Electroplating high hardness gold-antimony alloy process

By using multi-frequency perturbation sequence and impedance fingerprint technology, the problem of insufficient coating consistency in the electroplating process of high-hardness gold-antimony alloy was solved, realizing online endpoint identification and quality recording, and improving production efficiency and product consistency.

CN123108683APending Publication Date: 2026-11-17NINGBO DUJINHUI ENVIRONMENTAL PROTECTION CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610732207.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-26
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing high-hardness gold-antimony alloy electroplating processes cannot accurately reflect changes in the surface condition and local polarization state of different workpieces during electroplating, resulting in insufficient coating consistency. Furthermore, offline detection methods cannot be adjusted in real time, making it difficult to meet the requirements for high consistency and online production.

Method used

By establishing a multi-frequency perturbation sequence, collecting the voltage response between the cathode and the reference terminal, constructing a multi-frequency impedance fingerprint, and combining the qualified endpoint center fingerprint and the allowable deviation of the endpoint, the electroplating endpoint is judged in real time, and the key quality indicators of the coating are recorded.

Benefits of technology

This technology achieves stable and consistent online endpoint identification of high-hardness gold-antimony alloy coatings, reduces precious metal consumption, improves batch consistency and process traceability, and reduces testing cycle and sample damage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123108683A_ABST
    Figure CN123108683A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of electroplating process control, and discloses an electroplating high-hardness gold-antimony alloy process. By establishing an effective cathode surface of an electroplated plated part and a pre-plating reference, a direct-current electroplating main current and an alternating-current disturbance current amplitude are determined, and multi-frequency disturbance collection is carried out during gold-antimony alloy deposition to construct a multi-frequency impedance fingerprint of the current plated part; then, a qualified end center fingerprint and an end point allowable deviation amount are established based on a qualified standard sample, the current multi-frequency impedance fingerprint is compared with the qualified end center fingerprint, an end point deviation amount is formed, electroplating termination judgment is executed, and online thickness, antimony content, grain record amount and process stop results are generated. The scheme solves the problem that the prior art relies on fixed time, experience parameters or offline sampling judgment of the end point, and it is difficult to reflect the real state of the plating layer in a timely manner, and helps to reduce the overplating, underplating and rework risks, and improve the controllability and traceability of the electroplating process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of electroplating process control technology, specifically to an electroplating process for high-hardness gold-antimony alloy. Background Technology

[0002] Gold-antimony alloy coatings, due to their high hardness, good wear resistance, and stable surface properties, are commonly used for surface treatment of precision connectors, miniature functional components, wear-resistant contact parts, and parts requiring high-hardness precious metal coatings. Current high-hardness gold-antimony alloy electroplating processes typically rely on constant current density, electroplating time, cumulative charge, and empirically calculated film thickness as the main process control criteria. After electroplating, offline film thickness testing, composition analysis, hardness testing, or microstructure analysis are used to determine whether the coating meets process requirements. Some processes also collect conventional parameters such as tank voltage, current, conductivity, and temperature during electroplating to determine whether the electroplating process is in a stable state.

[0003] However, gold-antimony alloys belong to a co-deposition system, and the coating hardness is related to the coating thickness, antimony co-deposition ratio, grain size, and interface state. Relying solely on electroplating time or cumulative charge for endpoint control is insufficient to accurately reflect the impact of different substrate surface conditions, local polarization states, and deposition interface changes on the final coating performance. When there are differences in substrate surface roughness, effective cathode area, contact state, or local current distribution, gold-antimony alloy coatings formed under the same electroplating time may exhibit thickness deviations, antimony content deviations, or differences in microstructure density, leading to insufficient consistency in high-hardness coatings. While existing offline detection methods can obtain quality information such as film thickness, composition, or grain size after electroplating, the detection results are usually lagging behind the electroplating process and cannot be directly used for endpoint determination during deposition. For precious metal gold-antimony alloy coatings, excessively long electroplating times can easily lead to increased gold material consumption, increased internal stress in the coating, or abnormal surface microstructure; excessively short electroplating times may result in insufficient coating thickness, insufficient hardness, or unstable wear resistance. Offline sampling inspection also suffers from problems such as sample damage, long testing cycles, and inability to cover every piece being plated, making it difficult to meet the requirements of high consistency and online production. Furthermore, existing online electrochemical monitoring often focuses on single voltage, current, or single-frequency impedance parameters, resulting in limited information dimensions and making it difficult to simultaneously characterize thickness growth, phase changes, inter-frequency response differences, and interface structure evolution during the gold-antimony alloy deposition process. Single-point threshold control is also easily affected by instrument resolution, transient noise, and differences in the initial state of the substrate, leading to insufficient stability in endpoint determination.

[0004] Therefore, this case aims to propose a high-hardness gold-antimony alloy electroplating process that transforms the interface state changes of the high-hardness gold-antimony alloy coating during the deposition process into voltage response changes under multi-frequency perturbation, and then characterizes the degree to which the coating approaches the qualified endpoint through impedance modulus shift, phase shift and their combined fingerprints. Summary of the Invention

[0005] This invention provides a process for electroplating high-hardness gold-antimony alloy, which helps to solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a process for electroplating high-hardness gold-antimony alloy, comprising:

[0007] Establish the effective cathode surface of the workpiece to be electroplated, obtain the effective cathode area, set the DC current density for the gold-antimony alloy electroplating stage, and calculate the DC electroplating main current and AC disturbance current amplitudes according to the effective cathode area and DC current density; establish a multi-frequency disturbance sequence and a single-frequency sampling window, perform pre-plating reference disturbance acquisition, and calculate the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response; during the gold-antimony alloy deposition process, acquire the online voltage response between the cathode and the reference terminal according to the multi-frequency disturbance sequence, and extract the online voltage response amplitude and online voltage response phase; construct the current workpiece to be electroplated based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase. The process involves: generating multi-frequency impedance fingerprints of the workpiece; preparing qualified high-hardness gold-antimony alloy standard samples; collecting the endpoint impedance modulus and endpoint phase of each qualified standard sample; establishing the qualified endpoint center fingerprint and the allowable endpoint deviation; based on the multi-frequency impedance fingerprint of the current workpiece, generating online thickness records, online antimony content records, and online grain records; comparing the multi-frequency impedance fingerprint of the current workpiece with the qualified endpoint center fingerprint at each perturbation frequency to calculate the endpoint deviation; and performing electroplating termination judgment based on the endpoint deviation and the longest allowable electroplating time; cutting off the sinusoidal perturbation current corresponding to the amplitude of the DC electroplating main current and the AC perturbation current at the electroplating stop time, generating a stop data set, calculating the stop charge, and outputting the process stop result.

[0008] Optionally, the step of establishing the effective cathode surface of the workpiece to be electroplated, obtaining the effective cathode area, setting the DC current density for the gold-antimony alloy electroplating stage, and calculating the DC electroplating main current and AC disturbance current amplitude according to the effective cathode area and DC current density specifically includes:

[0009] The conductive surface of the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is taken as the effective cathode surface, and the effective cathode area of ​​the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is measured; the DC current density of the gold-antimony alloy electroplating stage is set as the calculation parameter of the electroplating main current; the effective cathode area is multiplied by the DC current density to calculate the DC electroplating main current of the gold-antimony alloy electroplating stage; the DC electroplating main current is converted according to a ratio of one-fiftieth to obtain the amplitude of the AC disturbance current superimposed during multi-frequency impedance acquisition.

[0010] Optionally, the step of establishing a multi-frequency disturbance sequence and a single-frequency sampling window, performing pre-plating reference disturbance acquisition, and calculating the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response specifically includes:

[0011] Establish a multi-frequency disturbance frequency number set, setting the total number of disturbance frequencies to no less than five, and assigning a corresponding disturbance frequency number index to each disturbance frequency; set the lowest and highest frequencies in the multi-frequency disturbance frequency sequence, ensuring the lowest frequency is greater than zero and less than the highest frequency; generate the disturbance frequencies corresponding to each disturbance frequency number index sequentially according to the geometric interval relationship from the lowest to the highest frequency; set the number of voltage sampling points under each disturbance frequency to no less than sixteen, and establish a single-frequency sampling number set; generate the sampling time corresponding to each sampling number under the current disturbance frequency according to the proportional relationship between the sampling number and the number of voltage sampling points, combined with the current disturbance frequency; during the gold-antimony alloy deposition stage, the DC... The electroplating main current is superimposed with the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current to form an online single-frequency sinusoidal disturbance current according to the current disturbance frequency; the effective lower limit and effective upper limit of the voltage acquisition device are obtained, and the effective lower limit is made smaller than the effective upper limit; before the gold-antimony alloy deposition begins, the DC electroplating main current is kept off, and only the AC disturbance current amplitude is used to perform a multi-frequency disturbance acquisition on the same workpiece according to the multi-frequency disturbance sequence to form the pre-plating reference disturbance current at each disturbance frequency; at each disturbance frequency, the pre-plating voltage response is acquired according to the corresponding sampling time; when all pre-plating sampled voltages at the current disturbance frequency are within the effective lower limit and effective range of the voltage acquisition device... When the voltage is within the effective range, retain all pre-plating sampled voltages at the current disturbance frequency; when there is a pre-plating sampled voltage at the current disturbance frequency that is lower than the lower limit or higher than the upper limit of the effective range, re-acquire the pre-plating voltage response data at the current disturbance frequency; when all pre-plating sampled voltages after re-acquisition are between the lower limit and the upper limit of the effective range of the voltage acquisition device, retain all pre-plating sampled voltages obtained after re-acquisition; when there is still a pre-plating sampled voltage that is lower than the lower limit or higher than the upper limit of the effective range after re-acquisition, stop the endpoint determination of the current plated part, record the pre-plating reference abnormal state, and do not generate the multi-frequency impedance fingerprint of the current plated part; for the pre-plating voltage response at the current disturbance frequency... Sine projection is performed by multiplying each pre-plating sampled voltage by the corresponding sinusoidal reference value at the sampling time, summing the results, and normalizing the sum according to the number of voltage sampling points to form the sinusoidal projection component of the pre-plating voltage response at the current disturbance frequency. Cosine projection is performed on the pre-plating voltage response at the current disturbance frequency by multiplying each pre-plating sampled voltage by the corresponding cosine reference value at the sampling time, summing the results, and normalizing the sum according to the number of voltage sampling points to form the cosine projection component of the pre-plating voltage response at the current disturbance frequency. The square of the sinusoidal projection component and the square of the cosine projection component of the pre-plating voltage response at the current disturbance frequency are added together, and the square root of the sum is taken to calculate the amplitude of the pre-plating voltage response at the current disturbance frequency.When the pre-plating voltage response amplitude is greater than zero, a binary arctangent operation is performed based on the cosine and sine projection components of the pre-plating voltage response at the current disturbance frequency to calculate the pre-plating reference phase at the current disturbance frequency. The pre-plating voltage response amplitude is then divided by the AC disturbance current amplitude to calculate the pre-plating reference impedance modulus at the current disturbance frequency. When the pre-plating voltage response amplitude is equal to zero, the endpoint determination of the current workpiece is stopped, the pre-plating reference abnormal state is recorded, and no multi-frequency impedance fingerprint of the current workpiece is generated.

