Online in-situ characterization method of chemical mechanical polishing through silicon via heterostructure

Abstract

The invention discloses an online in-situ characterization method of a chemical mechanical polishing through silicon via heterostructure, and belongs to the technical field of chip manufacturing. According to the TSV heterostructure sample, through the remarkable difference of different components of a TSV heterostructure sample matrix, an insulating layer, a barrier layer and a transmission layer in refractive index, work function and conductivity, a PiFM is taken as a core, a KPFM and a CAFM are combined, and the TSV heterostructure sample can be detected on the same platform by detecting light induction force, surface potential and current signals at a needle tip-sample position. And the real-time in-situ measurement of the TSV sample surface appearance height, roughness and interface component change is realized. According to the method, the light, electricity and force response differences between TSV heterostructure materials are fully utilized, dynamic feedback of the structure morphology and components in the CMP process is achieved, and support is provided for nano-scale and even atomic-scale polishing precision control.

Description

Technical Field

[0001] This invention belongs to the field of chip manufacturing technology, specifically relating to an online in-situ characterization method for chemically mechanically polished silicon through-hole heterostructures. Background Technology

[0002] With the increasing demands on the performance of core functional components in high-end equipment fields such as integrated circuits, aerospace, and defense, chip manufacturing precision and defect control technologies are continuously evolving towards the atomic scale. In advanced chip manufacturing, through-silicon vias (TSVs) and nano-through-silicon vias (n-TSVs), as key structures forming vertical interconnect channels through silicon wafers, are currently the only mature technology path to achieve vertical electrical interconnects between chips. For typical silicon / copper heterostructure TSV systems, surface roughness needs to be controlled to the atomic level and near-zero defects, posing a significant challenge to existing chemical mechanical polishing (CMP) technologies. Simultaneously, high-precision characterization of the TSV surface after polishing is crucial for guiding the optimization of material deposition and polishing processes. However, currently, there is a lack of systematic methods to achieve online in-situ characterization during polishing, making it difficult to obtain real-time information on the surface morphology, roughness, and compositional changes of TSV heterostructures.

[0003] For the detection and characterization of TSV heterostructures, existing research mainly focuses on the following three technical approaches:

[0004] (1) Optical methods

[0005] Traditional optical microscopes and white-light interferometers are limited by the Abbe diffraction limit, with spatial resolution typically on the order of hundreds of nanometers. They can only characterize the surface morphology of micron-sized through-silicon vias (TSVs) and cannot achieve morphological resolution for n-TSVs. Scanning near-field optical microscopy (SNOM) and tip-enhanced Raman spectroscopy (TERS) can overcome the diffraction limit by utilizing near-field scattering effects, achieving simultaneous nanometer-resolution characterization of morphology and composition. However, such systems typically include complex near-field optical coupling modules, pseudo-heterodyne interferometer modules, and high-end spectroscopic detection components (CCD and spectrometer), resulting in high equipment costs and extremely high system stability requirements. Furthermore, the signal light is susceptible to environmental scattering interference, limiting the signal-to-noise ratio and sample contrast. Additionally, specially designed metallized tips are required to improve coupling efficiency, all of which restrict their application in in-situ detection of TSV heterostructures.

[0006] (2) Electronic methods

[0007] Electronic characterization techniques primarily include scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HR-TEM). Both can achieve spatial resolution at the nanometer to sub-nanometer level and possess elemental and chemical composition analysis capabilities, making them core methods for studying the microstructure and interface characteristics of TSV structures. However, both SEM and TEM require high-vacuum conditions and struggle to achieve in-situ monitoring of sample height, roughness, and polishing thickness. Furthermore, for TSV heterostructures containing insulating layers, electron beam irradiation easily induces surface charge accumulation, leading to image distortion or failure to form an image. Gold sputtering or carbon film conductive treatment is typically used to eliminate the charging effect, but this process causes irreversible damage to the sample surface. In particular, TEM testing requires precise sample preparation via focused ion beam (FIB), resulting in long preparation cycles and high costs. It may also introduce ion bombardment damage and structural defects into the sample surface, affecting the accuracy of subsequent interface precision and compositional analysis. Therefore, while traditional electronic detection methods offer high resolution advantages, they still have significant limitations in the dynamic characterization and non-destructive monitoring of the CMP process in TSV heterostructures.

