A copper electroplating apparatus and method that dynamically monitors and adjusts process parameters

Abstract

The application provides a copper electroplating device and method capable of dynamically monitoring and adjusting process parameters, and belongs to the field of semiconductor advanced packaging manufacturing. The device comprises an electroplating tank, which is provided with an anode, a cathode jig and an electroplating liquid circulation and bubbling system. The core improvement of the application is that an X-Ray imaging system, a processing unit and a regulation and control unit are integrated. The X-Ray imaging system is used for acquiring a transmission image sequence of a TGV / BGV / TSV / BSV through-hole structure on a workpiece in real time during the electroplating process. The processing unit calculates the real-time copper filling saturation of each through-hole through a specific formula based on the image sequence, and can identify a cavity defect. The regulation and control unit automatically adjusts process parameters such as local current density, pulse parameters, stirring rate or temperature when the saturation is lower than a preset threshold or a defect is detected. The application realizes in-situ, real-time and visual monitoring and closed-loop intelligent regulation and control of the electroplating filling process in a deep hole, can actively inhibit the generation of cavities, and significantly improves the filling yield and process stability.

Description

Technical Field

[0001] This invention relates to the field of advanced semiconductor packaging manufacturing. More specifically, it relates to an apparatus and method for copper electroplating filling of high aspect ratio through-hole structures (such as glass through-hole (TGV), buried glass through-hole (BGV), silicon through-hole (TSV), etc.). More specifically, it relates to an apparatus and intelligent control method that can monitor the copper filling state in the through-hole in situ and in real time during the electroplating process and dynamically adjust the electroplating parameters based on the monitoring results. Background Technology

[0002] In advanced packaging technologies such as 3D integration, fan-out packaging, and GlassCore interconnect, vertical interconnection is achieved by filling vias with copper through electroplating. However, due to factors such as uneven electric field distribution within high aspect ratio vias, limited mass transfer of the electroplating solution, and uneven consumption of additives, defects such as voids, gaps, and pinch-offs are easily generated during the electroplating filling process, which seriously affects the reliability of the interconnect and the yield of the final product.

[0003] Currently, monitoring of the electroplating process mainly relies on monitoring macroscopic process parameters, such as total charge (ampere-hours), average current density, bath temperature, pH value, and chemical concentration. These methods are indirect monitoring and cannot directly and accurately reflect the dynamic process and three-dimensional morphology of metal deposition inside each through-hole. Therefore, process development is highly dependent on experience and cannot intervene when defects occur.

[0004] For the inspection of electroplating results, offline and destructive methods are commonly used. For example, this involves cutting the sample and observing it using a scanning electron microscope (SEM), or using a high-resolution microfocus X-ray imaging system to image and analyze the electroplated sample. These post-inspection methods suffer from significant delays; once defects are discovered, the workpiece is often beyond repair, resulting in a huge waste of materials and time. Furthermore, existing technologies (such as some schemes that improve uniformity by optimizing bath circulation) do not address the direct sensing and feedback control of the filling dynamics within the pores.

[0005] Therefore, there is an urgent need in this field for a technical solution that can directly and non-destructively obtain the internal filling status information of through holes in situ during the electroplating process, and can adjust the process parameters in real time and automatically based on the information, thereby proactively preventing the generation of defects. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a copper electroplating equipment and method capable of dynamically monitoring and adjusting process parameters. Specifically, based on a traditional electroplating tank, it creatively integrates an online X-ray imaging system, a high-speed processing unit, and an intelligent control unit, forming a real-time closed-loop control system for sensing, analysis, decision-making, and execution.

[0007] This invention provides the following technical solution: A copper electroplating apparatus capable of dynamically monitoring and adjusting process parameters includes an electroplating tank, wherein the electroplating tank is provided with an anode, a fixture for fixing the cathode workpiece, and an electroplating solution circulation and bubbling system. The improvement lies in that it further includes: An X-Ray imaging system, integrated inside or outside the electroplating tank, is used to acquire in real time a sequence of through-hole images of the cathode workpiece during the electroplating process. The processing unit is connected to the X-Ray imaging system signal and is used to calculate the real-time copper filling saturation in the via based on the image sequence; The control unit, which is signal-connected to the processing unit, is used to generate control commands to automatically adjust one or more electroplating process parameters when the real-time copper filling saturation is lower than a preset threshold.

[0008] Preferably, the X-Ray imaging system includes a microfocus X-Ray source and a high frame rate flat panel detector (such as FPXD); the flat panel detector has an imaging frame rate ≥ 50fps and a spatial resolution ≤ 50μm to meet the requirements for capturing dynamic processes.