[0012] Optionally, the step of acquiring the online voltage response between the cathode and the reference terminal according to a multi-frequency perturbation sequence during the gold-antimony alloy deposition process, and extracting the online voltage response amplitude and phase, specifically includes:

[0013] Establish a complete multi-frequency acquisition count index and record each complete multi-frequency acquisition process according to the complete multi-frequency acquisition count index; during the disturbance period of the current disturbance frequency in any complete multi-frequency acquisition, acquire the voltage between the cathode and the reference terminal at each sampling moment under the current disturbance frequency to form the online sampling voltage; use the effective lower limit and effective upper limit of the voltage acquisition device to perform integrity judgment on the online sampling voltage; when all online sampling voltages under the current disturbance frequency in any complete multi-frequency acquisition are between the effective lower limit and effective upper limit, retain the current disturbance frequency in the current complete multi-frequency acquisition. All online sampled voltages at the current disturbance frequency; if, during any complete multi-frequency acquisition, an online sampled voltage at the current disturbance frequency is lower than the lower limit or higher than the upper limit of the effective range, the voltage response data at the current disturbance frequency in the current complete multi-frequency acquisition is reacquired; if all online sampled voltages after reacquisition are between the lower and upper limits of the effective range, all online sampled voltages obtained after reacquisition are retained; if, after reacquisition, an online sampled voltage is still lower than the lower limit or higher than the upper limit of the effective range, the endpoint determination of the current plated part is stopped, the sampling abnormal state is recorded, and no current endpoint determination is generated. The multi-frequency impedance fingerprint of the plated component; For any complete multi-frequency acquisition, perform sinusoidal projection on all effective voltage sample values ​​at the current disturbance frequency, multiply each effective voltage sample value by the corresponding sinusoidal reference value at the sampling time, sum the results, and normalize according to the number of voltage sampling points to form an online voltage sinusoidal projection component; For any complete multi-frequency acquisition, perform cosine projection on all effective voltage sample values ​​at the current disturbance frequency, multiply each effective voltage sample value by the corresponding cosine reference value at the sampling time, sum the results, and normalize according to the number of voltage sampling points to form an online voltage cosine projection component; [The text abruptly ends here, likely due to an incomplete translation or source material.] The square of the sinusoidal projection component of the line voltage is added to the square of the cosine projection component of the line voltage, and the square root of the sum is taken to calculate the amplitude of the line voltage response at the current disturbance frequency in the current complete multi-frequency acquisition. When the amplitude of the line voltage response is greater than zero, a binary arctangent operation is performed on the cosine projection component and the sinusoidal projection component of the line voltage to calculate the phase of the line voltage response at the current disturbance frequency in the current complete multi-frequency acquisition. When the amplitude of the line voltage response is equal to zero, the endpoint determination of the current plated part is stopped, the abnormal state of the impedance signal is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0014] Optionally, constructing the multi-frequency impedance fingerprint of the workpiece to be plated based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase specifically includes:

[0015] The online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition is calculated by dividing the online voltage response amplitude at the current disturbance frequency by the AC disturbance current amplitude. The online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition is then compared with the pre-plating reference impedance modulus at the corresponding disturbance frequency using a dimensionless ratio transformation. The result of the ratio transformation is then subjected to a natural logarithmic operation to form the impedance modulus offset at the current disturbance frequency in the current complete multi-frequency acquisition. The phase of the online voltage response at the current disturbance frequency in the current complete multi-frequency acquisition is subtracted from the pre-plating reference phase at the corresponding disturbance frequency, and then dimensionlessized according to the circumferential angle reference to form the phase offset at the current disturbance frequency in the current complete multi-frequency acquisition. The impedance modulus offsets and phase offsets obtained from each disturbance frequency in the current complete multi-frequency acquisition are arranged alternately according to the disturbance frequency numbering order to form the multi-frequency impedance fingerprint vector corresponding to the current complete multi-frequency acquisition. The dimension of the multi-frequency impedance fingerprint vector is set to twice the total number of disturbance frequencies.

[0016] Optionally, the preparation of qualified high-hardness gold-antimony alloy standard samples, the acquisition of the endpoint impedance modulus and endpoint phase of each qualified standard sample, and the establishment of a qualified endpoint center fingerprint and an allowable endpoint deviation specifically include:

[0017] Using the same substrate shape, effective cathode area, and DC current density as the workpiece to be electroplated, qualified high-hardness gold-antimony alloy standard samples are prepared, with a minimum of three qualified standard samples. A set of qualified standard samples with qualified inspection records established according to process inspection specifications is created, and a qualified standard sample number index is assigned to each sample. For each qualified standard sample, using the same multi-frequency disturbance sequence, sampling time, AC disturbance current amplitude, voltage projection process, and impedance modulus calculation process as the workpiece to be plated, the pre-plating reference impedance modulus and pre-plating reference phase of the current qualified standard sample at each disturbance frequency are established before the gold-antimony alloy deposition begins. The endpoint of qualification is determined by collecting the endpoint impedance modulus and endpoint phase of the current qualified standard sample at each disturbance frequency, and using the electroplating stop time corresponding to the time when the current qualified standard sample obtains the qualified inspection record as the endpoint. The endpoint impedance modulus of the current qualified standard sample at each disturbance frequency is transformed into a dimensionless ratio with the pre-plating reference impedance modulus of the standard sample at the corresponding disturbance frequency. The natural logarithm of the transformation result is then used to calculate the standard modulus offset of the current qualified standard sample at each disturbance frequency. Finally, the endpoint phase of the current qualified standard sample at each disturbance frequency is subtracted from the pre-plating reference phase of the standard sample at the corresponding disturbance frequency, and then dimensionlessized according to the circumferential angle reference to obtain the standard modulus offset of the current qualified standard sample at each disturbance frequency. The standard phase offset is calculated; according to the perturbation frequency numbering order, the standard modulus offset and standard phase offset of the current qualified standard sample at each perturbation frequency are alternately arranged to form the endpoint fingerprint vector of the current qualified standard sample; the arithmetic mean of each component of the endpoint fingerprint vector of all qualified standard samples is calculated to form the qualified endpoint center fingerprint of all qualified standard sample endpoint fingerprints, and the dimension of the qualified endpoint center fingerprint is set to twice the total number of perturbation frequencies; for each qualified standard sample, the standard modulus offset of the current qualified standard sample is compared with the modulus offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint, and the difference is normalized according to the result of adding one to the current modulus offset center value. The modulus deviation component at the current perturbation frequency is generated. For each qualified standard sample, the standard phase offset of the current qualified standard sample is compared with the phase offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint at each perturbation frequency. The difference is normalized by adding one to the current phase offset center value to generate the phase deviation component at the current perturbation frequency. For each qualified standard sample, the modulus deviation component and the phase deviation component at all perturbation frequencies are summed and averaged at twice the total number of perturbation frequencies to calculate the standard deviation of the current qualified standard sample relative to the qualified endpoint center fingerprint. The maximum value is selected from the standard deviations of all qualified standard samples to form the standard sample fingerprint dispersion.The instrument impedance modulus resolution and instrument phase resolution are obtained. The pre-plating reference impedance modulus of the current workpiece at each disturbance frequency is added to the instrument impedance modulus resolution. This sum is then compared to the pre-plating reference impedance modulus of the current workpiece at the same disturbance frequency. The ratio of this ratio is transformed, and the natural logarithm of the ratio transformation result is performed to form the modulus resolution deviation term at each disturbance frequency. The modulus resolution deviation term after natural logarithmic operation at each disturbance frequency is summed with the instrument phase resolution over the disturbance frequency, and then averaged at twice the total number of disturbance frequencies to calculate the instrument resolution deviation caused by the instrument impedance modulus resolution and instrument phase resolution. The fingerprint dispersion of the standard sample is compared with the instrument resolution deviation, and the maximum value between the fingerprint dispersion of the standard sample and the instrument resolution deviation is selected as the allowable deviation at the endpoint.

[0018] Optionally, the step of forming online thickness records, online antimony content records, and online grain records based on the multi-frequency impedance fingerprint of the current plated part specifically includes:

[0019] For the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the impedance modulus offset at each perturbation frequency is extracted; the impedance modulus offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the thickness-related projection value corresponding to the current complete multi-frequency acquisition; for the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the phase offset at each perturbation frequency is extracted; the phase offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the antimony content-related projection value corresponding to the current complete multi-frequency acquisition; the impedance modulus offset difference between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency numbering order ... thickness-related projection value corresponding to the current complete multi-frequency acquisition; the impedance modulus offset difference between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency numbering order; the phase offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the thickness-related projection value corresponding to the current complete multi-frequency acquisition; the impedance modulus offset difference between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency numbering order; the phase offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the thickness-related projection value corresponding to the current complete multi-frequency acquisition; the impedance modulus offset difference between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency numbering order; the impedance modulus offset difference between The impedance modulus offset differences between adjacent perturbation frequencies are squared, summed, and then averaged according to the total number of perturbation frequencies minus one to form the grain-scale related projection quantity corresponding to the current complete multi-frequency acquisition. The offline thickness inspection value, offline antimony mass fraction inspection value, and offline grain-scale inspection value of each qualified standard sample are obtained. The offline thickness inspection values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average thickness of the qualified standard samples. The offline antimony mass fraction inspection values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average antimony mass fraction of the qualified standard samples. The offline grain size test values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average grain size of qualified standard samples. For the center fingerprint of the qualified endpoint, the modulus offset center value at each perturbation frequency is extracted, summed, and averaged according to the total number of perturbation frequencies to form the thickness-related projection of the center of the qualified endpoint. For the center fingerprint of the qualified endpoint, the phase offset center value at each perturbation frequency is extracted, summed, and averaged according to the total number of perturbation frequencies to form the antimony content-related projection of the center of the qualified endpoint. The central fingerprint is calculated by sequentially determining the difference in modulus offset center value between adjacent perturbation frequencies according to their frequency numbers. The difference is then squared and summed, and averaged using the total number of perturbation frequencies minus one, resulting in the grain-scale related projection quantity of the qualified endpoint center. The thickness-related projection quantity corresponding to the current complete multi-frequency acquisition is then multiplied by one, and the ratio of this result to the thickness-related projection quantity of the qualified endpoint center is transformed to form the online thickness ratio transformation result. This online thickness ratio transformation result is then multiplied by the average thickness of the qualified standard sample to form the online thickness record quantity corresponding to the current complete multi-frequency acquisition.The online antimony content ratio transformation result is obtained by adding one to the antimony content related projection value corresponding to the current complete multi-frequency acquisition and then performing a ratio transformation with the same result as adding one to the antimony content related projection value at the center of the qualified endpoint. This online antimony content ratio transformation result is then multiplied by the average antimony mass fraction of the qualified standard sample to obtain the online antimony content record value corresponding to the current complete multi-frequency acquisition. Similarly, the online grain ratio transformation result is obtained by adding one to the grain size related projection value at the center of the qualified endpoint and then performing a ratio transformation with the same result as adding one to the grain size related projection value at the current complete multi-frequency acquisition. This online grain ratio transformation result is then multiplied by the average grain size of the qualified standard sample to obtain the online grain record value corresponding to the current complete multi-frequency acquisition.