[0008] (3) Mechanical methods

[0009] Mechanical probe techniques include scanning tunneling microscopy (STM) and atomic force microscopy (AFM). STM can achieve atomic resolution imaging at the 0.1 nm level, but the object being detected must have good conductivity, limiting its application in heterogeneous systems. AFM, on the other hand, is not limited by the conductivity of the sample and can accurately acquire surface morphology, but traditional AFM cannot simultaneously resolve the compositional characteristics of the material. Summary of the Invention

[0010] To address the aforementioned technological limitations, this invention provides an online in-situ characterization method for chemically mechanically polished (CMP) through-silicon via (TSV) heterostructures. By leveraging the significant differences in refractive index, work function, and conductivity among the different components of the substrate, insulating layer, barrier layer, and transport layer of the TSV heterostructure sample, a real-time in-situ measurement method is achieved using photoinduced force microscopy (PiFM) as the core, coupled with scanning Kelvin probe microscopy (KPFM) and conductive atomic force microscopy (CAFM). This method detects the photoinduced force, surface potential, and current signals at the tip-sample interface on the same platform, enabling the real-time in-situ measurement of the surface morphology, roughness, and interfacial composition changes of the TSV sample.

[0011] This invention fully utilizes the differences in optical, electrical, and mechanical responses between TSV heterostructure materials to achieve dynamic feedback on structural morphology and composition during CMP, providing support for precision control at the nanometer and even atomic levels. This method is not only applicable to the quality assessment of through-silicon vias (TSVs) and nano-TSVs, but also provides a novel characterization method for precise control of heterostructure interfaces and defect detection in high-end chip manufacturing processes.

[0012] To achieve the above objectives, the present invention provides the following technical solution:

[0013] One of the technical solutions of this invention is to provide an online in-situ characterization method for chemically mechanically polished silicon through-hole heterostructures, comprising the following steps:

[0014] Using photoinduced force microscopy (PiFM), scanning Kelvin probe microscopy (KPFM), and conductive atomic force microscopy (CAFM), online in-situ characterization of through-silicon via (TSV) heterostructures or nano-TSV heterostructures during chemical mechanical polishing (CMP) was performed. Real-time information on surface morphology height, photomechanical amplitude, photomechanical phase, surface potential, and probe current was obtained. The obtained information was analyzed to achieve in-situ dynamic non-destructive testing of surface morphology height, roughness, and compositional changes of TSV heterostructures or nano-TSV heterostructures during CMP.

[0015] The schematic diagrams of PiFM (a), KPFM (b), and CAFM (c) characterizations of the heterostructure of through-silicon vias during chemical mechanical polishing are shown below. Figure 1 .

[0016] This invention integrates PiFM, KPFM, and CAFM modes on a single platform, establishing an in-situ dynamic characterization system for the coordinated detection of optical, electrical, and mechanical signals. By scanning TSV heterostructure samples under different polishing conditions in an atmospheric environment, this invention not only obtains information on sample morphology and height but also utilizes the differences in refractive index, work function, and conductivity among the different materials (Si, Si3N4, SiO2, TiN, Co) of the matrix, insulating layer, barrier layer, and transport layer in the TSV heterostructure sample. Material group differentiation is achieved through signals of photoinduced force (refractive index, local optical polarization response), surface potential (sample work function), and probe current (sample conductivity, current carrying capacity).

[0017] Preferably, the optically induced force microscope is the visible light optically induced force microscope (PiFM) disclosed in the invention patent with publication number CN120703413A.

[0018] Compared to conventional optically induced force microscopy (OIF), visible light optically induced force microscopy (VIF) directly modulates the laser using a signal generator within a lock-in amplifier. This avoids the environmental sensitivity and phase drift issues associated with using acousto-optic modulators (AOMs) or choppers during modulation. The laser excitation frequency is directly controlled by the lock-in amplifier, ensuring that the excitation and demodulation signals remain synchronized. This prevents the accumulation of phase drift, significantly improving the overall system's phase stability. This results in stable and accurate optical force signals, achieving high signal-to-noise ratio measurements with a laser power as low as 2 mW. The low excitation power also eliminates the damage to the sample caused by laser thermal effects.

