a heterostructure composed of TiN, W, and SiO2 stacked sequentially. The power density of the pump light 3 needs to be taken as the median value covering the target film, i.e., within 10. 5 W / cm²~10 6 The W / cm² range is selected to ensure that both the top TiN layer and the intermediate W layer can be distorted, avoiding incompleteness caused by heating only the surface layer.

[0042] In some embodiments, the diameter of the spot formed by the pump light 3 at the test location is in the range of 5 μm to 10 μm. By reducing the spot size, the power density can be increased. In some embodiments, the pump light 3 can be focused into a micrometer-scale spot using a high numerical aperture objective lens, i.e., the spot diameter is in the range of 5 μm to 10 μm, so that the power density is concentrated in a specific area. In this way, on the one hand, the "large-area weak distortion" caused by power dispersion can be avoided, ensuring that the broken area and the detection area of ​​the subsequent probe light 2 are accurately coincided; on the other hand, the localized high power density can make the lattice distortion more oriented, for example, the stretching or compression along the stress direction is more obvious, further enhancing the effect of central inversion symmetry breaking.

[0043] Since the core function of probe light 2 is to passively detect changes in the second-order nonlinear optical coefficient to form local perturbations, within the "symmetry-breaking environment" created by pump light 3, it interacts with the material's second-order nonlinear optical coefficient χ. (2) The process generates frequency-doubled signal light 4. Since the probe light 2 itself does not participate in lattice heating, its parameter design needs to aim for "weak perturbation and high matching degree". The wavelength needs to take into account both "frequency-doubled signal detectability" and "material transparency". For "frequency-doubled signal detectability", for example, when the wavelength of probe light 2 is 800nm ​​and the wavelength of frequency-doubled signal light 4 is 400nm, a silicon-based detector can be used to collect the frequency-doubled signal light 4. For "material transparency", for example, when the material of the film layer 1 to be tested is SiO2, the wavelength of probe light 2 needs to be selected as 1064nm and the wavelength of frequency-doubled signal light 4 needs to be 532nm to ensure that the frequency-doubled signal light 4 can penetrate the SiO2 film and reduce signal loss.

[0044] Furthermore, the power density of probe light 2 needs to be much lower than that of pump light 3; the power density of probe light 2 is typically between 10³ W / cm² and 10³ W / cm². 5 Within the range of W / cm², to avoid additional thermal distortion induced by probe light 2.

[0045] In some embodiments, the incident timing of the probe light 2 is matched with the lattice distortion dynamics of the pump light 3 (i.e., timing synchronization), and the pulse width of the probe light 2 is not greater than the stable window duration of the lattice distortion, so as to provide sufficient time resolution; for example, the probe light 2 is delayed by about 100 picoseconds after the pump light 3 pulse ends, which can ensure that the probe light and the distortion region are precisely overlapped in space and that the signal acquisition is completely within the stable window in time, thereby improving the detection signal-to-noise ratio.

[0046] In some embodiments, the polarization direction of the probe light 2 is aligned with the direction of the non-zero component of the second-order nonlinear optical coefficient induced by the pump light 3, to ensure that the polarization state of the probe light 2 is precisely aligned with the non-zero component of the second-order nonlinear optical coefficient. Specifically, since the intensity of the frequency-doubled signal light 4 is related to the angle between the polarization direction of the probe light and the non-zero component of the second-order nonlinear optical coefficient, the smaller the angle between the polarization direction of the probe light and the non-zero component of the second-order nonlinear optical coefficient, the stronger the frequency-doubled signal light 4. When the non-zero components of the coefficients are aligned (for example, pump light 3 induces lattice distortion along the x-direction, the second-order nonlinear optical coefficient has a non-zero component in the x-direction, and the polarization direction of probe light 2 is also along the x-direction), the nonlinear coupling is strongest, and the intensity of the frequency-doubled signal light 4 can reach its maximum value; if the angle between the polarization direction of probe light 2 and the non-zero component direction of the second-order nonlinear optical coefficient is 90°, almost no frequency-doubled signal light 4 will be generated. In practical applications, the polarization direction of probe light 2 can be adjusted by rotating the polarizer, and then combined with the generated frequency-doubled signal light 4, the optimal polarization direction can be determined.

[0047] In some embodiments, after the probe light 2 illuminates the target location, a portion of the probe light is reflected by the target location and transmitted together with the frequency-doubled signal light 4. The portion of the probe light transmitted together with the frequency-doubled signal light 4 is the fundamental frequency signal light. Figure 1 (Not shown), before receiving the frequency-doubled signal light 4, it also includes: filtering out the base frequency signal light.