[0009] Furthermore, the integration of the microfocus X-ray source with the flat panel detector has three feasible embodiments: Example 1: Both are integrated into the inside of the electroplating tank through a waterproof sealing structure; Example 2: The X-ray source is located outside the tank, and the flat panel detector is integrated inside the tank through a waterproof sealing structure; Example 3: Both are located outside the tank, and the tank has an X-ray transmission window (such as a beryllium window) at the corresponding position.

[0010] Preferably, the flat panel detector is an FPXD flat panel detector, the surface of which is coated with an aluminum oxide (Al2O3) anti-corrosion layer formed by atomic layer deposition to withstand long-term corrosion by acidic electroplating solutions.

[0011] Furthermore, the processing unit includes a saturation calculation module, wherein the copper filling saturation S(x,y,t) is calculated using the following formula: S(x,y,t)=(I0(x,y)-I(x,y,t))*K / I0(x,y) Where I0(x,y) is the baseline X-ray image intensity of the via region without copper deposition, I(x,y,t) is the X-ray image intensity at the current time t, and K is the linear attenuation coefficient of the copper material on the characteristic energy of the X-ray emitted by the microfocus X-ray source. This formula quantifies the change in X-ray intensity as the copper deposition thickness (saturation).

[0012] Furthermore, the processing unit also includes an artificial intelligence-based defect identification module (such as a convolutional neural network CNN) for performing spatiotemporal analysis on continuous image sequences, automatically segmenting and identifying defective regions such as holes, and outputting their three-dimensional position coordinates and volume proportions, providing more detailed input for precise control.

[0013] Preferably, the electroplating process parameters that the control unit can adjust include, but are not limited to, one or more of the following: local current density applied to the cathode workpiece, frequency and duty cycle of the pulse waveform, stirring rate of the electroplating solution circulation and bubbling system, and temperature of the electroplating solution.

[0014] According to another aspect of the present invention, a method for dynamic monitoring and adjustment of a copper electroplating process based on the above-mentioned equipment is provided, comprising the following steps: S1: During the electroplating process, a real-time transmission image sequence of the through-hole structure on the cathode workpiece is acquired through an X-Ray imaging system integrated inside or outside the electroplating tank. S2: The processing unit calculates the real-time copper fill saturation of each via based on the image sequence; S3: Compare the real-time copper filling saturation with a preset saturation threshold; S4: When the real-time copper filling saturation is lower than the saturation threshold, the control unit automatically adjusts one or more electroplating process parameters to increase the copper deposition rate in the corresponding area.

[0015] Preferably, the method further includes: in step S2, analyzing the image sequence through the defect identification module to dynamically identify the initiation location and expansion trend of the voids; in step S4, the control unit adjusts the electroplating parameters of the corresponding area in a targeted manner according to the location and trend of the voids.

[0016] The beneficial effects of this invention are: 1: This invention is the first to deeply integrate X-Ray online detection with electroplating production process, allowing for non-destructive direct observation of the filling process inside the through hole while electroplating is in progress, turning the process from a black box to a white box. 2: The copper filling saturation, a physical quantity that directly reflects the state of matter inside the hole, was creatively defined, and its quantitative calculation formula was given. It was used as the core control index to replace the traditional indirect macroscopic parameters, resulting in a qualitative leap in control precision and targeting. 3: A complete real-time feedback chain has been formed, which includes image acquisition, saturation calculation / defect recognition, intelligent decision-making, and parameter control. This transforms the electroplating process from an open-loop mode based on a fixed formula to a closed-loop intelligent mode based on real-time adaptive adjustment. It can proactively predict and suppress defects, fundamentally improving the process window and product yield. 4. The three proposed X-Ray system integration schemes balance imaging quality, system reliability, maintenance convenience and cost, and can adapt to different application scenarios and customer needs. At the same time, the schemes are compatible with a variety of adjustable parameters and have flexible control strategies. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the device structure in Embodiment 1 of the present invention (both the X-Ray source and the detector are built into the slot).

[0018] Figure 2 This is a schematic diagram of the device structure in Embodiment 2 of the present invention (external X-Ray source, internal detector).

[0019] Figure 3 This is a schematic diagram of the device structure in Embodiment 3 of the present invention (both the X-Ray source and the detector are external).

[0020] Markings in the diagram: 1. Electroplating tank; 2. Cathode workpiece; 3. Anode; 4. Electroplating solution circulation and bubbling system; 5. Micro-focus X-ray source; 6. Flat panel detector. Detailed Implementation

[0021] The core of this invention lies in integrating an online X-ray imaging system into the electroplating equipment to acquire image information reflecting the filling state within the holes in real time, and then converting this information into quantifiable copper filling saturation through an algorithm. When the system detects that the saturation growth in a certain area is slow or abnormal, it can immediately trigger a control mechanism to adjust the corresponding electroplating parameters (such as locally increasing the current or changing the stirring), thereby correcting uneven deposition and achieving precise control by addressing only the areas that are not satisfactory.

