A method for evaluating power loss of gallium oxide semiconductor material

Abstract

The application discloses a kind of power loss evaluation methods of gallium oxide semiconductor material, it is related to semiconductor material testing technical field, including, to the shutdown and commutation electric field fast estimation of metrological reference master file, identify weak medium partition, unify medium reliability index, summary is medium partition stress monitoring table;To the partition stress on-line monitoring of medium partition stress monitoring table, determine the upper limit of pre-failure field intensity, map safety working condition domain, generate dynamic safety working condition upper limit table;According to dynamic safety working condition upper limit table, build hot stack matrix, formulate temperature measurement calibration specification and sample according to test measurement chain, output electric heat synchronous measurement data summary table;Based on electric heat synchronous measurement data summary table, construct equivalent thermal resistance heat capacity network, and calculate temperature rise drift coefficient while loss conversion, obtain reference thermal boundary normalization result total table.The application improves the consistency and comparability of power loss evaluation results under different conditions.

Description

Technical Field

[0001] This invention relates to the field of semiconductor material testing technology, and in particular to a method for evaluating the power loss of gallium oxide semiconductor materials. Background Technology

[0002] Gallium oxide (GaO) semiconductors, as a new generation of wide-bandgap semiconductors, exhibit excellent material properties in power electronic devices and have attracted widespread attention in high-voltage and high-frequency applications. Accurate power loss assessment of GaO semiconductors is a crucial step in evaluating their efficiency and reliability. Existing assessment methods are typically based on electrothermal parameter testing. By measuring key parameters such as voltage, current, and junction temperature under specific operating conditions, and combining these with device structural characteristics and material properties, steady-state or transient electrothermal models are established to analyze and calculate power loss components such as conduction loss and switching loss. These methods rely on comprehensive measurements of multiple parameters, including epitaxial layer thickness, doping concentration, crystal orientation, and heat dissipation structure, and require ensuring the synchronization and accuracy of each metering channel in the testing system to achieve a quantitative assessment of power loss under device operating conditions.

[0003] However, existing assessment methods still have certain limitations in implementation. On the one hand, under complex electric and thermal field coupling conditions, traditional methods are not comprehensive enough in monitoring the stress in the local electric field concentration areas of the dielectric layer, leading to deviations in the mapping of the safe operating condition domain and affecting the accuracy of the assessment results. On the other hand, in the electrothermal synchronous measurement stage, temperature calibration and data acquisition often rely on fixed thermal boundary conditions, resulting in insufficient consistency in the normalization of power loss under different operating conditions, affecting the accuracy and comparability of the assessment results. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a power loss assessment method for gallium oxide semiconductor materials to solve the problem of insufficient accuracy and comparability of assessment results in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides a method for evaluating the power loss of gallium oxide semiconductor materials, comprising,

[0008] A metrology reference table for gallium oxide semiconductor materials is collected, and the metrology reference table is processed to achieve metrology chain consistency, resulting in a test metrology chain. This chain is then summarized with the metrology reference table to obtain the metrology reference master file.

[0009] The main metrology reference file is used to quickly predict the turn-off and commutation electric fields, identify weak medium zones, unify medium reliability indicators, and summarize them into a medium zone stress monitoring table.

[0010] Implement online monitoring of zoned stress in the medium zoned stress monitoring table, determine the upper limit of the field strength before failure, map the safe operating condition domain, and generate a dynamic safe operating condition upper limit table;

[0011] A heat stack matrix is ​​built based on the dynamic safety operating condition upper limit table, temperature measurement calibration specifications are formulated and sampling is performed according to the test and measurement chain, and a statistical table of synchronous electrothermal measurement data is output.

[0012] Based on the statistical table of synchronous electrothermal measurement data, an equivalent thermal resistance and thermal capacity network is constructed, and the temperature rise drift coefficient is calculated while loss is converted to obtain a general table of reference thermal boundary normalization results.

[0013] The reference thermal boundary normalization result summary table is time-mapped and periodically summed to generate a gallium oxide power loss assessment report.

[0014] As a preferred embodiment of the power loss assessment method for the gallium oxide semiconductor material of the present invention, the structure of the gallium oxide semiconductor material includes a gallium oxide functional epitaxial layer, a dielectric overlay stack, a metallization layer and a surface heat dissipation contact layer.

[0015] The surface heat dissipation contact layer includes the material and thickness of the surface heat dissipation contact layer;

[0016] The metrological reference table includes the thickness and doping parameters, crystallographic orientation, and step characteristic parameters of the gallium oxide functional epitaxial layer.