[0048] This is because when the pump light 3 heats the lattice and the probe light 2 excites the frequency-doubled signal light 4, the collected reflected light is not a single component. It contains both the probe light (fundamental frequency signal light) that did not participate in the reaction and the frequency-doubled signal light 4. Since the frequency-doubled signal light 4 is weaker and the fundamental frequency signal light is stronger, when the frequency-doubled signal light 4 and the fundamental frequency signal light are mixed together, the stronger fundamental frequency signal light will directly mask the weaker frequency-doubled signal light 4, making it impossible to accurately detect the frequency-doubled signal light 4 in the future. Therefore, it is necessary to filter out the fundamental frequency signal light.

[0049] In some embodiments, the probe light 2 and the frequency-doubled signal light 4 satisfy the phase-matching condition. In other words, the fundamental frequency signal light and the frequency-doubled signal light 4 satisfy the phase-matching condition, meaning their wave vectors are matched (Δk=0), ensuring that the phase difference remains constant and there is no cumulative mismatch during propagation. This design effectively avoids the intensity attenuation of the frequency-doubled signal light due to phase interference cancellation, significantly improving frequency doubling efficiency and signal stability. Specifically, the phase matching can be achieved by adjusting the incident angle of the probe light: utilizing the birefringence properties of the nonlinear medium, the wave vector directions of the probe light and the frequency-doubled signal light are changed, ensuring that the two beams satisfy momentum conservation, 2k1=k2, where k1 is the wave vector of the probe light and k2 is the wave vector of the frequency-doubled signal light. For example, when measuring a 100nm thick SiO2 film, when photothermal induction breaks its symmetry, adjusting the incident angle of the probe light 2 to the range of 30°~45° ensures that the phase mismatch coefficient Δk ≤ 10 between the probe light and the frequency-doubled signal light. 3 m -1 The effect is best when the intensity of the frequency-doubled signal light reaches its peak (frequency doubling efficiency ≥ 30%).

[0050] In some embodiments, obtaining the thickness of the film layer 1 under test includes: obtaining the time interval Δt between two adjacent light intensity peaks of the frequency-doubled signal light 4, and obtaining the propagation speed of local disturbances within the film layer 1 under test. The thickness of the film layer 1 to be tested Among them, the propagation speed of local disturbances within the test film layer 1. It is equal to the speed of sound.

[0051] The film measurement method provided by this invention is applicable to non-destructive online measurement of wafer-level metal interconnect layers, dielectric films, and compound semiconductor devices.

[0052] refer to Figure 1 and Figure 2 In another aspect, this invention provides a film measurement device for implementing the aforementioned film measurement method. The film measurement device includes: a light source, which emits pump light 3 and probe light 2 to illuminate the test position on the upper surface of the film 1 to be measured. The pump light 3 induces a change in the second-order nonlinear optical coefficient of the film 1 to be measured and forms a local disturbance. The local disturbance propagates back and forth between the upper and lower surfaces of the film 1 to be measured. The probe light 2 generates a frequency-doubled signal light 4 based on the second-order nonlinear optical effect; a retroreflector 13, which is located in the transmission optical path of the probe light 2 or in the transmission optical path of the pump light 3. The retroreflector 13 moves back and forth to time-delay the probe light 2 or the pump light 3; a signal receiver, which receives the frequency-doubled signal light 4; and a signal processing module 104, which obtains the thickness of the film 1 to be measured based on the frequency-doubled signal light 4.

[0053] In some embodiments, the light source includes a laser 10 and a beam splitter 11. The laser 10 can output laser pulses with a pulse width of femtosecond, picosecond, or nanosecond. The laser pulses are split into two beams with orthogonal polarization by the beam splitter 11. The transmitted light can be used as the probe light 2, and the reflected light can be used as the pump light 3.

[0054] In some embodiments, a high-precision linear electric displacement stage 12 is used to drive the retroreflector 13 to move back and forth, thereby creating a time delay between the pump light and the probe light. In other embodiments, other delay methods may also be used.

[0055] In some embodiments, the retroreflector 13 is located in the transmission optical path of the probe light 2. The probe light 2 enters the retroreflector 13, and the probe light 2 emitted from the retroreflector 13 is adjusted in transmission direction by a reflector and then focused on the test position on the sample 105 by the second optical lens 14.

[0056] In some embodiments, pump light 3 is reflected by beam splitter 11 and incident on modulator 101. The modulation frequency of modulator 101 is provided by signal generator 102 and synchronized to the reference signal input of lock-in amplifier 103. After being reflected by electrically adjustable mirror 17, pump light 3 is incident on the test position on sample 105. In some examples, sample 105 can be a wafer. In some embodiments, modulator 101 can be any one of acousto-optic modulator, electro-optic modulator, or magneto-optic modulator to achieve intensity or phase modulation. The attitude of electrically adjustable mirror 17 can be adjusted by piezoelectric tilting stage.