[0020] Optionally, the step of comparing the multi-frequency impedance fingerprint of the current part to be plated with the qualified endpoint center fingerprint by perturbation frequency, calculating the endpoint deviation, and performing the plating termination determination based on the endpoint deviation and the longest allowable plating time specifically includes:

[0021] The multi-frequency impedance fingerprint corresponding to the current complete multi-frequency acquisition is compared with the qualified endpoint center fingerprint for each perturbation frequency. For each perturbation frequency, the impedance modulus offset corresponding to the current complete multi-frequency acquisition is subtracted from the modulus offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized by adding one to the current modulus offset center value to form the modulus endpoint deviation component at the current perturbation frequency. For each perturbation frequency, the phase offset corresponding to the current complete multi-frequency acquisition is subtracted from the phase offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized by adding one to the current phase offset center value to form the current phase offset offset component. The phase endpoint deviation component at the previous perturbation frequency is calculated; the modulus endpoint deviation component and the phase endpoint deviation component at all perturbation frequencies are summed and averaged at twice the total number of perturbation frequencies to calculate the endpoint deviation corresponding to the current complete multi-frequency acquisition; when the endpoint deviation is not greater than the allowable endpoint deviation, the high-hardness gold-antimony alloy plating is recorded as reaching the non-destructive online endpoint; when the endpoint deviation is greater than the allowable endpoint deviation, the DC electroplating main current is maintained, and the endpoint deviation corresponding to the next complete multi-frequency acquisition is calculated after the next multi-frequency perturbation acquisition is completed; the longest allowable electroplating time given by the process card is obtained, and the completion time of each complete multi-frequency acquisition is recorded; when there is a completion When a complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the allowed endpoint deviation occurs, the endpoint status is recorded as "reached endpoint state". If no complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the allowed endpoint deviation occurs, the endpoint status is recorded as "not reached endpoint state". When the endpoint status is "reached endpoint state", the earliest completion time of any complete multi-frequency acquisition satisfying the condition of "completion time no later than the longest allowed electroplating time and endpoint deviation no greater than the allowed endpoint deviation" is selected as the acquisition index. When the endpoint status is "not reached endpoint state", the earliest completion time of any complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time is selected as the acquisition index. In the multi-frequency acquisition process, the latest completion time of the complete multi-frequency acquisition is selected as the acquisition index. When the endpoint state is reached, the completion time of the complete multi-frequency acquisition corresponding to the acquisition index is taken as the electroplating stop time. When the endpoint state is not reached, the longest allowable electroplating time is taken as the electroplating stop time. The endpoint deviation corresponding to the acquisition index is taken as the stop deviation. At the electroplating stop time, the DC electroplating main current and the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current are cut off, and the workpiece is removed from the gold-antimony alloy deposition state. When the endpoint state is reached, the endpoint state is recorded. When the endpoint state is not reached, the endpoint state is recorded.

[0022] Optionally, the step of cutting off the sinusoidal disturbance current corresponding to the amplitude of the DC electroplating main current and the AC disturbance current at the electroplating stop time, generating a stop data group, calculating the stop charge, and outputting the process stop result specifically includes: forming a stop data group using complete multi-frequency acquisition data corresponding to the recording acquisition index; and writing the electroplating stop time, end point state quantity, stop deviation quantity, online thickness record quantity, online antimony content record quantity, online grain record quantity, and multi-frequency impedance fingerprint vector into the stop data group.

[0023] Obtain the electroplating start time and the DC electroplating main current; subtract the electroplating start time from the electroplating stop time to form the electroplating duration; multiply the DC electroplating main current by the electroplating duration to form the stop charge; output the process stop result, which includes the stop deviation, endpoint state, electroplating stop time, stop charge, online thickness record, online antimony content record, and online grain record.

[0024] The present invention has the following beneficial effects:

[0025] 1. The conductive surface of the workpiece to be electroplated, which actually participates in the deposition of the gold-antimony alloy, is used as the effective cathode surface. The amplitudes of the DC electroplating main current and AC disturbance current are determined based on the effective cathode area. This design avoids the crude control method of setting the electroplating current solely according to the workpiece's external dimensions, nominal area, or equipment default parameters, ensuring that the electroplating main current matches the actual surface area involved in deposition. This reduces current density deviations caused by variations in local non-conductive areas, shielded areas, or actual contact areas of the workpiece. It also ensures that the disturbance intensity of subsequent multi-frequency disturbance acquisition maintains a reasonable ratio with the main current, preventing excessive disturbance from affecting the deposition process and excessively small disturbance from resulting in an indistinct response signal. This scheme unifies the electroplating main process and the online detection process under the same workpiece reference, enhancing the compatibility between workpieces of different specifications.

[0026] 2. Before the gold-antimony alloy deposition begins, a pre-deposition reference perturbation is performed on the same workpiece, and the pre-deposition reference impedance modulus and phase are obtained through a multi-frequency perturbation sequence. This scheme does not directly use the online measured impedance data as the endpoint determination criterion, but first establishes an initial electrochemical response reference for the workpiece itself, allowing subsequent data to be evaluated in a relative change manner. This reduces the impact of differences in the workpiece substrate, surface pretreatment, initial oxidation state, fixture contact state, and reference end arrangement on endpoint determination. The multi-frequency perturbation setting allows this scheme to simultaneously observe interface response characteristics at different frequencies, avoiding the situation where a single frequency only reflects a local or partial state. Through effective range determination, abnormal re-acquisition, and abnormal stop mechanisms, this scheme also improves data reliability and prevents abnormal sampling data from entering the subsequent fingerprint construction process. This scheme gives each workpiece its own starting point reference, thereby improving the stability and anti-interference capability of online endpoint identification.

[0027] 3. During the actual deposition of gold-antimony alloy, the online voltage response between the cathode and the reference terminal is continuously acquired according to a multi-frequency perturbation sequence. The amplitude and phase of the online voltage response are extracted through sine and cosine projection. This scheme transforms the dynamic interface changes during electroplating into calculable and comparable response characteristics, making the identification of the electroplating endpoint real-time and process-dependent. The online voltage response amplitude reflects the changes in interface impedance during coating growth, while the online voltage response phase reflects changes in the interface electrochemical process, deposition state, and response hysteresis characteristics. The combination of these two factors is more sensitive than simply measuring the tank voltage or current. The scheme also includes sampling integrity judgment and abnormal re-acquisition procedures to eliminate unreliable data caused by exceeding the measurement range, signal abnormalities, or transient interference. This scheme provides continuous, non-destructive, and online process sensing capabilities, enabling timely capture of electrochemical response changes as the coating approaches its endpoint, providing high-quality input for multi-frequency impedance fingerprint construction.

[0028] 4. The online impedance modulus and the pre-plating reference impedance modulus are transformed into dimensionless values, and the online phase and the pre-plating reference phase are made dimensionless. Then, they are combined according to the perturbation frequency sequence to form a multi-frequency impedance fingerprint vector. This design does not simply record the absolute value of the online impedance, but extracts the amplitude and direction of change of the current deposition state relative to the pre-plating state, forming a comprehensive fingerprint that represents the current coating state. This reduces the influence of initial differences in the workpiece and static deviations of the system on the judgment results, making the fingerprint more concentrated in reflecting the true evolution of the gold-antimony alloy coating during the deposition process. The multi-frequency impedance fingerprint simultaneously includes modulus changes and phase changes at different frequencies, and compared to a single index, it can better cover complex information such as coating thickness growth, alloy co-deposition, surface densification, and changes in grain state. This scheme forms multi-dimensional state characteristics, which can improve the ability to distinguish complex coating quality states and is suitable for electroplating systems such as high-hardness gold-antimony alloys that have requirements for both composition and microstructure.

[0029] 5. An endpoint fingerprint is established using multiple qualified high-hardness gold-antimony alloy standard samples. These endpoint fingerprints are then centrally processed to form a qualified endpoint center fingerprint. Simultaneously, the allowable deviation of the endpoint is established by considering the dispersion among the standard samples and the instrument's resolution. This design transforms the endpoint judgment criterion from an empirical threshold to a quality boundary formed by actual data from a group of qualified samples. The qualified endpoint center fingerprint represents the endpoint state of the coating that conforms to the process inspection specifications. The allowable endpoint deviation provides a reasonable tolerance for normal sample differences and the instrument's measurement resolution, avoiding misjudgment due to minor measurement fluctuations. This scheme also addresses the issue of natural fluctuations in the qualified endpoint state between different batches and different coated parts, ensuring that the endpoint judgment is neither too lenient, leading to the release of unqualified coatings, nor too strict, leading to the misjudgment of qualified coatings.

[0030] 6. The multi-frequency impedance fingerprint is further converted into online records related to thickness, antimony content, and grain size, rather than simply outputting a single judgment result indicating whether the endpoint has been reached. This scheme utilizes impedance modulus shift, phase shift, and modulus shift differences between adjacent perturbation frequencies to form thickness-related projections, antimony content-related projections, and grain size-related projections, respectively, and combines these with offline inspection values ​​of qualified standard samples for online record conversion. This allows the online electrochemical response to not only serve endpoint determination but also form a process record of key coating quality indicators. For high-hardness gold-antimony alloy coatings, insufficient thickness affects wear resistance and lifespan, antimony content deviation affects hardness and alloy properties, and grain size changes affect density and surface quality. This scheme establishes a process quality profile through online records, which is beneficial for tracking quality trends without damaging the coated part, improving batch consistency, process traceability, and anomaly analysis capabilities.

[0031] 7. The multi-frequency impedance fingerprint of the current part to be plated is compared with the center fingerprint of the qualified endpoint by perturbation frequency to form the endpoint deviation. This deviation is then combined with the allowable endpoint deviation and the maximum allowable plating time to determine termination. This scheme does not simply judge whether the plating time has been reached, nor does it judge whether a single parameter has exceeded a threshold. Instead, it comprehensively compares the overall closeness between the current plating state and the qualified endpoint state. When the endpoint deviation enters the allowable range, plating can be stopped in time to reduce excessive deposition of precious metals and process costs. When the endpoint deviation does not enter the allowable range, plating continues and the next multi-frequency sampling is performed to reduce the risk of under-plating. When the maximum allowable plating time is reached but the conditions are still not met, a record of failure to reach the endpoint is generated according to the boundary conditions to prevent the abnormal expansion caused by indefinitely extending the plating process. Compared with existing fixed-duration or fixed-charge control methods, this scheme has stronger state awareness and abnormal protection capabilities, and can simultaneously ensure quality compliance, cost control, and equipment safety.

[0032] 8. At the moment of electroplating stoppage, not only are the DC electroplating main current and sinusoidal disturbance current cut off, but the system also records and collects complete multi-frequency acquisition data, endpoint status quantities, stop deviation quantities, online thickness records, online antimony content records, online grain records, and multi-frequency impedance fingerprint vectors to form a stop data set. Furthermore, it calculates the stop charge and outputs the process stop result. This scheme transforms electroplating stoppage from a mere control action into a simultaneous creation of complete quality records and process documentation. The production end can trace the quality status of the plated part at the moment of stoppage through the stop data set. The quality end can determine whether the part reached the endpoint normally or stopped due to the longest permissible time based on the endpoint status quantity and stop deviation quantity. The process end can analyze electroplating efficiency, coating growth stability, and causes of anomalies by combining the stop charge quantity and online records. For high-value gold-antimony alloy coatings, complete stop records help reduce controversial judgments and improve the objectivity and verifiability of product release criteria. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the process of the present invention.

[0034] Figure 2 This is a schematic diagram of the multi-frequency disturbance and pre-plating reference establishment process of the present invention.

[0035] Figure 3 This is a schematic diagram of the online voltage response acquisition and amplitude-phase extraction process of the present invention.

[0036] Figure 4 This is a schematic diagram of the multi-frequency impedance fingerprint construction process of the present invention.

[0037] Figure 5 This is a schematic diagram of the process for calculating the endpoint deviation and determining the electroplating termination of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Example, refer to Figure 1 A process for electroplating high-hardness gold-antimony alloy, comprising:

[0040] Establish the effective cathode surface of the workpiece to be electroplated, obtain the effective cathode area, set the DC current density for the gold-antimony alloy electroplating stage, and calculate the DC electroplating main current and AC disturbance current amplitude according to the effective cathode area and DC current density.