[0017] As a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the steps for obtaining the metrological benchmark master file are as follows:

[0018] Voltage measurement points, current measurement points, gate signal measurement points, and temperature measurement points are connected to the metallization layer to obtain voltage channels, current channels, gate channels, and temperature channels. Zero-point calibration, phase calibration, and bandwidth calibration are performed on the voltage channels, current channels, gate channels, and temperature channels.

[0019] The sampling time of the voltage channel, current channel, gate channel and temperature channel is calibrated and summarized into a test measurement chain;

[0020] The metrology reference table and the test metrology chain are merged and registered to obtain the metrology reference master file.

[0021] In a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the steps for identifying weak dielectric partitions are as follows:

[0022] Based on the main metrological reference file, the gallium oxide functional epitaxial layer is layered, and the voltage change generated by the metallization layer is distributed according to the layer ratio to obtain the electric field value. The step transition point is determined based on the crystallographic orientation and step characteristic parameters, and the electric field concentration point is identified. The start and end points of the dielectric cover are extracted from the dielectric cover stack.

[0023] By comparing the electric field values ​​at the step transition point, electric field concentration point, and dielectric coverage start and end point with the surrounding locations, the sidewall passivation transition region, gate oxide critical region, and terminal inflection point are obtained.

[0024] The sidewall passivation transition region, gate oxide critical region, and terminal inflection point are uniformly registered as a list of dielectric-sensitive locations. The electric field values ​​of each location in the list of dielectric-sensitive locations are compared one by one with the electric field values ​​of other locations outside the list of dielectric-sensitive locations to obtain the weak dielectric partitions of gallium oxide semiconductor materials.

[0025] In a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the step of summarizing the data into a dielectric zone stress monitoring table is as follows:

[0026] Leakage current, threshold drift, small signal capacitance change, and breakdown event records are read for the gate oxide critical region, sidewall passivation transition region, and terminal inflection point in the weak medium partition, and the medium is judged to be in an unacceptable state or an acceptable state.

[0027] In the critical region of gate oxide, the sidewall passivation transition region, and the terminal inflection point, records of media failure and media satisfaction are recorded separately according to whether the media is in an unpermitted state or in a permitted state, and then merged into a unified media reliability index.

[0028] By summarizing the unified medium reliability index and the list of medium-sensitive locations, a medium zoning stress monitoring table is obtained.

[0029] As a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the steps for generating the dynamic safe operating condition upper limit table are as follows:

[0030] The dielectric partition stress monitoring table is turned on in the metallization layer, and voltage change, current change, gate signal change and temperature change are acquired simultaneously. At the same time, the corresponding leakage current, threshold drift, small signal capacitance change and breakdown event record are obtained.

[0031] The leakage current, threshold drift, small signal capacitance change and breakdown event records at each detection location are compared with the unified medium reliability index to confirm the upper limit of the pre-failure field strength and the allowable value of the upper limit of the pre-failure field strength at the detection location.

[0032] The upper limit of the field strength before failure and the allowable value of the upper limit of the field strength before failure are converted into the bus voltage range, gate bias range, voltage change time range and junction temperature range respectively, and then merged into a safe operating condition domain.

[0033] By summarizing all safety operating condition domains, a dynamic safety operating condition upper limit table is obtained.

[0034] As a preferred embodiment of the power loss assessment method for the gallium oxide semiconductor material described in this invention, the steps for compiling the output electrothermal synchronous measurement data are as follows:

[0035] The junction temperature ranges of each detection location in the dynamic safety condition upper limit table are classified and junction temperature targets are extracted. For the junction temperature targets, surface heat dissipation contact layers that can make temperature changes fall within the junction temperature range are selected. After registration, the junction temperature targets are sorted from low to high to form a heat stack matrix.

[0036] According to the junction temperature target, the recording order of the temperature before operation, the starting temperature of operation, the temperature during operation, and the temperature after operation is specified, and zero-point calibration is performed for each, forming a temperature measurement calibration specification.

[0037] The voltage, current, gate, and temperature channels of the test metering chain are invoked to perform a turn-on operation on the metallization layer and perform zero-point calibration to obtain electrothermal synchronous data records corresponding to the junction temperature target and the surface heat dissipation contact layer.

[0038] The electrothermal synchronization data were sorted according to the junction temperature target, and the surface heat dissipation contact layer was labeled one by one. The data were then summarized with the temperature measurement calibration specifications to obtain a summary table of electrothermal synchronization measurement data.

[0039] As a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the construction of the equivalent thermal resistance and thermal capacity network specifically refers to calculating the loss amount by compiling the electrothermal synchronous measurement data table, performing zero-point correction on the post-operation temperature and the operation process temperature respectively, and forming an equivalent thermal resistance set and an equivalent thermal capacity set with the ratio of the loss amount, and registering them as the equivalent thermal resistance and thermal capacity network.