[0057] In some embodiments, the film measurement device further includes a first optical lens 15, through which the pump light 3 is focused onto the position to be measured. The numerical aperture of the first optical lens 15 is in the range of 0.8 to 1.0, so as to ensure that the diameter of the light spot formed by the pump light 3 at the position to be measured is in the range of 5 μm to 10 μm.

[0058] In some embodiments, after the probe light 2 illuminates the position to be measured, a portion of the probe light 2 is reflected by the position to be measured and transmitted together with the frequency-doubled signal light 4. The portion of the probe light transmitted together with the frequency-doubled signal light 4 is the fundamental frequency signal light 5. The film measurement device also includes a dichroic mirror 18, which is used to reflect the frequency-doubled signal light 4 and transmit the fundamental frequency signal light 5. The incident angle of the mixed beam of the frequency-doubled signal light 4 and the fundamental frequency signal light 5 into the dichroic mirror 18 is 45°.

[0059] The surface of the dichroic mirror 18 is coated with multiple dielectric films. The thickness and material of the dielectric films are designed based on the wavelength difference between the frequency-doubled signal light 4 and the fundamental frequency signal light 5 to ensure that the dichroic mirror 18 has high reflectivity for the frequency-doubled signal light 4 and high transmittance for the fundamental frequency signal light 5. Taking the 1064nm pump light 3 and 532nm frequency-doubled signal light 4 commonly used in semiconductor detection as an example, a dichroic mirror 18 with high reflectivity at 532nm and high transmittance at 1064nm should be selected. When the mixed light containing the 1064nm fundamental frequency signal light 5 and the 532nm frequency-doubled signal light 4 is incident on the dichroic mirror 18, the 532nm frequency-doubled signal light 4 will be reflected to the frequency-doubled signal light collection path, while the 1064nm fundamental frequency signal light 5 will pass directly through the lens surface and be processed by the subsequent fundamental frequency signal light 5 absorber, thus preventing the fundamental frequency signal light 5 from entering the collection end and generating noise.

[0060] The incident angle of the mixed beam of frequency-doubled signal light 4 and fundamental frequency signal light 5 onto the dichroic mirror 18 is 45°. This is to ensure that the optical path of the reflected frequency-doubled signal light 4 is perpendicular to the optical path of the original fundamental frequency signal light 5 (probe light 2), so that the frequency-doubled signal light 4 can be directly guided to the subsequent collection end without complex optical path detours, reducing optical loss and meeting the compact optical path requirements in semiconductor detection. In addition, if the incident angle of the mixed beam of frequency-doubled signal light 4 and fundamental frequency signal light 5 onto the dichroic mirror 18 deviates from 45°, some of the fundamental frequency signal light 5 may be reflected into the collection end, increasing noise. Therefore, the 45° angle ensures that the fundamental frequency signal light 5 passes through the dichroic mirror 18 with minimal loss, avoiding increased noise at the collection end.

[0061] In some embodiments, the mixed beam of the frequency-doubled signal light 4 and the fundamental frequency signal light 5 passes through the third optical lens 16 and then enters the dichroic mirror 18.

[0062] It should be noted that after the pump light 3 is incident on the test position, the energy of the pump light 3 is absorbed by the electrons in the test film 1, which excites the electrons to jump from a low energy state to a high energy state, forming hyperthermal electrons with energy much higher than the thermal equilibrium state. These hyperthermal electrons exchange energy with the surrounding electrons rapidly through Coulomb interaction, and at the same time transfer energy to the lattice through electron-phonon coupling, which eventually causes the lattice to heat up and reach a new thermal equilibrium. During the lattice heating process, its structure is deformed, that is, the lattice constant changes, which leads to a change in the second-order nonlinear optical coefficient. At this time, the probe light 2 incident on the same position will generate a frequency-doubled signal light 4 (purple dashed line) through the second-order nonlinear optical effect. The frequency-doubled signal light 4 is collimated by the third optical lens 16 and then guided out by the reflector. The dichroic mirror 18 filters out the fundamental frequency signal light 5 and reflects the frequency-doubled signal light 4 to the signal receiver.

[0063] In some embodiments, the signal receiver includes an optical probe 100 and a lock-in amplifier 103. The optical probe 100 converts the frequency-doubled signal light 4 into an electrical signal and inputs it into the input terminal of the lock-in amplifier 103. The lock-in amplifier 103 extracts the target signal based on the reference frequency to filter out noise frequencies and improve the signal-to-noise ratio. The lock-in amplifier 103 transmits the target signal to the signal processing module 104.