[0019] Compared to conventional optically induced force microscopy, visible light optically induced force microscopy avoids the need for optical parametric oscillating lasers (OPO) or quantum cascade lasers (QCL) as laser sources, as well as the requirement for laser modulation using choppers or AOMs. It directly uses semiconductor lasers to modulate the laser. With silicon tips or metal-coated tips, it can detect optical force signals in the atmosphere, greatly reducing the complexity and cost of the system and simplifying operation.

[0020] Compared to conventional optical inductive force microscopes, visible light optical inductive force microscopes use lenses for focusing instead of off-axis parabolic mirrors for focusing lasers, thus reducing the cost of the optical system.

[0021] Visible light optical induction force microscopy eliminates the need for objective focusing compared to conventional optical induction force microscopy, significantly reducing the cost of the optical system.

[0022] Visible light optical induction force microscopy offers greater versatility than conventional optical induction force microscopy, as it allows for a wider variety of substrates and eliminates the need for transparency.

[0023] Optionally, when examining samples with through-silicon via heterostructures, the visible light optically induced force microscope uses amplitude modulation mode with the following parameters: amplitude 27.06 nm, probe excitation frequency f1 = 69.6158 kHz, and laser modulation frequency f... 调制 =361.6103kHz, optically induced demodulation frequency f 解调 =431.2261kHz, laser power 1.6mW.

[0024] Optionally, when examining nanocrystalline silicon through-hole heterostructure samples, the visible light optically induced force microscope uses amplitude modulation mode with the following parameters: amplitude 27.06 nm, probe excitation frequency f1 = 72.0612 kHz, and laser modulation frequency f... 调制 =372.9967kHz, optically induced demodulation frequency f 解调 =445.0579kHz, laser power 1.6mW.

[0025] Optionally, when examining samples with through-silicon via heterostructures, the scanning Kelvin probe microscope is used in amplitude modulation mode with the following parameters: the probe tip is 50 nm away from the sample, and a 0.5 V AC voltage is applied.

[0026] Optionally, when examining nano-silicon through-hole heterostructure samples, the conductive atomic force microscope is used in contact mode with the following parameters: load 80 nN and 1 V voltage applied to the tip.

[0027] The second technical solution of the present invention provides an optimized chemical mechanical polishing method based on the above-mentioned online in-situ characterization method for through-silicon via heterostructures, comprising the following steps: real-time information on the surface morphology height, roughness, and composition changes of through-silicon via heterostructures or nano-through-silicon via heterostructures obtained by the above-mentioned online in-situ characterization method for through-silicon via heterostructures, combined with preset polishing process parameters and target surface accuracy, and real-time adjustment of polishing process parameters to achieve the target surface accuracy.

[0028] The beneficial technical effects of the present invention are as follows:

[0029] This invention enables online in-situ characterization of the CMP process for TSV and n-TSV heterostructure samples. Based on the feedback obtained, operators can adjust polishing parameters in real time, thereby achieving controllable polishing and sub-nanometer and atomic-level surface smoothing of different materials. It also allows for quantitative analysis of material removal rate, interface residue, and polishing uniformity, providing reliable support for atomic-level smoothing and precise interface control.

[0030] The technical system of this invention is not only applicable to through-silicon via structures, but can also be extended to fields such as three-dimensional integrated circuit interconnects, MEMS devices, and nanocomposite films.

[0031] Compared to optical microscopes, this invention overcomes the optical diffraction limit, achieving nanometer-level lateral resolution in detecting sample morphology and composition.

[0032] Compared to SNOM and TERS, the PiFM system module of this invention employs a modulated laser-induced tip-sample photo-induced force excitation cantilever, which offers a higher signal-to-noise ratio compared to scattered light signals, is less affected by background light signals, and can detect weak photo-force signals (on the order of μV). Furthermore, it eliminates the need for complex near-field optical path coupling modules, pseudo-heterodyne interferometer modules, and high-end spectral detection components (CCD and spectrometer), reducing equipment costs. In addition, all instrument modes can be used with ordinary commercial metal tips, eliminating the need for specialized tips.

[0033] Compared to SEM, TEM, and STM, this invention can obtain sample morphology and material composition information under atmospheric conditions, with sub-nanometer resolution in the z-direction. It can also image insulating samples without requiring external processing that could damage the sample surface.

[0034] This invention employs in-situ detection, avoiding sample damage introduced by FIB cutting. The detection operation is simple, low-cost, and does not affect subsequent polishing, morphology, and composition analysis of the sample.