[0040] As a preferred embodiment of the power loss assessment method for gallium oxide semiconductor materials described in this invention, the steps for obtaining the reference thermal boundary normalization result summary table are as follows:

[0041] In the equivalent thermal resistance and thermal capacity network, the surface heat dissipation contact layers are classified and sorted according to the junction temperature target. Adjacent electrothermal synchronous data records are extracted to calculate the temperature change increment and loss increment, and the temperature rise drift coefficient of each surface heat dissipation contact layer is obtained and summarized into a set of temperature rise drift coefficients.

[0042] In the electrothermal synchronous measurement data table, a reference thermal boundary conversion benchmark is selected, and the temperature rise drift coefficient set is called to convert the loss of each record of electrothermal synchronous data, generating a reference thermal boundary normalization result summary table.

[0043] In a preferred embodiment of the power loss assessment method for gallium oxide semiconductor material according to the present invention, the steps for forming a gallium oxide power loss assessment report are as follows:

[0044] By performing time mapping between the reference thermal boundary normalization result summary table and the electrothermal synchronization data record, a reference thermal boundary normalization result summary table with time sequence is obtained.

[0045] The operation group is obtained from the reference thermal boundary normalization result summary table with time sequence information. The number of times the operation group is registered is counted, multiplied with the loss amount, and summed to obtain the total converted loss amount. This total loss amount is then registered with the reference thermal boundary conversion benchmark to form a gallium oxide power loss assessment report.

[0046] The beneficial effects of this invention are as follows: by obtaining the medium partition stress monitoring table, it is possible to provide early warning of local failure risks and provide an accurate data basis for dynamic safe operating condition domain mapping; by obtaining the reference thermal boundary normalization result summary table, the influence of thermal boundary condition differences on loss assessment is eliminated, and the consistency and comparability of power loss assessment results under different operating conditions are improved. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a method for evaluating the power loss of gallium oxide semiconductor materials.

[0049] Figure 2 A flowchart for generating a medium zoning stress monitoring table.

[0050] Figure 3 A flowchart for generating a table of synchronous electrothermal measurement data.

[0051] Figure 4 A flowchart for generating a summary table of reference thermal boundary normalization results. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0055] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a method for evaluating the power loss of gallium oxide semiconductor materials, including the following steps:

[0056] S1. Analyze the structure of gallium oxide semiconductor materials, collect the measurement reference table of gallium oxide semiconductor materials, perform measurement chain consistency processing on the measurement reference table to obtain the test measurement chain, and summarize it with the measurement reference table to obtain the measurement reference master file.

[0057] Furthermore, the structure of gallium oxide semiconductor materials includes a gallium oxide functional epitaxial layer, a dielectric overlay stack, a metallization layer, and a surface heat dissipation contact layer.

[0058] It should be noted that the surface heat dissipation contact layer includes the material and thickness of the surface heat dissipation contact layer.

[0059] The metrological reference table includes the thickness and doping parameters of the gallium oxide functional epitaxial layer, as well as the crystallographic orientation and step characteristic parameters.

[0060] It should be noted that crystallographic orientation and step characteristic parameters refer to the crystal plane orientation corresponding to the surface of gallium oxide semiconductor material, the steps set according to the minute undulations on the crystal plane, and the height, spacing and orientation of the steps.

[0061] Voltage measurement points, current measurement points, gate signal measurement points, and temperature measurement points are connected to the metallization layer to obtain voltage channels, current channels, gate channels, and temperature channels. Zero-point calibration, phase calibration, and bandwidth calibration are then performed on the voltage channels, current channels, gate channels, and temperature channels.

[0062] Furthermore, the metrological reference meter undergoes metrological chain consistency processing. Voltage, current, gate signal, and temperature measurement points are connected to the metallization layer to obtain voltage, current, gate, and temperature channels. Baseline data for these channels in a non-operational state is recorded, and zero-point calibration is performed based on this baseline data. Simultaneously, the voltage, current, gate, and temperature channels are activated on the metallization layer. Voltage, current, gate signal, and temperature changes are recorded during these activation operations. The start and end times of all changes are extracted. Using the start time of the voltage change as the phase reference, channels with earlier start times than the voltage change are entered into the next phase. The time shift is performed by moving channels whose start time is later than the voltage change to the front, making the start and end times of the change the same for all channels, eliminating phase differences and completing phase calibration. The slope of all changes is extracted, and the gate change slope is used as the bandwidth reference. The slopes of voltage, current, and temperature changes are compared with the gate change slope. If these slopes differ from the gate change slope, the voltage, current, and temperature measurement points for each channel are adjusted until they match the gate change slope, completing bandwidth calibration.