[0041] Establish a multi-frequency disturbance frequency sequence and a single-frequency sampling window, perform pre-plating reference disturbance acquisition, and calculate the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response;

[0042] During the gold-antimony alloy deposition process, the online voltage response between the cathode and the reference terminal was acquired according to the multi-frequency perturbation sequence, and the amplitude and phase of the online voltage response were extracted.

[0043] Based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase, construct the multi-frequency impedance fingerprint of the current part to be plated;

[0044] Prepare qualified high-hardness gold-antimony alloy standard samples, collect the endpoint impedance modulus and endpoint phase of each qualified standard sample, and establish the qualified endpoint center fingerprint and the allowable deviation of the endpoint.

[0045] Based on the multi-frequency impedance fingerprint of the plated parts, online thickness recording, online antimony content recording, and online grain size recording are generated;

[0046] The multi-frequency impedance fingerprint of the current part to be plated is compared with the qualified endpoint center fingerprint by perturbation frequency to calculate the endpoint deviation. The electroplating termination judgment is then performed based on the endpoint deviation and the longest allowable electroplating time.

[0047] At the moment of electroplating stop, the DC electroplating main current and the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current are cut off, a stop data set is generated, the stop charge is calculated, and the process stop result is output.

[0048] By first establishing an effective cathode surface and pre-plating reference for the workpiece itself, and then conducting multi-frequency perturbation acquisition during the gold-antimony alloy deposition process, and comparing the current multi-frequency impedance fingerprint with the qualified endpoint center fingerprint, this solves the problem that existing electroplating processes mainly rely on fixed electroplating time, empirical current efficiency, or offline sampling to determine the endpoint, making it difficult to reflect the true deposition state of the coating in a timely manner. The endpoint determination no longer relies solely on external empirical parameters but is correlated with the real-time electrochemical response of the workpiece during the electroplating process, allowing for a more objective assessment of whether the coating is close to a qualified state. Simultaneously, this claim also sets online thickness recording, online antimony content recording, and online grain recording, enabling the synchronous generation of quality records when electroplating stops. Compared to the delayed detection method after electroplating in existing technologies, this scheme forms endpoint judgment and process records during the electroplating process, helping to reduce the risks of overplating, underplating, and rework, and improving the controllability and traceability of the high-hardness gold-antimony alloy electroplating process.

[0049] The process of establishing an effective cathode surface for the workpiece to be electroplated, obtaining the effective cathode area, setting the DC current density for the gold-antimony alloy electroplating stage, and calculating the DC electroplating main current and AC disturbance current amplitude based on the effective cathode area and DC current density specifically includes:

[0050] The conductive surface of the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is taken as the effective cathode surface, and the effective cathode area of ​​the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is measured.

[0051] The DC current density during the gold-antimony alloy electroplating stage is set as the calculation parameter for the main electroplating current.

[0052] The DC electroplating main current for the gold-antimony alloy electroplating stage is calculated by multiplying the effective cathode area by the DC current density.

[0053] The DC electroplating main current is converted to one-fiftieth of its original value to obtain the amplitude of the AC disturbance current superimposed during multi-frequency impedance acquisition.

[0054] The conductive surface of the workpiece to be electroplated that actually participates in the deposition of the gold-antimony alloy is defined as the effective cathode surface. The effective cathode area is measured and recorded as follows: ;in, The effective cathode area that actually participates in the deposition of gold-antimony alloy in the workpiece to be electroplated;

[0055] The DC current density during the gold-antimony alloy electroplating stage is set to... ;

[0056] The electroplating main current is calculated based on the effective cathode area and DC current density as follows:

[0057] ;in, This refers to the DC electroplating main current during the gold-antimony alloy electroplating stage.

[0058] Set the AC disturbance current amplitude to: ;in, This represents the amplitude of the AC disturbance current superimposed during multi-frequency impedance acquisition.

[0059] Reference Figure 2 The process of establishing a multi-frequency disturbance sequence and a single-frequency sampling window, performing pre-plating reference disturbance acquisition, and calculating the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response specifically includes:

[0060] Establish a set of multi-frequency disturbance frequency numbers, set the total number of disturbance frequencies to be no less than five, and configure a corresponding disturbance frequency number index for each disturbance frequency;

[0061] Set the lowest and highest frequencies in the multi-frequency perturbation frequency sequence, and make the lowest frequency greater than zero and less than the highest frequency;

[0062] Based on the geometric interval relationship from the lowest frequency to the highest frequency, the perturbation frequencies corresponding to each perturbation frequency number index are generated sequentially.

[0063] Set the number of voltage sampling points at each disturbance frequency to be no less than sixteen, and establish a single-frequency sampling number set;

[0064] Based on the ratio between the sampling number and the number of voltage sampling points, and combined with the current disturbance frequency, the sampling time corresponding to each sampling number under the current disturbance frequency is generated;

[0065] During the gold-antimony alloy deposition stage, the DC electroplating main current is superimposed with the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current, and an online single-frequency sinusoidal disturbance current is formed according to the current disturbance frequency.

[0066] Obtain the effective lower limit and effective upper limit of the voltage acquisition device, and ensure that the effective lower limit is less than the effective upper limit;

[0067] Before the gold-antimony alloy deposition begins, the DC electroplating main current is kept off. Only the AC disturbance current amplitude is used to perform a multi-frequency disturbance acquisition on the same workpiece according to the multi-frequency disturbance sequence to form the pre-plating reference disturbance current at each disturbance frequency.

[0068] At each perturbation frequency, the pre-plating voltage response is collected according to the corresponding sampling time.

[0069] When all pre-plating sampled voltages at the current disturbance frequency are between the lower limit and the upper limit of the effective range of the voltage acquisition device, all pre-plating sampled voltages at the current disturbance frequency are retained.

[0070] If the pre-plating sampling voltage is lower than the lower limit of the effective range or higher than the upper limit of the effective range at the current disturbance frequency, the pre-plating voltage response data at the current disturbance frequency will be re-acquired.

[0071] When all the pre-plating sampled voltages after re-acquisition are within the lower limit and upper limit of the effective range of the voltage acquisition device, all the pre-plating sampled voltages obtained after re-acquisition are retained.

[0072] If, after re-collection, the pre-plating sampling voltage is still lower than the lower limit of the effective range or higher than the upper limit of the effective range, the endpoint determination of the current plated part is stopped, the abnormal state of the pre-plating reference is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0073] The pre-plating voltage response under the current disturbance frequency is sinusoidally projected. Each pre-plating sampled voltage is multiplied by the sinusoidal reference value at the corresponding sampling time and summed. The result is then normalized according to the number of voltage sampling points to form the sinusoidal projection component of the pre-plating voltage response under the current disturbance frequency.

[0074] Cosine projection is performed on the pre-plating voltage response at the current disturbance frequency. The pre-plating sampled voltage is multiplied by the cosine reference value at the corresponding sampling time and summed. The result is then normalized according to the number of voltage sampling points to form the cosine projection component of the pre-plating voltage response at the current disturbance frequency.

[0075] The square of the sinusoidal projection component and the square of the cosine projection component of the pre-plating voltage response at the current disturbance frequency are added together, and the square root of the sum is taken to calculate the amplitude of the pre-plating voltage response at the current disturbance frequency.

[0076] When the pre-plating voltage response amplitude is greater than zero, perform a binary arctangent operation based on the cosine projection component and sine projection component of the pre-plating voltage response at the current disturbance frequency to calculate the pre-plating reference phase at the current disturbance frequency. Divide the pre-plating voltage response amplitude by the AC disturbance current amplitude to calculate the pre-plating reference impedance modulus at the current disturbance frequency.

[0077] When the pre-plating voltage response amplitude is zero, stop the endpoint determination of the current plated part, record the pre-plating reference abnormal state, and do not generate the multi-frequency impedance fingerprint of the current plated part.

[0078] The set of multi-frequency disturbance numbers is established as follows: ;in, This is a set of frequency numbers for multi-frequency disturbances. This represents the total number of disturbance frequencies; Indexed by frequency number;

[0079] The lowest frequency in the multi-frequency perturbation sequence is denoted as . The highest frequency in the multi-frequency perturbation sequence is denoted as and satisfy ;

[0080] Generate the first The perturbation frequency is ;

[0081] Acquisition at each frequency There are voltage sampling points, and they satisfy the following conditions: ;in, The number of voltage sampling points at each frequency;

[0082] Establish a single-frequency sampling number set as ;

[0083] Construct the first At the frequency of the first Each sampling time is ;in, For sampling number index;

[0084] The online single-frequency sinusoidal perturbation current function for the gold-antimony alloy deposition stage is established as follows:

[0085] ;in, For the first At a frequency at continuous time The online instantaneous current; For continuous time variables; This is the main current for DC electroplating;

[0086] The lower limit of the effective range of the voltage acquisition device is denoted as... The upper limit of the effective range is denoted as and satisfy ;

[0087] Maintain DC plating current before gold-antimony alloy deposition begins. Not connected; only AC disturbance current amplitude is used. A multi-frequency disturbance acquisition is performed on the same workpiece to be plated according to the frequency sequence, wherein the pre-plating reference disturbance current is: ;in, For the first At a frequency at continuous time The pre-plating reference disturbance current;

[0088] In the At each frequency, at the sampling time The voltage response before plating was collected and denoted as... ;

[0089] When all pre-plating sampling voltages meet At that time, the first one is reserved. All pre-plating sampled voltages at each frequency;

[0090] When the pre-plating sampling voltage does not meet the requirements At that time, re-collect the first Pre-plating voltage response data at each frequency;

[0091] After re-collection, all pre-plating sampling voltages meet the requirements. At that time, all pre-plating sampling voltages obtained by re-acquiring data are retained;

[0092] Even after re-collection, the pre-plating sampling voltage still does not meet the requirements. When the current plated part is stopped from determining the end point, the pre-plating reference abnormal state is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0093] The pre-plating voltage response is subjected to sinusoidal and cosine projections, specifically as follows:

[0094] ,

[0095] ;in, For the first The sinusoidal projection component of the pre-plating voltage response at a given frequency; For the first Cosine projection component of the pre-plating voltage response at a given frequency.

[0096] Set the pre-plating voltage amplitude to: ;in, For the first The amplitude of the pre-plating voltage response at each frequency;

[0097] when At that time, the reference phase and reference impedance modulus before plating are calculated as follows:

[0098] , ;in, For the first Pre-plating reference phase at a specific frequency; For the first Reference impedance modulus before plating at a given frequency; It is a bivariate arctangent function;

[0099] when When the time is reached, stop determining the endpoint of the current plated part, record the abnormal state of the pre-plating reference, and do not generate a multi-frequency impedance fingerprint of the current plated part.

[0100] Reference Figure 3 The process of acquiring the online voltage response between the cathode and the reference terminal according to a multi-frequency perturbation sequence during the gold-antimony alloy deposition, and extracting the online voltage response amplitude and phase, specifically includes:

[0101] Establish a complete multi-frequency acquisition count index, and record each complete multi-frequency acquisition process according to the complete multi-frequency acquisition count index;

[0102] During the disturbance period of the current disturbance frequency in any complete multi-frequency acquisition, the voltage between the cathode and the reference terminal is acquired at each sampling moment under the current disturbance frequency to form an online sampling voltage;

[0103] Integrity determination is performed on the online sampled voltage using the lower and upper limits of the effective range of the voltage acquisition device.