[0035] Compared to traditional AFM, this invention can simultaneously analyze the compositional characteristics of materials. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram illustrating the PiFM (a), KPFM (b), and CAFM (c) characterization of heterostructures with through-silicon vias during chemical mechanical polishing (CMP) according to the present invention.

[0038] Figure 2 Images of TSV-A sample (a) and TSV-B sample (b) from Example 1 under an optical microscope.

[0039] Figure 3 This is a SEM image of the cross-sectional profile of the TSV-A sample in Example 1.

[0040] Figure 4 The images show an optical microscope image (a) of the etched area of ​​the n-TSV sample in Example 1, an EDS elemental analysis result of the cross-section of the n-TSV sample (b), an AFM characterization image (c) of the etched area of ​​the n-TSV sample with a scanning range of 20 μm × 20 μm, and an AFM characterization image (d) with a scanning range of 5 μm × 5 μm.

[0041] Figure 5 The images are (a) to (e) and (f) to (j) of the polished TSV-A samples PiFM 1 to 5 in Example 2.

[0042] Figure 6 The optical amplitude diagrams (a) to (e) and the optical phase diagrams (f) to (j) of the polished TSV-A samples PiFM 1 to 5 in Example 2 are shown.

[0043] Figure 7The images show the morphology (a) to (e) and surface potential (f) to (j) of polished TSV-B samples 1 to 5 in Example 2.

[0044] Figure 8 The images are (a) to (e) and (f) to (j) of the polished TSV-B samples PiFM 1 to 5 in Example 2.

[0045] Figure 9 The optical amplitude diagrams (a) to (e) and optical phase diagrams (f) to (j) of the polished TSV-B samples PiFM 1 to 5 in Example 2 are shown.

[0046] Figure 10 The images show the morphology (a) to (e) and surface potential (f) to (j) of polished TSV-A samples 1 to 5 in Example 2.

[0047] Figure 11 The images (a) to (b) and the probe current diagrams (c) to (d) show the morphology of the CAFM etched region of polished n-TSV samples 1 and 2 in Example 2.

[0048] Figure 12 The images show the photomechanical amplitude (a) to (b) and photomechanical phase (c) to (d) of the PiFM etching region of polished n-TSV samples 1-2 in Example 2. Detailed Implementation

[0049] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention. It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the present invention.

[0050] It should be noted that any aspects not described in detail in this invention are conventional practices in the field and are not the focus of this invention.

[0051] Furthermore, regarding the numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Any stated value or intermediate value within a stated range, as well as each smaller range between any other stated value or intermediate value within said range, are also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0052] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods and materials have been described herein, any methods and materials similar to or equivalent to those described herein may be used in the implementation or testing of this invention.

[0053] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0054] Example 1

[0055] Preparation of TSV heterostructure samples: Holes with a diameter of 3μm and a depth of 150nm were etched on a silicon wafer; then Si3N4 was deposited on the sample surface using the ALD process with a thickness of 50nm; two TSV heterostructure samples, TSV-A and TSV-B, were prepared.

[0056] Images of TSV-A sample (a) and TSV-B sample (b) under an optical microscope are shown below. Figure 2 The cross-sectional profile of the TSV-A sample was characterized using SEM, and the results are shown in [Figure number missing]. Figure 3 . Figure 3 The effectiveness of the deposition process was confirmed, and silicon nitride was successfully deposited both inside and outside the pores of the TSV-A.

[0057] Preparation of n-TSV heterostructure samples: Holes with a diameter of 10 nm and a depth of 350 nm were etched in a 10 μm × 10 μm region on a silicon wafer; then, SiO2 (thickness of about 10 nm), TiN (thickness of about 11 nm) and Co (the holes were not completely filled after deposition) were deposited sequentially in the holes using the ALD process to obtain n-TSV heterostructure samples, abbreviated as n-TSV samples.

[0058] The morphology of the etched region was characterized using optical microscopy and AFM, and elemental analysis of the cross-section of the n-TSV sample was performed using EDS. Optical micrograph of the etched region of the n-TSV sample (a), EDS elemental analysis results of the cross-section of the n-TSV sample (b), AFM characterization images of the etched region of the n-TSV sample with a scanning range of 20 μm × 20 μm (c) and 5 μm × 5 μm (d) are shown below. Figure 4 . Figure 4 (a), (c), and (d) show that the n-TSV heterostructure is uniformly etched within the etched area, and (b) shows that all three materials were successfully deposited according to the preset scheme.