[0063] It should be noted that zero-point calibration refers to using the baseline data of the voltage channel, current channel, gate channel, and temperature channel as the zero-point calibration reference value, and then calculating the difference between the subsequently acquired measurement values ​​of the voltage channel, current channel, gate channel, and temperature channel and the zero-point calibration reference value to ensure that the measurement output of all channels is recorded from the same starting point.

[0064] The sampling time of the voltage channel, current channel, gate channel and temperature channel is calibrated and summarized into a test measurement chain.

[0065] Furthermore, based on the metrological reference table, the voltage channel, current channel, gate channel, and temperature channel are synchronized one by one in terms of timing. A unified clock source is set, and the sampling time of the voltage channel, current channel, gate channel, and temperature channel is calibrated according to the unified clock source to ensure that the sampling start time of the voltage channel, current channel, gate channel, and temperature channel is consistent.

[0066] The voltage channel, current channel, gate channel, and temperature channel are combined with the calibration parameters to obtain the test measurement chain.

[0067] The metrology reference table and the test metrology chain are merged and registered to obtain the metrology reference master file.

[0068] S2. Perform rapid prediction of the turn-off and commutation electric fields in the main metrology reference file, identify weak medium zones, unify medium reliability indicators, and summarize them into a medium zone stress monitoring table.

[0069] The gallium oxide functional epitaxial layer is layered according to the metrological reference master file. The voltage changes generated by the metallization layer activation operation are distributed according to the layer ratio to obtain the electric field value. The step transition point is determined based on the crystallographic orientation and step characteristic parameters, and the electric field concentration point is identified. The start and end points of the dielectric overlay are extracted from the dielectric overlay stack.

[0070] Furthermore, based on the thickness and doping parameters, crystallographic orientation and step characteristic parameters, and metallization layer of the gallium oxide functional epitaxial layer in the metrology master file, the turn-off and commutation electric fields of the gallium oxide semiconductor material are rapidly estimated. The gallium oxide functional epitaxial layer is layered according to the epitaxial layer thickness to obtain the layering results. The metallization layer of the gallium oxide semiconductor material is turned on, and the voltage change during the turn-on operation is recorded. According to the proportion of each layer in the gallium oxide functional epitaxial layer in the layering results, the voltage change is distributed to each layer. The ratio of the voltage change distributed to each layer to the thickness of the gallium oxide functional epitaxial layer is calculated to obtain the electric field value at each layer. The location where the step height or step direction changes is set as the step turning point. The region where the electric field value is continuously higher than that of the surrounding layers is divided into the electric field concentration area. The dielectric overlay stack of the gallium oxide semiconductor material is read, and the start and end overlay positions of the dielectric overlay stack are extracted as the start and end points of the dielectric overlay.

[0071] By comparing the electric field values ​​at the step transition point, the electric field concentration point, and the start and end points of dielectric coverage with the electric field values ​​at the surrounding locations, the sidewall passivation transition region, the gate oxide critical region, and the terminal inflection point are obtained.

[0072] Furthermore, the electric field value at the step transition point is compared with the electric field values ​​at the surrounding locations. If the electric field value at the step transition point is larger than the electric field values ​​at the surrounding locations, then the step transition point is designated as the sidewall passivation transition zone. The electric field value at the electric field concentration point is compared with the electric field values ​​at the surrounding locations. If the electric field value at the electric field concentration point is larger than the electric field values ​​at the surrounding locations, then the electric field concentration point is designated as the gate oxide critical region. The electric field values ​​at the start and end of the dielectric coverage are compared with the electric field values ​​at the surrounding locations. If the electric field value at the start or end of the coverage is larger than the electric field values ​​at the surrounding locations, then the start or end of the coverage is designated as the terminal inflection point.

[0073] The sidewall passivation transition region, gate oxide critical region, and terminal inflection point are uniformly registered as a list of dielectric-sensitive locations. The electric field values ​​of each location in the list of dielectric-sensitive locations are compared one by one with the electric field values ​​of other locations outside the list of dielectric-sensitive locations to obtain the weak dielectric partitions of gallium oxide semiconductor materials.

[0074] The sidewall passivation transition region, gate oxide critical region, and terminal inflection point are registered to obtain a list of dielectric-sensitive locations. The electric field values ​​of the sidewall passivation transition region, gate oxide critical region, and terminal inflection point in the list of dielectric-sensitive locations are compared one by one with the electric field values ​​of other locations outside the list of dielectric-sensitive locations. The locations of the electric field values ​​of the sidewall passivation transition region, gate oxide critical region, and terminal inflection point in the list of dielectric-sensitive locations that are higher than the electric field values ​​of other locations are set as weak dielectric partitions of gallium oxide semiconductor material.