[0104] When all online sampled voltages at the current disturbance frequency in any complete multi-frequency acquisition are between the lower limit and the upper limit of the effective range, all online sampled voltages at the current disturbance frequency in the current complete multi-frequency acquisition are retained.

[0105] If, during any complete multi-frequency acquisition, the online sampling voltage at the current disturbance frequency is lower than the lower limit of the effective range or higher than the upper limit of the effective range, the voltage response data at the current disturbance frequency in the current complete multi-frequency acquisition will be reacquired.

[0106] When all online sampled voltages after re-acquisition are between the lower limit and the upper limit of the effective range, all online sampled voltages obtained after re-acquisition are retained;

[0107] If, after re-collection, the online sampling voltage is still below the lower limit of the effective range or above the upper limit of the effective range, the endpoint determination of the current plated part is stopped, the abnormal sampling state is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0108] For any complete multi-frequency acquisition, perform sinusoidal projection on all effective voltage sample values ​​at the current disturbance frequency, multiply each effective voltage sample value by the sinusoidal reference value at the corresponding sampling time, sum the results, and normalize according to the number of voltage sampling points to form online voltage sinusoidal projection components.

[0109] Cosine projection is performed on all effective voltage sample values ​​at the current disturbance frequency in any complete multi-frequency acquisition. Each effective voltage sample value is multiplied by the cosine reference value at the corresponding sampling time and summed. Then, the values ​​are normalized according to the number of voltage sampling points to form online voltage cosine projection components.

[0110] The square value of the sinusoidal projection component of the online voltage is added to the square value of the cosine projection component of the online voltage, and the square root of the sum is taken to calculate the amplitude of the online voltage response at the current disturbance frequency in the current complete multi-frequency acquisition.

[0111] When the amplitude of the online voltage response is greater than zero, the phase of the online voltage response at the current disturbance frequency in the current complete multi-frequency acquisition is calculated by performing a binary arctangent operation based on the cosine projection component and the sine projection component of the online voltage.

[0112] When the online voltage response amplitude is zero, stop the endpoint determination of the current plated part, record the abnormal state of the impedance signal, and do not generate a multi-frequency impedance fingerprint of the current plated part.

[0113] The index of the complete multi-frequency acquisition count is denoted as... ;

[0114] In the The first complete multi-frequency acquisition During each frequency disturbance, at each sampling time The voltage between the cathode and the reference terminal is collected and denoted as . ;

[0115] Lower effective range limit of voltage acquisition equipment and the upper limit of the effective range Perform online sampling integrity determination:

[0116] When all sampled voltages satisfy At that time, the first one is reserved. The first complete multi-frequency acquisition All sampled voltages at each frequency;

[0117] When the sampling voltage does not meet the requirements At that time, re-collect the first The first complete multi-frequency acquisition Voltage response data at each frequency;

[0118] After re-acquisition, all sampled voltages meet the requirements. At that time, all sampled voltages obtained from the re-acquisition will be retained;

[0119] Even after re-acquisition, the sampling voltage still does not meet the requirements. When the current plated part is stopped, the endpoint determination is stopped, the abnormal sampling state is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0120] For the The first complete multi-frequency acquisition All effective voltage samples at each frequency are subjected to sine and cosine projections, respectively, as follows:

[0121] ,

[0122] ;in, , The first The first complete multi-frequency acquisition The sinusoidal and cosine projection components of the online voltage at each frequency;

[0123] The two projected components are combined into the voltage response amplitude, specifically:

[0124] ;in, For the first The first complete multi-frequency acquisition The amplitude of the online voltage response at each frequency;

[0125] when When the voltage response phase is calculated, it is: ;in, For the first The first complete multi-frequency acquisition The online voltage response phase at each frequency;

[0126] when When the current plated part is stopped, the endpoint determination is stopped, the abnormal state of the impedance signal is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated.

[0127] Reference Figure 4 The construction of a multi-frequency impedance fingerprint of the workpiece to be plated based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase specifically includes:

[0128] Divide the online voltage response amplitude at the current disturbance frequency in the current complete multi-frequency acquisition by the AC disturbance current amplitude to calculate the online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition.

[0129] The online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition is transformed into a dimensionless ratio with the pre-plating reference impedance modulus at the corresponding disturbance frequency. The natural logarithm of the ratio transformation result is then used to form the impedance modulus offset at the current disturbance frequency in the current complete multi-frequency acquisition.

[0130] The phase offset at the current disturbance frequency is obtained by subtracting the online voltage response phase at the current disturbance frequency from the pre-plating reference phase at the corresponding disturbance frequency in the current complete multi-frequency acquisition, and then performing dimensionless processing according to the circumferential angle reference.

[0131] According to the perturbation frequency numbering order, the impedance modulus offset and phase offset at each perturbation frequency obtained from the current complete multi-frequency acquisition are arranged alternately to form the multi-frequency impedance fingerprint vector corresponding to the current complete multi-frequency acquisition.

[0132] Set the dimension of the multi-frequency impedance fingerprint vector to twice the total number of perturbation frequencies.

[0133] The impedance modulus is calculated using the voltage response amplitude and the disturbance current amplitude as follows: ;in, For the first The first complete multi-frequency acquisition Online impedance modulus at each frequency;

[0134] The impedance modulus during the electroplating process is transformed into a dimensionless ratio with the reference impedance modulus before plating, specifically as follows: ;in, For the first The first complete multi-frequency acquisition Impedance modulus offset at each frequency;

[0135] The phase during the electroplating process is made dimensionless relative to the reference phase before plating, specifically as follows:

[0136] ;in, For the first The first complete multi-frequency acquisition Phase offset at each frequency;

[0137] The first All frequencies obtained from a complete multi-frequency acquisition and The arrangement is a multi-frequency impedance fingerprint vector, specifically: ;in, For the first The multi-frequency impedance fingerprint vector formed by a complete multi-frequency acquisition has a dimension of [missing information]. .

[0138] The preparation of qualified high-hardness gold-antimony alloy standard samples involves collecting the endpoint impedance modulus and endpoint phase of each qualified standard sample, establishing the qualified endpoint center fingerprint and the allowable endpoint deviation, specifically including:

[0139] Using the same substrate shape, effective cathode area, and DC current density as the workpiece to be electroplated, prepare qualified high-hardness gold-antimony alloy standard samples, and set the number of qualified standard samples to be no less than three.

[0140] Establish a set of qualified standard samples with qualified inspection records formed by process inspection specifications as qualified standard sample number set, and configure a qualified standard sample number index for each qualified standard sample;

[0141] For each qualified standard sample, the same multi-frequency disturbance frequency sequence, sampling time, AC disturbance current amplitude, voltage projection process and impedance modulus calculation process as the part to be plated are used to establish the standard sample pre-plating reference impedance modulus and standard sample pre-plating reference phase at each disturbance frequency for the current qualified standard sample before the gold-antimony alloy deposition begins.

[0142] At the qualified endpoint, the current qualified standard sample is collected at each disturbance frequency to obtain the endpoint impedance modulus and endpoint phase, and the electroplating stop time corresponding to the current qualified standard sample when the qualified inspection record is obtained is taken as the qualified endpoint.

[0143] The current qualified standard sample at each disturbance frequency is subjected to a dimensionless ratio transformation of the endpoint impedance modulus at each disturbance frequency and the reference impedance modulus of the standard sample before plating at the corresponding disturbance frequency. The natural logarithm of the ratio transformation result is then used to form the standard modulus offset of the current qualified standard sample at each disturbance frequency.

[0144] Subtract the pre-plating reference phase of the standard sample at the corresponding perturbation frequency from the endpoint phase of the current qualified standard sample at each perturbation frequency, and perform dimensionless processing according to the circumferential angle reference to form the standard phase offset of the current qualified standard sample at each perturbation frequency.

[0145] According to the perturbation frequency numbering order, the standard modulus offset and standard phase offset of the current qualified standard sample at each perturbation frequency are arranged alternately to form the endpoint fingerprint vector of the current qualified standard sample.

[0146] The arithmetic mean of the endpoint fingerprint vectors of all qualified standard samples is calculated for each component to form the qualified endpoint center fingerprint of all qualified standard samples, and the dimension of the qualified endpoint center fingerprint is set to twice the total number of perturbation frequencies.

[0147] For each qualified standard sample, the standard modulus offset of the current qualified standard sample is compared with the modulus offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint at each perturbation frequency. The difference is then normalized according to the result of adding one to the current modulus offset center value to form the modulus deviation component at the current perturbation frequency.

[0148] For each qualified standard sample, the standard phase offset of the current qualified standard sample is compared with the phase offset center value of the corresponding perturbation frequency in the center fingerprint of the qualified endpoint at each perturbation frequency. The difference is then normalized according to the result of adding one to the current phase offset center value to form the phase deviation component at the current perturbation frequency.

[0149] For each qualified standard sample, the modulus deviation component and phase deviation component at all disturbance frequencies are summed and averaged at twice the total number of disturbance frequencies to calculate the standard deviation of the current qualified standard sample relative to the center fingerprint of the qualified endpoint.

[0150] The maximum value is selected from the standard deviations of all qualified standard samples to form the fingerprint dispersion of the standard samples.

[0151] Obtain the instrument impedance modulus resolution and instrument phase resolution. Add the pre-plating reference impedance modulus of the current part to be plated at each disturbance frequency to the instrument impedance modulus resolution, and then perform a ratio transformation with the pre-plating reference impedance modulus of the current part to be plated at the same disturbance frequency. Perform a natural logarithmic operation on the ratio transformation result to form the modulus resolution deviation term at each disturbance frequency.

[0152] The modulus resolution deviation term calculated by natural logarithm at each perturbation frequency is summed with the instrument phase resolution according to the perturbation frequency, and then averaged according to twice the total number of perturbation frequencies to calculate the instrument resolution deviation caused by the instrument impedance modulus resolution and the instrument phase resolution.

[0153] The fingerprint dispersion of the standard sample is compared with the instrument resolution deviation, and the maximum value between the fingerprint dispersion of the standard sample and the instrument resolution deviation is selected as the allowable deviation of the endpoint.

[0154] Using the same substrate shape, effective cathode area, and electroplating main current density as the workpiece to be electroplated, prepare... One high-hardness gold-antimony alloy standard sample; among them... The number of qualified standard samples;

[0155] The standard samples that have formed qualified inspection records according to process inspection specifications will be established as a set of standard sample numbers. ;in, Index for qualified standard sample numbers;

[0156] For the For each qualified standard sample, using the same frequency sequence, sampling time, AC disturbance amplitude, voltage projection, and impedance modulus calculation process as the workpiece to be plated, a pre-plating baseline impedance modulus of the standard sample was established before the gold-antimony alloy deposition began. Pre-plating reference phase of standard sample ;in, For the first The qualified standard sample was in the first Pre-plating reference impedance modulus at a given frequency; For the first The qualified standard sample was in the first Pre-plating reference phase at a specific frequency;

[0157] At the qualified endpoint, the impedance modulus of the standard sample was collected. Phase of the end point of the standard sample The qualified endpoint is the first The electroplating stop time corresponding to the issuance of a qualified inspection record for each standard sample; among which... For the first The qualified standard sample at the qualification endpoint, the first Endpoint impedance modulus at each frequency; For the first The qualified standard sample at the qualification endpoint, the first The endpoint phase at each frequency;

[0158] Calculate the modulus offset and phase offset of the standard sample separately, as follows:

[0159] , ;in, For the first The qualified standard sample was in the first Standard modulus offset at each frequency; For the first The qualified standard sample was in the first Standard phase offset at each frequency;

[0160] Record the endpoint fingerprint as: ;in, For the first The endpoint fingerprint vector of a qualified standard sample;

[0161] The arithmetic mean of the endpoint fingerprints of all qualified standard samples is used to obtain the qualified endpoint center fingerprint: ;in, The center fingerprint vector of the endpoint fingerprint of all qualified standard samples, with dimension . ;

[0162] Construct the first The deviation of each qualified standard sample from the center fingerprint is:

[0163] ;in, For the first The standard deviation of a qualified standard sample from the center fingerprint of the qualified endpoint; The first fingerprint in the qualified endpoint center The modulus offset center value corresponding to each frequency; The qualified endpoint center fingerprint of the first The phase offset center value corresponding to each frequency;

[0164] The fingerprint dispersion of the standard sample is constructed as follows: ;in, The maximum value among all qualified standard samples;

[0165] Let the impedance modulus resolution be denoted as The phase resolution is denoted as ;

[0166] The resolving deviation of the calculation instrument is: ;in, This refers to the instrument resolution deviation caused by the instrument's impedance modulus resolution and phase resolution.