[0022] Example 1: like Figure 1 As shown, in this embodiment, both the micro-focus X-Ray source 5 and the high frame rate FPXD flat panel detector 6 are directly mounted inside the electroplating tank 1 through a high-grade sealing structure. The cathode workpiece 2 (such as a glass wafer with TGV) is mounted on a fixture and placed between the X-Ray source 5 and the detector 6. The X-Ray source 5 can be mounted on a servo drive mechanism to achieve two-dimensional scanning to monitor different positions of the workpiece.

[0023] The X-Ray source 5 uses an open-type microfocusing tube with a focal spot size ≤0.8μm. The FPXD flat panel detector 6 has a pixel size of 30μm, a dynamic range of 16bit, and an 80nm thick Al2O3 protective layer grown on its surface using atomic layer deposition (ALD) technology.

[0024] The detector 6 is encapsulated in an acrylic shell and vacuum potted with acid-resistant epoxy resin. After curing, it forms a bubble-free sealed body, achieving an IP68 protection rating, and can be immersed in copper sulfate plating solution for a long time.

[0025] This scheme has the shortest optical path, no additional window attenuation, the highest imaging clarity, excellent spatial resolution, and is easy to implement multi-angle scanning imaging.

[0026] Example 2: like Figure 2 As shown, in this embodiment, the micro-focus X-Ray source 5 is placed outside the electroplating tank 1, while the FPXD flat panel detector 6 is still sealed and built into the tank. The electroplating tank 1 has a beryllium window that transmits X-Ray light on the side wall facing the X-Ray source 5.

[0027] The X-Ray source 5 is completely isolated from corrosive environments, making maintenance and replacement convenient and extending its service life. Furthermore, due to the availability of higher power and stronger penetration options for external X-Ray sources, good image quality is still guaranteed. The internal structure of the tank is relatively simplified.

[0028] Example 3: like Figure 3 As shown, in this embodiment, the microfocus X-ray source 5 and the FPXD flat panel detector 6 are both located outside the electroplating tank 1, positioned on opposite sides of the tank. Beryllium windows are installed at corresponding positions on both sides of the tank.

[0029] The entire X-ray imaging system is physically isolated from the electroplating environment, requiring no anti-corrosion design, resulting in the highest system reliability and the easiest maintenance. The disadvantages are that X-rays require two penetration windows, leading to some signal attenuation and necessitating a higher-power X-ray source.

[0030] Core Algorithms and Workflow The following details the working method of the device of the present invention: The process begins in step S1, when the electroplating process is started. At the same time, the online X-Ray imaging system starts working at a high frame rate (e.g., 60fps), acquiring image sequences that penetrate the workpiece in real time and transmitting them to the processing unit.

[0031] In step S2, the processing unit performs image processing and calculations. First, it retrieves the baseline image I0(x,y) acquired before electroplating begins. This is based on the X-Ray absorption law. As electroplating progresses, copper is deposited within the vias. Because copper's density is much higher than the electroplating solution, its absorption of X-Ray is stronger, causing the intensity I(x,y,t) received by the detector to gradually decrease. The processing unit uses the formula S(x,y,t)=(I0(x,y)-I(x,y,t))*K / I0(x,y) to calculate the copper filling saturation since the start of electroplating for each pixel in the image (corresponding to a small area on the workpiece), thereby generating a global saturation distribution map.

[0032] In parallel, in step S3, the defect identification module (a pre-trained convolutional neural network model) in the processing unit analyzes consecutive image frames. The model can identify abnormal regions that do not conform to the normal filling pattern, such as the initial nucleus of a hole, and accurately segment the hole region, calculate the percentage of its volume to the volume of the corresponding through-hole, and locate its three-dimensional coordinates.

[0033] Step S4 is the judgment step. The processing unit compares the calculated real-time saturation with the preset ideal process curve and combines it with the defect identification results. The judgment criteria may be: the saturation growth rate of a certain area is less than 70% of the average value, or voids with a volume ratio greater than 0.1% are identified.

[0034] If an anomaly is detected (step S5), proceed to step S6, where the control unit generates specific control commands based on a preset control strategy library. For example: If the saturation of a certain edge region is low, the multi-channel power supply is instructed to increase the current output of the cathode in that region.

[0035] If a void is detected at the bottom of the hole, the pulse power supply is instructed to increase the proportion of reverse pulses in subsequent cycles, and at the same time, the micro-nozzle located below the hole is instructed to enhance local stirring.