[0075] Leakage current, threshold drift, small-signal capacitance change, and breakdown event records are read for the gate oxide critical region, sidewall passivation transition region, and terminal inflection point in the weak medium partition, and the results are used to determine whether the medium is in an unacceptable state or an acceptable state.

[0076] The gate oxide critical region, sidewall passivation transition region, and terminal inflection point of the weak dielectric partition are judged by leakage current, threshold drift, small-signal capacitance change, and breakdown event record. If any of the following conditions are met in the gate oxide critical region, sidewall passivation transition region, and terminal inflection point: leakage current is greater than the baseline data of the current channel under no-operation state, threshold drift is not equal to zero, small-signal capacitance change is not equal to zero, or a breakdown event record is found, then the gate oxide critical region, sidewall passivation transition region, and terminal inflection point are judged to be in an unacceptable state; otherwise, the gate oxide critical region, sidewall passivation transition region, and terminal inflection point are judged to be in an acceptable state.

[0077] It should be noted that the requirements for collecting leakage current are as follows: when the metallization layer is turned on, the current value of the current channel in the stable section is read as the leakage current; the requirements for collecting threshold drift are as follows: before and after the metallization layer is turned on, the gate signal voltage corresponding to the gate channel is measured once, and the difference between the gate signal voltage before and after the operation is used as the threshold drift; the requirements for collecting small-signal capacitance change are as follows: the gate channel is applied twice with a gate signal that will not cause a change in the turn-on characteristics, and the voltage-current response of the voltage channel and the current channel under these two applications is read respectively. The difference between the two voltage-current responses is calculated to obtain the small-signal capacitance change; the requirements for collecting breakdown event records are as follows: when the metallization layer is turned on, if a sudden change occurs in the voltage channel or the current channel and cannot be eliminated by zero-point correction, the sudden change is recorded as a breakdown event.

[0078] In the critical region of gate oxide, the sidewall passivation transition region, and the terminal inflection point, records of media failure and media satisfaction are registered separately according to whether the media is in an unpermitted state or in a permitted state, and then merged into a unified media reliability index.

[0079] Furthermore, the gate oxide critical region, sidewall passivation transition region, and terminal inflection point, which are determined to be in an unacceptable state of the medium, are registered together with the corresponding leakage current exceeding the baseline data of the current channel under no-operation conditions, threshold drift not equal to zero, small-signal capacitance change not equal to zero, and breakdown event records as medium non-compliance records. The four conditions in the medium non-compliance records, namely leakage current exceeding the baseline data of the current channel under no-operation conditions, threshold drift not equal to zero, small-signal capacitance change not equal to zero, and breakdown event records, are used as the criteria for medium disallowability. The gate oxide critical region, sidewall passivation transition region, and terminal inflection point, which are determined to be in an acceptable state of the medium, are registered as medium-acceptance records, wherein the leakage current is not exceeding the baseline data under no-operation conditions, threshold drift is equal to zero, small-signal capacitance change is equal to zero, and there are no breakdown event records. The leakage current of the medium-acceptance records is not exceeding the baseline data under no-operation conditions, threshold drift is equal to zero, small-signal capacitance change is equal to zero, and there are no breakdown event records, which are used as the criteria for medium acceptance. The criteria for medium disallowability and medium acceptance are integrated into a unified medium reliability index.

[0080] By summarizing the unified medium reliability index and the list of medium-sensitive locations, a medium zoning stress monitoring table is obtained.

[0081] S3. Implement online monitoring of zoned stress in the medium zoned stress monitoring table, determine the upper limit of the field strength before failure, map the safe operating condition domain, and generate a dynamic safe operating condition upper limit table.

[0082] The dielectric zone stress monitoring table is turned on in the metallization layer to simultaneously acquire voltage changes, current changes, gate signal changes and temperature changes, and obtain the corresponding leakage current, threshold drift, small signal capacitance changes and breakdown event records.

[0083] Furthermore, the gate oxide critical region, sidewall passivation transition region, and terminal inflection point of the dielectric partition stress monitoring table are read as detection locations. Simultaneously, the corresponding turn-on operation, leakage current acquisition requirements, threshold drift acquisition requirements, small-signal capacitance change acquisition requirements, and breakdown event record acquisition requirements are read. The zero-point calibration results of the voltage channel, current channel, gate channel, and temperature channel already obtained from the metrological reference main file are called. A turn-on operation is performed on the detection location on the metallization layer, while simultaneously reading the voltage changes, current changes, gate signal changes, and temperature changes generated by the voltage channel, current channel, gate channel, and temperature channel. Based on the leakage current acquisition requirements, threshold drift acquisition requirements, small-signal capacitance change acquisition requirements, and breakdown event record acquisition requirements, the leakage current, threshold drift, small-signal capacitance change, and breakdown event record at the detection location are read respectively, and the detection location is saved correspondingly with the leakage current, threshold drift, small-signal capacitance change, and breakdown event record.