[0167] Set the allowable deviation from the endpoint as follows: ;in, This represents the allowable deviation from the endpoint.

[0168] The process of generating online thickness records, online antimony content records, and online grain records based on the multi-frequency impedance fingerprint of the current plated part specifically includes:

[0169] For the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the impedance modulus offset at each disturbance frequency is extracted.

[0170] The impedance modulus offsets at each disturbance frequency are summed and averaged according to the total number of disturbance frequencies to form the thickness-related projection corresponding to the current complete multi-frequency acquisition.

[0171] For the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the phase offset at each disturbance frequency is extracted;

[0172] The phase offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the antimony content related projection corresponding to the current complete multi-frequency acquisition.

[0173] The impedance modulus offset difference between adjacent disturbance frequencies is calculated one by one according to the disturbance frequency numbering order.

[0174] The impedance modulus offset difference between each adjacent perturbation frequency is squared and summed, and then averaged according to the total number of perturbation frequencies minus one, to form the grain-scale related projection quantity corresponding to the current complete multi-frequency acquisition.

[0175] Obtain the offline thickness test value, offline antimony mass fraction test value, and offline grain size test value for each qualified standard sample;

[0176] The offline thickness test values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average thickness of the qualified standard samples.

[0177] The offline antimony mass fraction test values ​​of all qualified standard samples are summed, and then averaged according to the number of qualified standard samples to form the average antimony mass fraction of qualified standard samples.

[0178] The offline grain size test values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average grain size of the qualified standard samples.

[0179] For the fingerprint of the qualified endpoint center, the modulus offset center value under each disturbance frequency is extracted, the modulus offset center value under each disturbance frequency is summed, and the average is processed according to the total number of disturbance frequencies to form the thickness-related projection of the qualified endpoint center.

[0180] For the fingerprint of the qualified endpoint center, the phase offset center value under each perturbation frequency is extracted, the phase offset center value under each perturbation frequency is summed, and the average is processed according to the total number of perturbation frequencies to form the relevant projection amount of antimony content of the qualified endpoint center.

[0181] For the fingerprint of the qualified endpoint center, the difference of the modulus offset center value between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency number order. The difference of the modulus offset center value between each adjacent perturbation frequency is squared and then summed. The average value is then calculated based on the total number of perturbation frequencies minus one, to form the grain-scale related projection of the qualified endpoint center.

[0182] The result of adding one to the thickness-related projection amount corresponding to the current complete multi-frequency acquisition is compared with the result of adding one to the thickness-related projection amount of the qualified endpoint center to form an online thickness ratio transformation result. The online thickness ratio transformation result is then multiplied with the average thickness of the qualified standard sample to form the online thickness record amount corresponding to the current complete multi-frequency acquisition.

[0183] The ratio of the antimony content related projection value corresponding to the current complete multi-frequency acquisition plus one is transformed with the ratio of the antimony content related projection value of the qualified endpoint center plus one, forming the online antimony content ratio transformation result. The online antimony content ratio transformation result is then multiplied with the average antimony mass fraction of the qualified standard sample to form the online antimony content record value corresponding to the current complete multi-frequency acquisition.

[0184] The result of adding one to the grain size-related projection of the qualified endpoint center is compared with the result of adding one to the grain size-related projection of the current complete multi-frequency acquisition to form an online grain ratio transformation result. The online grain ratio transformation result is then multiplied with the average grain size of the qualified standard sample to form the online grain recording quantity corresponding to the current complete multi-frequency acquisition.

[0185] The first The modulus offset in the sub-multi-frequency impedance fingerprint is converted into a thickness-related projection, specifically: ;in, For the first The thickness-related projection amount corresponding to sub-multi-frequency impedance fingerprints;

[0186] The first The phase shift in the sub-multi-frequency impedance fingerprint is converted into an antimony co-deposition correlation projection, specifically: ;in, For the first The antimony content-related projection amount corresponding to sub-multi-frequency impedance fingerprints;

[0187] The first In sub-multi-frequency impedance fingerprints, the modulus shift difference between adjacent frequencies is converted into a grain-scale correlated projection, specifically: ;in, For the first Grain-scale related projection amount corresponding to sub-multi-frequency impedance fingerprint;

[0188] The first The offline thickness test value of each qualified standard sample is recorded as follows: The offline antimony mass fraction test value is recorded as The offline grain size test value is recorded as ;

[0189] The average thickness of the qualified standard sample is calculated as follows: ;

[0190] The average antimony mass fraction of the qualified standard samples was calculated as follows: ;

[0191] The average grain size of the qualified standard sample was calculated as follows: ;

[0192] Calculate the first The number of online thickness records corresponding to each complete multi-frequency acquisition is:

[0193] ;

[0194] Calculate the first The number of online antimony content records corresponding to each complete multi-frequency acquisition is:

[0195] ;

[0196] Calculate the first The number of online grain records corresponding to each complete multi-frequency acquisition is:

[0197] .

[0198] Reference Figure 5 The step of comparing the multi-frequency impedance fingerprint of the current part to be plated with the qualified endpoint center fingerprint by perturbation frequency, calculating the endpoint deviation, and performing an electroplating termination determination based on the endpoint deviation and the longest allowable electroplating time specifically includes:

[0199] Compare the multi-frequency impedance fingerprint corresponding to the current complete multi-frequency acquisition with the qualified endpoint center fingerprint by perturbation frequency.

[0200] For each disturbance frequency, the impedance modulus offset corresponding to the current complete multi-frequency acquisition is subtracted from the modulus offset center value of the corresponding disturbance frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized according to the result of adding one to the current modulus offset center value to form the modulus endpoint deviation component under the current disturbance frequency.

[0201] For each disturbance frequency, the phase offset corresponding to the current complete multi-frequency acquisition is subtracted from the phase offset center value of the corresponding disturbance frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized according to the current phase offset center value plus one to form the phase endpoint deviation component under the current disturbance frequency.

[0202] The modulus endpoint deviation component and phase endpoint deviation component at all perturbation frequencies are summed and averaged at twice the total number of perturbation frequencies to calculate the endpoint deviation corresponding to the current complete multi-frequency acquisition.

[0203] When the deviation from the endpoint is not greater than the allowable deviation from the endpoint, the high-hardness gold-antimony alloy coating is recorded as having reached the non-destructive online endpoint.

[0204] When the endpoint deviation exceeds the allowable endpoint deviation, maintain the DC electroplating main current, and continue to calculate the endpoint deviation corresponding to the next complete multi-frequency acquisition after completing the next multi-frequency disturbance acquisition.

[0205] Obtain the maximum allowable electroplating time given by the process card, and record the completion time of each complete multi-frequency acquisition;

[0206] When there is a complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the endpoint allowable deviation, the endpoint status quantity is recorded as the endpoint reached status.

[0207] When there is no complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the endpoint allowable deviation, the endpoint status quantity is recorded as the endpoint not reached.

[0208] When the endpoint state is reached, select the earliest complete multi-frequency acquisition from the complete multi-frequency acquisitions that meet the conditions that the completion time is no later than the longest allowable electroplating time and the endpoint deviation is no greater than the endpoint allowable deviation as the record acquisition index.

[0209] When the endpoint state is not reached, select the latest complete multi-frequency acquisition with the latest completion time from the complete multi-frequency acquisitions with a completion time no later than the longest allowed electroplating time as the record acquisition index.

[0210] When the endpoint state is reached, the complete multi-frequency acquisition completion time corresponding to the record acquisition index will be used as the electroplating stop time.

[0211] When the endpoint state is not reached, the longest allowed electroplating time will be used as the electroplating stop time.

[0212] Use the endpoint deviation corresponding to the record collection index as the stop deviation;

[0213] At the moment when electroplating stops, the sinusoidal disturbance current corresponding to the amplitude of the DC electroplating main current and the AC disturbance current is cut off, and the workpiece is removed from the gold-antimony alloy deposition state.

[0214] When the endpoint state variable is "reached endpoint state", record the endpoint state.

[0215] When the endpoint state variable is "not reached", record the "not reached" state.

[0216] The first Sub-multi-frequency impedance fingerprint fingerprints at the qualified endpoint center After performing frequency-wise normalization comparison, we obtain: ;in, For the first The endpoint deviation corresponding to a complete multi-frequency acquisition;

[0217] When the deviation from the endpoint satisfies At that time, it was determined that the high-hardness gold-antimony alloy coating had reached the non-destructive online endpoint;

[0218] When the endpoint deviation does not meet the requirements At the same time, maintain DC electroplating current And calculate after completing the next multi-frequency disturbance acquisition. ;in, For the first The endpoint deviation corresponding to a complete multi-frequency acquisition;

[0219] Obtain the longest permissible plating time given in the process card, denoted as . ;

[0220] The first The time when the complete multi-frequency acquisition is completed is recorded as ;

[0221] Constructing the endpoint state variables Specifically:

[0222] ;in, It is an existential quantifier; This is a non-existent quantifier;

[0223] Build a record collection index Specifically:

[0224] ;

[0225] Constructing the electroplating stop time Specifically: ;

[0226] Set stop deviation amount Specifically: ;

[0227] exist The DC electroplating current and AC disturbance current are cut off at all times, and the workpiece is removed from the gold-antimony alloy deposition state.

[0228] when When the endpoint is reached, record the endpoint status.

[0229] when When the endpoint is not reached, record the status.

[0230] The process of cutting off the sinusoidal disturbance current corresponding to the amplitude of the DC electroplating main current and the AC disturbance current at the moment of electroplating stop, generating a stop data group, calculating the stop charge, and outputting the process stop result specifically includes: forming a stop data group using complete multi-frequency acquisition data corresponding to the recording acquisition index;

[0231] Write the electroplating stop time, endpoint state quantity, stop deviation quantity, online thickness record quantity, online antimony content record quantity, online grain record quantity, and multi-frequency impedance fingerprint vector into the stop data group;

[0232] Obtain the start time of electroplating and the DC electroplating main current;

[0233] The electroplating duration is calculated by subtracting the electroplating start time from the electroplating stop time.

[0234] Multiply the DC electroplating main current by the electroplating duration to form the stop charge;

[0235] The output process stop results include the stop deviation, endpoint status, electroplating stop time, stop charge, online thickness record, online antimony content record, and online grain size record.