[0036] In step S7, each actuator (power supply, pump, valve, etc.) receives and executes the control command. Subsequently, the system immediately returns to step S1 and enters the next monitoring-control cycle, forming a real-time closed-loop control until the electroplating process is successfully completed.

[0037] Thus, this invention, through the combination of hardware integration and software algorithms, provides an unprecedented real-time sensing and intelligent control capability for high aspect ratio through-hole copper electroplating processes. It not only greatly improves process development efficiency but also proactively ensures yield during production, which is of great significance to advanced semiconductor packaging manufacturing.

[0038] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A copper electroplating equipment capable of dynamically monitoring and adjusting process parameters, comprising an electroplating tank (1), wherein the electroplating tank (1) is provided with an anode (3), a fixture for fixing a cathode workpiece (2), and an electroplating solution circulation and bubbling system (4), characterized in that, Also includes: An X-Ray imaging system is integrated inside or outside the electroplating tank (1) to acquire a real-time transmission image sequence of the through-hole structure on the cathode workpiece (2) during the electroplating process. The processing unit is connected to the X-Ray imaging system signal and is used to calculate the real-time copper filling saturation in the via based on the image sequence; The control unit, which is signal-connected to the processing unit, is used to generate control commands to automatically adjust one or more electroplating process parameters when the real-time copper filling saturation is lower than a preset threshold.

2. The copper electroplating equipment according to claim 1, characterized in that, The X-Ray imaging system includes a microfocus X-Ray source (5) and a high frame rate flat panel detector (6); the flat panel detector (6) has an imaging frame rate ≥ 50fps and a spatial resolution ≤ 50μm.

3. The copper electroplating equipment according to claim 2, characterized in that, The integration of the micro-focus X-Ray source (5) and the flat panel detector (6) is as follows: the micro-focus X-Ray source (5) and the flat panel detector (6) are both integrated inside the electroplating tank (1) through a waterproof sealing structure; or, the micro-focus X-Ray source (5) is located outside the electroplating tank (1), and the flat panel detector (6) is integrated inside the electroplating tank (1) through a waterproof sealing structure; or, the micro-focus X-Ray source (5) and the flat panel detector (6) are both located outside the electroplating tank (1), and the electroplating tank (1) is provided with an X-Ray transmission window at the corresponding position.

4. The copper electroplating equipment according to claim 2 or 3, characterized in that, The flat panel detector (6) is an FPXD flat panel detector, and its surface is coated with an aluminum oxide anti-corrosion layer formed by atomic layer deposition.

5. The copper electroplating equipment according to claim 1, characterized in that, The processing unit includes a saturation calculation module, and the formula for calculating the copper filling saturation S(x,y,t) is as follows: S(x,y,t)=(I0(x,y)-I(x,y,t))*K / I0(x,y), Wherein, I0(x,y) is the reference X-Ray image intensity of the via region without copper deposition, I(x,y,t) is the X-Ray image intensity at the current time t, and K is the linear attenuation coefficient of the copper material on the characteristic energy of the X-ray emitted by the microfocus X-Ray source.

6. The copper electroplating equipment according to claim 5, characterized in that, The processing unit also includes a defect identification module, which uses a convolutional neural network to analyze the continuous image sequence, segment the hole region, and output the volume ratio and three-dimensional position coordinates of the hole.

7. The copper electroplating equipment according to claim 1, characterized in that, The electroplating process parameters that the control unit can adjust include one or more of the following: local current density applied to the cathode workpiece (2), frequency and duty cycle of the pulse waveform, stirring rate of the electroplating solution circulation and bubbling system (4), and temperature of the electroplating solution.

8. A method for dynamic monitoring and adjustment of a copper electroplating process based on the equipment described in any one of claims 1-7, characterized in that, Includes the following steps: During the electroplating process, a real-time transmission image sequence of the through hole structure on the cathode workpiece (2) is acquired by an X-Ray imaging system integrated inside or outside the electroplating tank (1). Based on the image sequence, the processing unit calculates the real-time copper fill saturation of each via; The real-time copper fill saturation is compared with a preset saturation threshold. When the real-time copper filling saturation is lower than the saturation threshold, the control unit automatically adjusts one or more electroplating process parameters to increase the copper deposition rate in the corresponding area.

9. The method according to claim 8, characterized in that, Also includes: In the step of calculating real-time copper filling saturation, the image sequence is analyzed by the defect identification module to dynamically identify the initiation location and expansion trend of voids; In the step of adjusting the electroplating process parameters, the control unit adjusts the electroplating parameters of the corresponding area according to the location and trend of the voids.

Images