[0084] The leakage current, threshold drift, small signal capacitance change, and breakdown event records at each detection location are compared with the unified medium reliability index to confirm the upper limit of the pre-failure field strength or the allowable value of the upper limit of the pre-failure field strength at the detection location.

[0085] Furthermore, the electric field values ​​of the detection location in the list of medium-sensitive locations are read. The electric field values ​​of the detection location in the list of medium-sensitive locations, the corresponding saved leakage current, threshold drift, small signal capacitance change and breakdown event records are compared with the unified medium reliability index one by one. If any one of them meets the judgment condition that the medium does not allow, the electric field value of the detection location in the list of medium-sensitive locations is determined as the upper limit of the field strength before failure of the detection location. If all four comparison results meet the judgment condition that the medium allows, the electric field value of the detection location in the list of medium-sensitive locations is determined as the allowable value of the upper limit of the field strength before failure of the detection location.

[0086] The upper limit of the field strength before failure and the allowable value of the upper limit of the field strength before failure are converted into the bus voltage range, gate bias range, voltage change time range and junction temperature range respectively, and then combined into a safe operating condition domain.

[0087] Furthermore, the upper limit of the pre-failure field strength or the allowable value of the upper limit of the pre-failure field strength at the detection location is multiplied by the thickness of the gallium oxide functional epitaxial layer at the detection location to obtain the bus voltage range; the gate signal change at the detection location is taken as the gate bias range; the time interval between the start time and the end time of the voltage change at the detection location is taken as the voltage change time range; and the temperature change at the detection location is taken as the junction temperature range.

[0088] The bus voltage range, gate bias range, voltage change time range, and junction temperature range are combined into a safe operating condition domain.

[0089] By summarizing all safety operating condition domains, a dynamic safety operating condition upper limit table is obtained.

[0090] S4. Based on the dynamic safety operating condition upper limit table, build a heat stack matrix, formulate temperature measurement calibration specifications, and sample according to the test measurement chain, and output a statistical table of synchronous electrothermal measurement data.

[0091] The junction temperature ranges of each detection location in the dynamic safety condition upper limit table are classified and junction temperature targets are extracted. For the junction temperature targets, surface heat dissipation contact layers that can make temperature changes fall within the junction temperature range are selected. After registration, the junction temperature targets are sorted from low to high to form a heat stack matrix.

[0092] Furthermore, the junction temperature ranges given for each detection location in the dynamic safety condition upper limit table are classified. Detection locations with identical upper and lower limits of junction temperature range are grouped into the same group. Detection locations with identical upper or lower limits of junction temperature range are grouped into adjacent groups. Detection locations with different upper and lower limits of junction temperature range but overlapping intervals are merged according to the overlapping intervals. The center value of the junction temperature range or overlapping interval of each group is used as the junction temperature target.

[0093] Junction temperature targets are extracted sequentially from low to high according to the lower limit of the junction temperature range. For each junction temperature target, the same activation operation is performed on the surface heat dissipation contact layer of different materials and thicknesses, and the temperature change of the temperature channel is read. The surface heat dissipation contact layer type that can make the temperature change of the temperature channel fall within the junction temperature range corresponding to the junction temperature target is selected. The junction temperature targets and their corresponding surface heat dissipation contact layers are registered one by one and sorted from low to high according to the junction temperature target to form a heat stack matrix.

[0094] According to the target junction temperature, the recording order of the temperature before operation, the starting temperature of operation, the temperature during operation, and the temperature after operation is specified, and zero-point calibration is performed for each, thus forming a temperature measurement calibration specification.

[0095] Furthermore, a temperature measurement calibration specification is formulated based on the junction temperature target, specifying the recording order of the temperature channels. The temperature of the temperature channel is recorded once before the metallization layer is turned on, as the pre-operation temperature; the temperature of the temperature channel is recorded once at the same moment as the start of the voltage change in the voltage channel, as the operation start temperature; the temperature of the temperature channel is recorded once at the same moment as the end of the voltage change in the voltage channel, as the operation process temperature; and the temperature of the temperature channel is recorded again after the voltage channel, current channel, and gate channel have returned to a stable state, as the post-operation temperature. The difference between the pre-operation temperature, operation start temperature, operation process temperature, and post-operation temperature and the zero-point calibration reference value of the temperature channel is calculated to ensure that the temperature change is calculated using the same reference value.

[0096] The voltage, current, gate, and temperature channels of the test metering chain are invoked to perform a turn-on operation on the metallization layer and perform zero-point calibration to obtain electrothermal synchronization data records corresponding to the junction temperature target and the surface heat dissipation contact layer.