[0236] To record collection index The corresponding complete multi-frequency acquisition data forms a stop data set, which includes the electroplating stop time. Endpoint state quantity Stop deviation amount Thickness recording volume Antimony content recording amount Grain recording amount and multi-frequency impedance fingerprint vector ;in, To record the number of online thickness records corresponding to the acquisition index; To record the number of online antimony content records corresponding to the collection index; To record the number of online grain records corresponding to the acquisition index; To record the multi-frequency impedance fingerprint vector corresponding to the acquisition index;

[0237] The start time of electroplating is recorded as... The stop charge is calculated based on the DC electroplating current as follows:

[0238] ;in, To stop the amount of charge;

[0239] Output process stop results, including: stop deviation amount Endpoint state quantity Electroplating stop time Stop charge quantity Thickness recording volume Antimony content recording amount Grain recording amount .

[0240] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0241] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A process for electroplating high-hardness gold-antimony alloy, characterized in that, include: Establish the effective cathode surface of the workpiece to be electroplated, obtain the effective cathode area, set the DC current density for the gold-antimony alloy electroplating stage, and calculate the DC electroplating main current and AC disturbance current amplitude according to the effective cathode area and DC current density. Establish a multi-frequency disturbance frequency sequence and a single-frequency sampling window, perform pre-plating reference disturbance acquisition, and calculate the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response; During the gold-antimony alloy deposition process, the online voltage response between the cathode and the reference terminal was acquired according to the multi-frequency perturbation sequence, and the amplitude and phase of the online voltage response were extracted. Based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase, construct the multi-frequency impedance fingerprint of the current part to be plated; Prepare qualified high-hardness gold-antimony alloy standard samples, collect the endpoint impedance modulus and endpoint phase of each qualified standard sample, and establish the qualified endpoint center fingerprint and the allowable deviation of the endpoint. Based on the multi-frequency impedance fingerprint of the plated parts, online thickness recording, online antimony content recording, and online grain size recording are generated; The multi-frequency impedance fingerprint of the current part to be plated is compared with the qualified endpoint center fingerprint by perturbation frequency to calculate the endpoint deviation. The electroplating termination judgment is then performed based on the endpoint deviation and the longest allowable electroplating time. At the moment of electroplating stop, the DC electroplating main current and the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current are cut off, a stop data set is generated, the stop charge is calculated, and the process stop result is output.

2. The electroplating process for a high-hardness gold-antimony alloy according to claim 1, characterized in that, The process of establishing an effective cathode surface for the workpiece to be electroplated, obtaining the effective cathode area, setting the DC current density for the gold-antimony alloy electroplating stage, and calculating the DC electroplating main current and AC disturbance current amplitude based on the effective cathode area and DC current density specifically includes: The conductive surface of the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is taken as the effective cathode surface, and the effective cathode area of ​​the workpiece to be electroplated that actually participates in the deposition of gold-antimony alloy is measured. The DC current density during the gold-antimony alloy electroplating stage is set as the calculation parameter for the main electroplating current. The DC electroplating main current for the gold-antimony alloy electroplating stage is calculated by multiplying the effective cathode area by the DC current density. The DC electroplating main current is converted to one-fiftieth of its original value to obtain the amplitude of the AC disturbance current superimposed during multi-frequency impedance acquisition.

3. The electroplating process for a high-hardness gold-antimony alloy according to claim 2, characterized in that, The process of establishing a multi-frequency disturbance sequence and a single-frequency sampling window, performing pre-plating reference disturbance acquisition, and calculating the pre-plating reference impedance modulus and pre-plating reference phase based on the pre-plating voltage response specifically includes: Establish a set of multi-frequency disturbance frequency numbers, set the total number of disturbance frequencies to be no less than five, and configure a corresponding disturbance frequency number index for each disturbance frequency; Set the lowest and highest frequencies in the multi-frequency perturbation frequency sequence, and make the lowest frequency greater than zero and less than the highest frequency; Based on the geometric interval relationship from the lowest frequency to the highest frequency, the perturbation frequencies corresponding to each perturbation frequency number index are generated sequentially. Set the number of voltage sampling points at each disturbance frequency to be no less than sixteen, and establish a single-frequency sampling number set; Based on the ratio between the sampling number and the number of voltage sampling points, and combined with the current disturbance frequency, the sampling time corresponding to each sampling number under the current disturbance frequency is generated; During the gold-antimony alloy deposition stage, the DC electroplating main current is superimposed with the sinusoidal disturbance current corresponding to the amplitude of the AC disturbance current, and an online single-frequency sinusoidal disturbance current is formed according to the current disturbance frequency. Obtain the effective lower limit and effective upper limit of the voltage acquisition device, and ensure that the effective lower limit is less than the effective upper limit; Before the gold-antimony alloy deposition begins, the DC electroplating main current is kept off. Only the AC disturbance current amplitude is used to perform a multi-frequency disturbance acquisition on the same workpiece according to the multi-frequency disturbance sequence to form the pre-plating reference disturbance current at each disturbance frequency. At each perturbation frequency, the pre-plating voltage response is collected according to the corresponding sampling time. When all pre-plating sampled voltages at the current disturbance frequency are between the lower limit and the upper limit of the effective range of the voltage acquisition device, all pre-plating sampled voltages at the current disturbance frequency are retained. If the pre-plating sampling voltage is lower than the lower limit of the effective range or higher than the upper limit of the effective range at the current disturbance frequency, the pre-plating voltage response data at the current disturbance frequency will be re-acquired. When all the pre-plating sampled voltages after re-acquisition are within the lower limit and upper limit of the effective range of the voltage acquisition device, all the pre-plating sampled voltages obtained after re-acquisition are retained. If, after re-collection, the pre-plating sampling voltage is still lower than the lower limit of the effective range or higher than the upper limit of the effective range, the endpoint determination of the current plated part is stopped, the abnormal state of the pre-plating reference is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated. The pre-plating voltage response under the current disturbance frequency is sinusoidally projected. Each pre-plating sampled voltage is multiplied by the sinusoidal reference value at the corresponding sampling time and summed. The result is then normalized according to the number of voltage sampling points to form the sinusoidal projection component of the pre-plating voltage response under the current disturbance frequency. Cosine projection is performed on the pre-plating voltage response at the current disturbance frequency. The pre-plating sampled voltage is multiplied by the cosine reference value at the corresponding sampling time and summed. The result is then normalized according to the number of voltage sampling points to form the cosine projection component of the pre-plating voltage response at the current disturbance frequency. The square of the sinusoidal projection component and the square of the cosine projection component of the pre-plating voltage response at the current disturbance frequency are added together, and the square root of the sum is taken to calculate the amplitude of the pre-plating voltage response at the current disturbance frequency. When the pre-plating voltage response amplitude is greater than zero, perform a binary arctangent operation based on the cosine projection component and sine projection component of the pre-plating voltage response at the current disturbance frequency to calculate the pre-plating reference phase at the current disturbance frequency. Divide the pre-plating voltage response amplitude by the AC disturbance current amplitude to calculate the pre-plating reference impedance modulus at the current disturbance frequency. When the pre-plating voltage response amplitude is zero, stop the endpoint determination of the current plated part, record the pre-plating reference abnormal state, and do not generate the multi-frequency impedance fingerprint of the current plated part.

4. The electroplating process for a high-hardness gold-antimony alloy according to claim 3, characterized in that, The process of acquiring the online voltage response between the cathode and the reference terminal according to a multi-frequency perturbation sequence during the gold-antimony alloy deposition, and extracting the online voltage response amplitude and phase, specifically includes: Establish a complete multi-frequency acquisition count index, and record each complete multi-frequency acquisition process according to the complete multi-frequency acquisition count index; During the disturbance period of the current disturbance frequency in any complete multi-frequency acquisition, the voltage between the cathode and the reference terminal is acquired at each sampling moment under the current disturbance frequency to form an online sampling voltage; Integrity determination is performed on the online sampled voltage using the lower and upper limits of the effective range of the voltage acquisition device. When all online sampled voltages at the current disturbance frequency in any complete multi-frequency acquisition are between the lower limit and the upper limit of the effective range, all online sampled voltages at the current disturbance frequency in the current complete multi-frequency acquisition are retained. If, during any complete multi-frequency acquisition, the online sampling voltage at the current disturbance frequency is lower than the lower limit of the effective range or higher than the upper limit of the effective range, the voltage response data at the current disturbance frequency in the current complete multi-frequency acquisition will be reacquired. When all online sampled voltages after re-acquisition are between the lower limit and the upper limit of the effective range, all online sampled voltages obtained after re-acquisition are retained; If, after re-collection, the online sampling voltage is still below the lower limit of the effective range or above the upper limit of the effective range, the endpoint determination of the current plated part is stopped, the abnormal sampling state is recorded, and the multi-frequency impedance fingerprint of the current plated part is not generated. For any complete multi-frequency acquisition, perform sinusoidal projection on all effective voltage sample values ​​at the current disturbance frequency, multiply each effective voltage sample value by the sinusoidal reference value at the corresponding sampling time, sum the results, and normalize according to the number of voltage sampling points to form online voltage sinusoidal projection components. Cosine projection is performed on all effective voltage sample values ​​at the current disturbance frequency in any complete multi-frequency acquisition. Each effective voltage sample value is multiplied by the cosine reference value at the corresponding sampling time and summed. Then, the values ​​are normalized according to the number of voltage sampling points to form online voltage cosine projection components. The square value of the sinusoidal projection component of the online voltage is added to the square value of the cosine projection component of the online voltage, and the square root of the sum is taken to calculate the amplitude of the online voltage response at the current disturbance frequency in the current complete multi-frequency acquisition. When the amplitude of the online voltage response is greater than zero, the phase of the online voltage response at the current disturbance frequency in the current complete multi-frequency acquisition is calculated by performing a binary arctangent operation based on the cosine projection component and the sine projection component of the online voltage. When the online voltage response amplitude is zero, stop the endpoint determination of the current plated part, record the abnormal state of the impedance signal, and do not generate a multi-frequency impedance fingerprint of the current plated part.

5. The electroplating process for a high-hardness gold-antimony alloy according to claim 4, characterized in that, The construction of a multi-frequency impedance fingerprint of the workpiece to be plated based on the online voltage response amplitude, AC disturbance current amplitude, pre-plating reference impedance modulus, and pre-plating reference phase specifically includes: Divide the online voltage response amplitude at the current disturbance frequency in the current complete multi-frequency acquisition by the AC disturbance current amplitude to calculate the online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition. The online impedance modulus at the current disturbance frequency in the current complete multi-frequency acquisition is transformed into a dimensionless ratio with the pre-plating reference impedance modulus at the corresponding disturbance frequency. The natural logarithm of the ratio transformation result is then used to form the impedance modulus offset at the current disturbance frequency in the current complete multi-frequency acquisition. The phase offset at the current disturbance frequency is obtained by subtracting the online voltage response phase at the current disturbance frequency from the pre-plating reference phase at the corresponding disturbance frequency in the current complete multi-frequency acquisition, and then performing dimensionless processing according to the circumferential angle reference. According to the perturbation frequency numbering order, the impedance modulus offset and phase offset at each perturbation frequency obtained from the current complete multi-frequency acquisition are arranged alternately to form the multi-frequency impedance fingerprint vector corresponding to the current complete multi-frequency acquisition. Set the dimension of the multi-frequency impedance fingerprint vector to twice the total number of perturbation frequencies.