[0059] Example 2

[0060] Chemical mechanical polishing:

[0061] TSV-A, TSV-B, and n-TSV samples were polished using an automatic pressure polishing machine with polyurethane polishing pads featuring K-grooves. The polishing head and polishing disc speeds were both 60 rpm. After every 60 seconds of polishing, the polishing pads were surface-conditioned for 120 seconds using a diamond abrasive dressing pad. The pH of the polishing solution was adjusted using potassium hydroxide (KOH) and nitric acid (HNO3). For each type of TSV-A, TSV-B, and n-TSV, a corresponding number of pieces were selected for polishing with different pressures, polishing times, and polishing solution compositions. Specific parameters are shown in Tables 1-3.

[0062] Table 1 TSV-A Sample Polishing Settings

[0063]

[0064] Table 2 TSV-B Sample Polishing Settings

[0065]

[0066] Table 3 Polishing settings for n-TSV samples

[0067]

[0068] In-situ dynamic characterization equipment parameter settings:

[0069] The TSV-A and TSV-B samples were detected using a platinum-iridium alloy coated probe 1. In PiFM amplitude modulation mode, the amplitude was 27.06 nm, the probe excitation frequency f1 = 69.6158 kHz, and the laser modulation frequency f... 调制 =361.6103kHz, optically induced demodulation frequency f 解调 =431.2261kHz, laser power 1.6mW. In KPFM amplitude modulation mode, the amplitude is consistent, and when scanning the surface potential of the material, the needle tip is 50nm away from the sample, and the voltage is 0.5V AC.

[0070] The n-TSV sample was detected using a platinum-iridium alloy coated probe 2 in PiFM amplitude modulation mode with an amplitude of 27.06 nm, a probe excitation frequency of f1 = 72.0612 kHz, and a laser modulation frequency of f. 调制 =372.9967kHz, optically induced demodulation frequency f 解调 =445.0579kHz, laser power 1.6mW. Detection was performed in CAFM contact mode using an 80nN load and a 1V voltage applied to the needle tip.

[0071] Analysis of polished sample test results:

[0072] The morphology (a) to (e) and phase diagrams (f) to (j) of polished TSV-A samples PiFM (numbers 1-5) are shown below. Figure 5 (a) to (e) and (f) to (j) correspond to numbers 1 to 5 respectively. The morphology diagrams show the height and roughness of the TSV-A samples after polishing under different conditions. Although the material composition changes can be determined from the phase diagrams, the material contrast is poor, making it impossible to distinguish the changes in the material inside and outside the pores after polishing.

[0073] The optical power amplitude diagrams (a) to (e) and optical power phase diagrams (f) to (j) of the PiFM of polished TSV-A samples 1 to 5 are shown below. Figure 6 (a) to (e) and (f) to (j) correspond to numbers 1 to 5 respectively. Based on the optical amplitude and phase, it can be seen that without polishing, as... Figure 6 As shown in (a) and (f), the amplitude and phase of the optical force inside and outside the hole remain uniform, indicating that Si3N4 material exists both inside and outside the hole. However, as polishing progresses, as... Figure 6 As shown in (b) to (e) and (g) to (j), some pores contain shadowed areas of optical amplitude and unstable areas of optical phase. Combining the morphology and phase diagrams, it can be inferred that a mixture of intrinsic impurities and SiO2 polishing particles exists within the pores. The difference in refractive index between these particles and Si3N4 causes the difference in optical amplitude and phase. As polishing progresses, as... Figure 6 As shown in (b) to (e) and (g) to (j), the optical force amplitude and phase inside and outside the hole are still uniform, and the optical force amplitude does not increase compared with the surface of the unpolished sample, indicating that there is still a Si3N4 film on the material surface and the Si region is not exposed.