[0097] Furthermore, the test metering chain synchronously samples the voltage, current, gate, and temperature channels. It selects the junction temperature target and corresponding surface heat dissipation contact layer one by one in the order of the thermal stack matrix. A single turn-on operation, consistent with the dynamic safety operating condition upper limit table, is performed on the metallization layer. Simultaneously, the voltage, current, gate signal, and temperature values ​​of the voltage, current, gate, and temperature channels are recorded. The recorded voltage and current values ​​are then compared with the zero-point correction reference values ​​for the voltage and current channels, respectively, to obtain the voltage channel difference result and the current channel difference result. Similarly, the temperature value is compared with the zero-point correction reference value for the temperature channel to obtain the temperature channel difference result. Finally, the voltage channel difference result, current channel difference result, and temperature channel difference result are summarized to obtain the electrothermal synchronous data record corresponding to the junction temperature target and surface heat dissipation contact layer.

[0098] The electrothermal synchronization data were sorted according to the junction temperature target, and the surface heat dissipation contact layer was labeled one by one. The data were then summarized with the temperature measurement calibration specifications to obtain a summary table of electrothermal synchronization measurement data.

[0099] S5. Based on the statistical table of synchronous electrothermal measurement data, construct an equivalent thermal resistance and thermal capacity network, calculate the temperature rise drift coefficient and perform loss conversion at the same time, and obtain a summary table of reference thermal boundary normalization results.

[0100] The loss is calculated by summarizing the electrothermal synchronous measurement data. The temperature after operation and the temperature during operation are zero-point calibrated and compared with the loss to form an equivalent thermal resistance set and an equivalent thermal capacity set, which are then registered as an equivalent thermal resistance and thermal capacity network.

[0101] Furthermore, the electrothermal synchronous measurement data records belonging to the same junction temperature target but corresponding to different surface heat dissipation contact layers are read one by one from the electrothermal synchronous measurement data table. The voltage channel difference result and current channel difference result of the current electrothermal synchronous data record are extracted. The voltage channel difference result and current channel difference result are multiplied to calculate the loss of the current electrothermal synchronous data record. The post-operation temperature in the temperature measurement calibration specification corresponding to the current electrothermal synchronous data record is read. The difference between the post-operation temperature and the zero-point correction reference value of the temperature channel is calculated to obtain the post-operation temperature change. The post-operation temperature change is correspondingly recorded with the loss of the current electrothermal synchronous data record to obtain the correspondence between the loss of the current surface heat dissipation contact layer and the post-operation temperature change. All electrothermal synchronous data records under the same junction temperature target are processed. The equivalent thermal resistance set of the junction temperature target is obtained by calculating the ratio of the temperature change after operation to the loss of each record. The operating process temperature in the temperature measurement calibration specification corresponding to the current electrothermal synchronization data record is read, and the difference between the operating process temperature and the zero-point correction reference value of the temperature channel is calculated to obtain the operating process temperature change. The ratio of the operating process temperature change to the loss is calculated to obtain the equivalent heat capacity value of the current electrothermal synchronization data record. For all electrothermal synchronization data records under the same junction temperature target, the ratio of the operating process temperature change to the loss of each record is calculated to obtain the equivalent heat capacity set of the junction temperature target. The equivalent thermal resistance set and the equivalent heat capacity set are combined and registered according to the order of junction temperature target and surface heat dissipation contact layer to generate the equivalent thermal resistance and heat capacity network of the junction temperature target.

[0102] In the equivalent thermal resistance and thermal capacity network, the surface heat dissipation contact layers are classified and sorted according to the junction temperature target. Adjacent electrothermal synchronous data records are extracted to calculate the temperature change increment and loss increment, and the temperature rise drift coefficient of each surface heat dissipation contact layer is obtained and summarized into a set of temperature rise drift coefficients.

[0103] Furthermore, in the equivalent thermal resistance-capacity network, the electrothermal synchronization data records are classified according to the surface heat dissipation contact layer, and sorted according to the junction temperature target from low to high. Adjacent electrothermal synchronization data records are extracted, and the loss and post-operation temperature change for both low and high junction temperature targets are read. The difference between the post-operation temperature change for the high and low junction temperature targets is calculated to obtain the temperature change increment as the junction temperature target increases under the same surface heat dissipation contact layer. The high junction temperature target is then further analyzed. The difference between the loss of the target temperature and the loss of the target low junction temperature is calculated to obtain the loss increment as the junction temperature increases under the same surface heat dissipation contact layer. The ratio of the temperature change increment to the loss increment is calculated to obtain a set of temperature rise drift coefficients corresponding to the surface heat dissipation contact layer. The above calculation process is performed sequentially on other adjacent electrothermal synchronization data records under the same surface heat dissipation contact layer to obtain the temperature rise drift coefficient sequence of the surface heat dissipation contact layer. Then, the temperature rise drift coefficient sequences obtained from different surface heat dissipation contact layers are summarized according to the junction temperature target order to form a set of temperature rise drift coefficients.