[0064] This invention achieves a breakthrough in semiconductor film thickness measurement by utilizing the temporal characteristic peak localization mechanism of the frequency-doubled signal light 4, without requiring complex derivation of stress-related formulas. In terms of measurement efficiency, traditional ellipsometers rely on complex optical modeling, while this invention only needs to extract the position of the characteristic peak in the frequency-doubled signal light 4 curve to complete single-point measurement, significantly improving efficiency and eliminating the need for model fitting and equipment switching processes. In terms of measurement reliability, this invention directly calculates the thickness based on the physical laws of local perturbation propagation, avoiding the modeling errors that may exist in ellipsometers and the measurement aliasing problem caused by stress and density coupling in X-ray reflection methods, further reducing the standard deviation of film thickness measurement and improving the accuracy of film thickness measurement.

[0065] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0066] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for measuring film layers, characterized in that, include: The test position on the upper surface of the test film is irradiated with pump light. The pump light induces a change in the second-order nonlinear optical coefficient of the test film and forms a local disturbance. The local disturbance propagates back and forth between the upper and lower surfaces of the test film. The power density of the pump light is greater than the lattice thermal distortion threshold of the test film and less than the material damage threshold of the test film. The test position is illuminated by a probe light, which generates a frequency-doubled signal light based on a second-order nonlinear optical effect. When the local disturbance propagates to the upper surface, the second-order nonlinear optical coefficient of the test position illuminated by the probe light generates a peak value, so that the light intensity of the frequency-doubled signal light generated by the probe light illuminating the test position reaches the peak value. The thickness of the film to be tested is obtained based on the time interval between two adjacent light intensity peaks of the frequency-doubled signal light.

2. The film measurement method according to claim 1, characterized in that, The polarization direction of the pump light is parallel to that of the probe light, and there is a time delay between the pump light and the probe light.

3. The film measurement method according to claim 1, characterized in that, After the probe light illuminates the position to be measured, part of the probe light is reflected by the position to be measured and transmitted together with the frequency-doubled signal light. The part of the probe light transmitted together with the frequency-doubled signal light is the fundamental frequency signal light. Before receiving the frequency-doubled signal light, the method further includes filtering out the fundamental frequency signal light.

4. The film measurement method according to claim 1, characterized in that, The probe light is phase-matched with the frequency-doubled signal light.

5. The film measurement method according to claim 1, characterized in that, The wavelength of the pump light matches the strong absorption band of the film under test.

6. The film measurement method according to claim 1, characterized in that, The ratio of the power density of the pump light to the power density of the probe light is in the range of 10 to 100, and the power density of the pump light is in the range of 10. 4 W / cm 2 ~10 7 W / cm 2 Within the range.

7. The film measurement method according to claim 1, characterized in that, The diameter of the spot formed by the pump light at the measured position is in the range of 5μm to 10μm.

8. The film measurement method according to claim 1, characterized in that, The polarization direction of the probe light is consistent with the direction of the non-zero component of the second-order nonlinear optical coefficient caused by the pump light.

9. The film measurement method according to claim 1, characterized in that, Obtaining the thickness of the film under test includes: obtaining the time interval Δt between two adjacent intensity peaks of the frequency-doubled signal light, and obtaining the propagation speed of the local disturbance within the film under test. The thickness of the film to be tested .

10. A film layer measuring device, characterized in that, For implementing the film measurement method according to any one of claims 1 to 9, the film measurement device comprises: The light source is used to emit pump light and probe light to illuminate the test position on the upper surface of the film layer under test. The pump light induces a change in the second-order nonlinear optical coefficient of the film layer under test and forms a local disturbance. The local disturbance propagates back and forth between the upper and lower surfaces of the film layer under test. The probe light generates a frequency-doubled signal light based on the second-order nonlinear optical effect. A retroreflector, located in the transmission optical path of the probe light or in the transmission optical path of the pump light, the retroreflector moving back and forth to time delay the probe light or the pump light; A signal receiver, wherein the signal receiver is used to receive frequency-doubled signal light; A signal processing module is used to obtain the thickness of the film to be tested based on the frequency-doubled signal light.

11. The film layer measuring device according to claim 10, characterized in that, After the probe light illuminates the position to be measured, a portion of the probe light is reflected by the position to be measured and transmitted together with the frequency-doubled signal light. The portion of the probe light transmitted together with the frequency-doubled signal light is the fundamental frequency signal light. The film measurement device also includes a dichroic mirror, which is used to reflect the frequency-doubled signal light and transmit the fundamental frequency signal light. The incident angle of the mixed beam of the frequency-doubled signal light and the fundamental frequency signal light into the dichroic mirror is 45°.

12. The film layer measuring device according to claim 10, characterized in that, The film measurement device further includes a first optical lens, through which the pump light is focused onto the position to be measured, and the numerical aperture of the first optical lens is in the range of 0.8 to 1.0.

Images