[0074] The morphology (a) to (e) and surface potential (f) to (j) of the KPFM of polished TSV-A samples 1 to 5 are shown in the figure. Figure 7 (a) to (e) and (f) to (j) correspond to numbers 1 to 5 respectively. For example... Figure 7 As shown, the surface potentials inside and outside the holes of the unpolished TSV-A sample are consistent, indicating that the region is covered by a uniformly thick Si3N4 layer. Unlike the PiFM characterization results, the surface potential inside the holes of the polished TSV-A sample is lower than that outside, preventing the passage of particles. Figure 7(g) to (j) accurately determine whether Si3N4 still exists inside and outside the hole after polishing. This is because during polishing, the narrow space inside the hole restricts the movement of abrasive grains in the polishing fluid, resulting in a relatively dispersed pressure distribution and thus less material removal inside the hole. Conversely, the relatively open space on the outer surface allows abrasive grains to make better contact with the surface, leading to greater material removal. This greater material removal will generate more lattice defects and dangling bonds on the outer surface. These defects alter the electronic structure of the material surface, causing a change in the work function of Si3N4 and resulting in a difference between the surface potential inside and outside the hole.

[0075] Based on the feedback from PiFM and KPFM characterization results, operators can adjust polishing parameters in real time according to the test results, thereby achieving controllable polishing and sub-nanometer and atomic-level surface smoothing of TSV-A heterostructures, quantitatively analyzing material removal rate, interface residue and polishing uniformity, and providing real-time feedback for atomic-level smoothing and interface precision control.

[0076] The morphology (a) to (e) and phase diagrams (f) to (j) of polished TSV-B samples PiFM (numbers 1-5) are shown below. Figure 8 (a) to (e) and (f) to (j) correspond to numbers 1 to 5 respectively. The morphology diagrams show the height and roughness of the TSV-B samples after polishing under different conditions. Although the material composition changes can be determined from the phase diagrams, the material contrast is poor, making it impossible to distinguish the changes in the material inside and outside the pores after polishing.

[0077] The optical amplitude (a) to (e) and optical phase (f) to (j) of the PiFM of polished TSV-B samples 1 to 5 are shown in the figure. Figure 9 (a)~(e) and (f)~(j) correspond to numbers 1 to 5 respectively. Figure 9 As shown in (a) to (e), there is a difference in optical intensity amplitude inside and outside the hole in all five samples, with the optical intensity amplitude inside the hole being lower than outside. This is due to the lower refractive index of Si3N4 compared to Si, indicating that Si3N4 is still present inside the hole after polishing. Furthermore, regarding... Figure 9 Analysis of (b) and (g) with their corresponding morphology and phase diagrams reveals that the TSV-B sample exhibits localized high-protrusions, corresponding phase changes, and dark spots in both optical amplitude and phase on the outer surface of the pores. These characteristics, along with their optical amplitude and phase, differ from both Si and Si3N4, indicating the presence of impurities on the sample surface. (Observation) Figure 9Images (c) and (h), along with the corresponding morphology and phase diagrams, reveal a large number of granular protrusions within the pores of the TSV-B sample. This phenomenon is attributed to the use of a polishing slurry containing SiO2 particles in the polishing experiment, causing the agglomerated SiO2 particles to remain trapped within the pores of the TSV-B. These protrusions may also contain impurities remaining from the TSV-B preparation process. The low optical amplitude and unstable optical phase of these particles indicate that their optical power is lower than that of Si3N4 and Si, consistent with the optical properties that the refractive index of SiO2 is lower than that of Si3N4.

[0078] The morphology (a) to (e) and surface potential (f) to (j) of the KPFM of polished TSV-B samples 1 to 5 are shown in the figure. Figure 10 (a) to (e) and (f) to (j) correspond to numbers 1 to 5 respectively. For example... Figure 10 As shown, the surface potential inside the hole is lower than that outside the hole. This is due to the difference between the work function of Si3N4 inside the hole and the work function of Si outside the hole, which provides a certain auxiliary reference for the residual Si3N4 inside the hole.

[0079] Based on the feedback from PiFM and KPFM characterization results, operators can adjust polishing parameters in real time according to the test results, thereby achieving controllable polishing and sub-nanometer and atomic-level surface smoothing of TSV-B heterostructures, quantitatively analyzing material removal rate, interface residue and polishing uniformity, and providing real-time feedback for atomic-level smoothing and interface precision control.

[0080] The morphology images (a) to (b) and probe current maps (c) to (d) of the polished n-TSV samples 1 and 2 are shown in Figures 1 and 2, respectively. The morphology images reveal the changes in sample height and surface roughness. The probe current maps also show that during polishing, other materials on the n-TSV sample surface are removed, while the conductive material Co is gradually exposed, leading to an increase in the tip current.