6. The electroplating process for a high-hardness gold-antimony alloy according to claim 5, characterized in that, The preparation of qualified high-hardness gold-antimony alloy standard samples involves collecting the endpoint impedance modulus and endpoint phase of each qualified standard sample, establishing the qualified endpoint center fingerprint and the allowable endpoint deviation, specifically including: Using the same substrate shape, effective cathode area, and DC current density as the workpiece to be electroplated, prepare qualified high-hardness gold-antimony alloy standard samples, and set the number of qualified standard samples to be no less than three. Establish a set of qualified standard samples with qualified inspection records formed by process inspection specifications as qualified standard sample number set, and configure a qualified standard sample number index for each qualified standard sample; For each qualified standard sample, the same multi-frequency disturbance frequency sequence, sampling time, AC disturbance current amplitude, voltage projection process and impedance modulus calculation process as the part to be plated are used to establish the standard sample pre-plating reference impedance modulus and standard sample pre-plating reference phase at each disturbance frequency for the current qualified standard sample before the gold-antimony alloy deposition begins. At the qualified endpoint, the current qualified standard sample is collected at each disturbance frequency to obtain the endpoint impedance modulus and endpoint phase, and the electroplating stop time corresponding to the current qualified standard sample when the qualified inspection record is obtained is taken as the qualified endpoint. The current qualified standard sample at each disturbance frequency is subjected to a dimensionless ratio transformation of the endpoint impedance modulus at each disturbance frequency and the reference impedance modulus of the standard sample before plating at the corresponding disturbance frequency. The natural logarithm of the ratio transformation result is then used to form the standard modulus offset of the current qualified standard sample at each disturbance frequency. Subtract the pre-plating reference phase of the standard sample at the corresponding perturbation frequency from the endpoint phase of the current qualified standard sample at each perturbation frequency, and perform dimensionless processing according to the circumferential angle reference to form the standard phase offset of the current qualified standard sample at each perturbation frequency. According to the perturbation frequency numbering order, the standard modulus offset and standard phase offset of the current qualified standard sample at each perturbation frequency are arranged alternately to form the endpoint fingerprint vector of the current qualified standard sample. The arithmetic mean of the endpoint fingerprint vectors of all qualified standard samples is calculated for each component to form the qualified endpoint center fingerprint of all qualified standard samples, and the dimension of the qualified endpoint center fingerprint is set to twice the total number of perturbation frequencies. For each qualified standard sample, the standard modulus offset of the current qualified standard sample is compared with the modulus offset center value of the corresponding perturbation frequency in the qualified endpoint center fingerprint at each perturbation frequency. The difference is then normalized according to the result of adding one to the current modulus offset center value to form the modulus deviation component at the current perturbation frequency. For each qualified standard sample, the standard phase offset of the current qualified standard sample is compared with the phase offset center value of the corresponding perturbation frequency in the center fingerprint of the qualified endpoint at each perturbation frequency. The difference is then normalized according to the result of adding one to the current phase offset center value to form the phase deviation component at the current perturbation frequency. For each qualified standard sample, the modulus deviation component and phase deviation component at all disturbance frequencies are summed and averaged at twice the total number of disturbance frequencies to calculate the standard deviation of the current qualified standard sample relative to the center fingerprint of the qualified endpoint. The maximum value is selected from the standard deviations of all qualified standard samples to form the fingerprint dispersion of the standard samples. Obtain the instrument impedance modulus resolution and instrument phase resolution. Add the pre-plating reference impedance modulus of the current part to be plated at each disturbance frequency to the instrument impedance modulus resolution, and then perform a ratio transformation with the pre-plating reference impedance modulus of the current part to be plated at the same disturbance frequency. Perform a natural logarithmic operation on the ratio transformation result to form the modulus resolution deviation term at each disturbance frequency. The modulus resolution deviation term calculated by natural logarithm at each perturbation frequency is summed with the instrument phase resolution according to the perturbation frequency, and then averaged according to twice the total number of perturbation frequencies to calculate the instrument resolution deviation caused by the instrument impedance modulus resolution and the instrument phase resolution. The fingerprint dispersion of the standard sample is compared with the instrument resolution deviation, and the maximum value between the fingerprint dispersion of the standard sample and the instrument resolution deviation is selected as the allowable deviation of the endpoint.

7. The electroplating process for a high-hardness gold-antimony alloy according to claim 6, characterized in that, The process of generating online thickness records, online antimony content records, and online grain records based on the multi-frequency impedance fingerprint of the current plated part specifically includes: For the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the impedance modulus offset at each disturbance frequency is extracted. The impedance modulus offsets at each disturbance frequency are summed and averaged according to the total number of disturbance frequencies to form the thickness-related projection corresponding to the current complete multi-frequency acquisition. For the multi-frequency impedance fingerprint obtained from the current complete multi-frequency acquisition, the phase offset at each disturbance frequency is extracted; The phase offsets at each perturbation frequency are summed and averaged according to the total number of perturbation frequencies to form the antimony content related projection corresponding to the current complete multi-frequency acquisition. The impedance modulus offset difference between adjacent disturbance frequencies is calculated one by one according to the disturbance frequency numbering order. The impedance modulus offset difference between each adjacent perturbation frequency is squared and summed, and then averaged according to the total number of perturbation frequencies minus one, to form the grain-scale related projection quantity corresponding to the current complete multi-frequency acquisition. Obtain the offline thickness test value, offline antimony mass fraction test value, and offline grain size test value for each qualified standard sample; The offline thickness test values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average thickness of the qualified standard samples. The offline antimony mass fraction test values ​​of all qualified standard samples are summed, and then averaged according to the number of qualified standard samples to form the average antimony mass fraction of qualified standard samples. The offline grain size test values ​​of all qualified standard samples are summed and averaged according to the number of qualified standard samples to form the average grain size of the qualified standard samples. For the fingerprint of the qualified endpoint center, the modulus offset center value under each disturbance frequency is extracted, the modulus offset center value under each disturbance frequency is summed, and the average is processed according to the total number of disturbance frequencies to form the thickness-related projection of the qualified endpoint center. For the fingerprint of the qualified endpoint center, the phase offset center value under each perturbation frequency is extracted, the phase offset center value under each perturbation frequency is summed, and the average is processed according to the total number of perturbation frequencies to form the relevant projection amount of antimony content of the qualified endpoint center. For the fingerprint of the qualified endpoint center, the difference of the modulus offset center value between adjacent perturbation frequencies is calculated one by one according to the perturbation frequency number order. The difference of the modulus offset center value between each adjacent perturbation frequency is squared and then summed. The average value is then calculated based on the total number of perturbation frequencies minus one, to form the grain-scale related projection of the qualified endpoint center. The result of adding one to the thickness-related projection amount corresponding to the current complete multi-frequency acquisition is compared with the result of adding one to the thickness-related projection amount of the qualified endpoint center to form an online thickness ratio transformation result. The online thickness ratio transformation result is then multiplied with the average thickness of the qualified standard sample to form the online thickness record amount corresponding to the current complete multi-frequency acquisition. The ratio of the antimony content related projection value corresponding to the current complete multi-frequency acquisition plus one is transformed with the ratio of the antimony content related projection value of the qualified endpoint center plus one, forming the online antimony content ratio transformation result. The online antimony content ratio transformation result is then multiplied with the average antimony mass fraction of the qualified standard sample to form the online antimony content record value corresponding to the current complete multi-frequency acquisition. The result of adding one to the grain size-related projection of the qualified endpoint center is compared with the result of adding one to the grain size-related projection of the current complete multi-frequency acquisition to form an online grain ratio transformation result. The online grain ratio transformation result is then multiplied with the average grain size of the qualified standard sample to form the online grain recording quantity corresponding to the current complete multi-frequency acquisition.

8. The electroplating process for a high-hardness gold-antimony alloy according to claim 7, characterized in that, The process of comparing the multi-frequency impedance fingerprint of the current part to be plated with the qualified endpoint center fingerprint by frequency perturbation, calculating the endpoint deviation, and performing an electroplating termination determination based on the endpoint deviation and the longest allowable electroplating time specifically includes: Compare the multi-frequency impedance fingerprint corresponding to the current complete multi-frequency acquisition with the qualified endpoint center fingerprint by perturbation frequency. For each disturbance frequency, the impedance modulus offset corresponding to the current complete multi-frequency acquisition is subtracted from the modulus offset center value of the corresponding disturbance frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized according to the result of adding one to the current modulus offset center value to form the modulus endpoint deviation component under the current disturbance frequency. For each disturbance frequency, the phase offset corresponding to the current complete multi-frequency acquisition is subtracted from the phase offset center value of the corresponding disturbance frequency in the qualified endpoint center fingerprint, and the resulting difference is normalized according to the current phase offset center value plus one to form the phase endpoint deviation component under the current disturbance frequency. The modulus endpoint deviation component and phase endpoint deviation component at all perturbation frequencies are summed and averaged at twice the total number of perturbation frequencies to calculate the endpoint deviation corresponding to the current complete multi-frequency acquisition. When the deviation from the endpoint is not greater than the allowable deviation from the endpoint, the high-hardness gold-antimony alloy coating is recorded as having reached the non-destructive online endpoint. When the endpoint deviation exceeds the allowable endpoint deviation, maintain the DC electroplating main current, and continue to calculate the endpoint deviation corresponding to the next complete multi-frequency acquisition after completing the next multi-frequency disturbance acquisition. Obtain the maximum allowable electroplating time given by the process card, and record the completion time of each complete multi-frequency acquisition; When there is a complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the endpoint allowable deviation, the endpoint status quantity is recorded as the endpoint reached status. When there is no complete multi-frequency acquisition with a completion time no later than the longest allowed electroplating time and an endpoint deviation no greater than the endpoint allowable deviation, the endpoint status quantity is recorded as the endpoint not reached. When the endpoint state is reached, select the earliest complete multi-frequency acquisition from the complete multi-frequency acquisitions that meet the conditions that the completion time is no later than the longest allowable electroplating time and the endpoint deviation is no greater than the endpoint allowable deviation as the record acquisition index. When the endpoint state is not reached, select the latest complete multi-frequency acquisition with the latest completion time from the complete multi-frequency acquisitions with a completion time no later than the longest allowed electroplating time as the record acquisition index. When the endpoint state is reached, the complete multi-frequency acquisition completion time corresponding to the record acquisition index will be used as the electroplating stop time. When the endpoint state is not reached, the longest allowed electroplating time will be used as the electroplating stop time. Use the endpoint deviation corresponding to the record collection index as the stop deviation; At the moment when electroplating stops, the sinusoidal disturbance current corresponding to the amplitude of the DC electroplating main current and the AC disturbance current is cut off, and the workpiece is removed from the gold-antimony alloy deposition state. When the endpoint state variable is "reached endpoint state", record the endpoint state. When the endpoint state variable is "not reached", record the "not reached" state.

9. The electroplating process for a high-hardness gold-antimony alloy according to claim 8, characterized in that, The process of cutting off the sinusoidal disturbance current corresponding to the amplitude of the DC electroplating main current and the AC disturbance current at the moment of electroplating stop, generating a stop data group, calculating the stop charge, and outputting the process stop result specifically includes: forming a stop data group using complete multi-frequency acquisition data corresponding to the recording acquisition index; Write the electroplating stop time, endpoint state quantity, stop deviation quantity, online thickness record quantity, online antimony content record quantity, online grain record quantity, and multi-frequency impedance fingerprint vector into the stop data group; Obtain the start time of electroplating and the DC electroplating main current; The electroplating duration is calculated by subtracting the electroplating start time from the electroplating stop time. Multiply the DC electroplating main current by the electroplating duration to form the stop charge; The output process stop results include the stop deviation, endpoint status, electroplating stop time, stop charge, online thickness record, online antimony content record, and online grain size record.