[0081] The photodynamic amplitude (PFM) diagrams (a) to (b) and photodynamic phase (c) to (d) of the etched regions of polished n-TSV samples 1 and 2 correspond to samples 1 and 2, respectively. According to the PFM diagrams, the photoinduced force amplitude decreases in the etched regions. Combined with the probe current diagrams, this is because Co material is gradually exposed on the surface of the etched regions of the n-TSV samples. Since Co and Si have different refractive indices, this leads to a difference in the PFM amplitude between the etched and unetched regions. Simultaneously, the exposure of Co material also causes a difference in the photodynamic phase between the etched and unetched regions.

[0082] Based on the feedback from PiFM and CAFM characterization results, operators can adjust polishing parameters in real time according to the test results, thereby achieving controllable polishing and sub-nanometer and atomic-level surface smoothing of n-TSV heterostructures, quantitatively analyzing material removal rate, interface residue and polishing uniformity, and providing real-time feedback for atomic-level smoothing and interface precision control.

[0083] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. An in-situ online method for characterizing chemical mechanical polishing of a through-silicon via heterostructure, comprising: The method comprises the following steps: The silicon through hole heterostructure or nano silicon through hole heterostructure sample in the chemical mechanical polishing process is in-situ characterized on-line by using a light-induced force microscope, a scanning Kelvin probe microscope and a conductive atomic force microscope, real-time surface topography height information, light force amplitude information, light force phase information, surface potential information and probe current information of the sample are obtained, the obtained information is analyzed, and in-situ dynamic nondestructive detection of surface topography height, roughness and component change information of the silicon through hole heterostructure or nano silicon through hole heterostructure sample in the chemical mechanical polishing process is realized.

2. The method of claim 1, wherein the in-situ, in-line characterization of a chemical mechanical polishing of a through-silicon via heterostructure is performed by a method comprising: The light-induced force microscope is a visible light-induced force microscope disclosed in the patent application CN120703413A.

3. The in-situ online method for characterizing a chemical mechanical polishing of a through-silicon via heterostructure according to claim 1, wherein, For the sample of through-silicon via heterostructure, the amplitude modulation mode was used in the optical force microscopy. The parameters were set as follows: amplitude 27.06 nm, probe oscillation frequency f1=69.6158 kHz, laser modulation frequency f 调制 =361.6103 kHz, optical force demodulation frequency f 解调 =431.2261 kHz, and laser power 1.6 mW.

4. The in-situ online method for characterizing a chemical mechanical polishing through-silicon via heterostructure of claim 1, wherein, When detecting the sample of nano-silicon through-hole heterostructure, the amplitude modulation mode is used in the visible light-induced force microscope, and the parameters are set as follows: amplitude 27.06 nm, probe excitation frequency f1=72.0612 kHz, laser modulation frequency f 调制 =372.9967 kHz, light-induced force demodulation frequency f 解调 =445.0579 kHz, and laser power 1.6 mW.

5. The in-situ online method for characterizing a chemical mechanical polishing through-silicon via heterostructure of claim 1, wherein, When the silicon through hole heterostructure sample is detected, the scanning Kelvin probe microscope adopts an amplitude modulation mode, and the parameters are set as follows: the distance between the needle tip and the sample is 50 nm, and an alternating voltage of 0.5 V is applied.

6. The in-situ online method for characterizing a chemical mechanical polishing through-silicon via heterostructure of claim 1, wherein, When the nano silicon through hole heterostructure sample is detected, the conductive atomic force microscope adopts a contact mode, and the parameters are set as follows: the load is 80 nN, and a voltage of 1 V is applied to the needle tip.

7. An optimized chemical mechanical polishing method based on the in-situ online characterization method of the chemical mechanical polishing of a through-silicon via heterostructure according to any one of claims 1 to 6, characterized in that, The method comprises the following steps: the real-time surface topography height, roughness and component change information of the silicon through hole heterostructure or nano silicon through hole heterostructure sample in the chemical mechanical polishing process obtained by the in-situ characterization method of the silicon through hole heterostructure in the chemical mechanical polishing process according to any one of claims 1-6 is combined with preset polishing process parameters and target surface precision, and the polishing process parameters are adjusted in real time to achieve the target surface precision.

Images