[0104] In the electrothermal synchronous measurement data table, a reference thermal boundary conversion benchmark is selected, and the temperature rise drift coefficient set is called to convert the loss of each record of electrothermal synchronous data, generating a reference thermal boundary normalization result summary table.

[0105] Furthermore, in the electrothermal synchronous measurement data table, the electrothermal synchronous data record with the smallest temperature change after operation is identified. The junction temperature target, surface heat dissipation contact layer, loss, equivalent thermal resistance set, and equivalent heat capacity set corresponding to the electrothermal synchronous data record with the smallest temperature change after operation are determined as the reference thermal boundary conversion benchmark. For electrothermal synchronous data records with a junction temperature target higher than the reference thermal boundary conversion benchmark under the same surface heat dissipation contact layer, the temperature rise drift coefficient of the electrothermal synchronous data record that matches the surface heat dissipation contact layer and the corresponding junction temperature target in the temperature rise drift coefficient set is read. The loss of the electrothermal synchronous data record is then compared with that of the electrothermal synchronous data record. The temperature rise drift coefficients are multiplied to obtain the same equivalent loss. For the electrothermal synchronous data records of the surface heat dissipation contact layer that are different from the reference thermal boundary conversion benchmark, the temperature rise drift coefficients corresponding to each junction temperature target registered in the temperature rise drift coefficient set of the surface heat dissipation contact layer are read sequentially. The loss of these electrothermal synchronous data records is multiplied with the corresponding temperature rise drift coefficient to obtain different equivalent loss. Finally, all the same equivalent loss and different equivalent loss are registered together with their respective junction temperature targets, surface heat dissipation contact layers, equivalent thermal resistance sets and equivalent heat capacity sets to obtain the reference thermal boundary normalization result summary table.

[0106] S6. Perform time mapping and period summation on the reference thermal boundary normalization result summary table to generate a gallium oxide power loss assessment report.

[0107] By performing time mapping between the reference thermal boundary normalization result summary table and the electrothermal synchronization data record, a reference thermal boundary normalization result summary table with time sequence is obtained.

[0108] Furthermore, a time mapping is performed on the reference thermal boundary normalization result summary table. Each record in the table, including the junction temperature target, surface heat dissipation contact layer, and the same or different equivalent losses, is read. The electrothermal synchronous measurement data record matching the junction temperature target and surface heat dissipation contact layer is searched in the electrothermal synchronous measurement data summary table. The start and end times of the voltage change corresponding to the turn-on operation in the test measurement chain are read. The same or different equivalent losses in the reference thermal boundary normalization result summary table are matched one-to-one with the read start and end times of the voltage change. This matching process is then performed on the remaining records in the table, establishing a correspondence between all the same and different equivalent losses and the start and end times of the voltage changes that have occurred in the test measurement chain. This results in a reference thermal boundary normalization result summary table with a chronological order.

[0109] The operation group is obtained from the reference thermal boundary normalization result summary table with time sequence information. The number of times the operation group is registered is counted, multiplied with the loss amount, and summed to obtain the total converted loss amount. This total loss amount is then registered with the reference thermal boundary conversion benchmark to form a gallium oxide power loss assessment report.

[0110] Furthermore, the reference thermal boundary normalization result table with time sequence information is traversed. Reference thermal boundary normalization result records with the same voltage change start time and the same voltage change end time are identified as the same turn-on operation. The same or different equivalent losses belonging to the same turn-on operation are grouped into the same operation group. For each operation group, the number of times this operation group is registered in the reference thermal boundary normalization result table is counted. The same or different equivalent losses in this operation group are multiplied by the number of registrations respectively, and the calculation results are summed to obtain the periodic equivalent loss of the current operation group. The periodic equivalent losses of all operation groups are summed in sequence to obtain the total equivalent loss. The total equivalent loss is registered together with the junction temperature target, surface heat dissipation contact layer, equivalent thermal resistance set and equivalent heat capacity set corresponding to the reference thermal boundary conversion benchmark to form a gallium oxide power loss assessment report.

[0111] It should be noted that the reference thermal boundary normalization result is recorded as one data point in the reference thermal boundary normalization result summary table, which includes the junction temperature target, surface heat dissipation contact layer, and the same or different calculated loss.

[0112] In summary, this invention achieves the following: by obtaining a medium partition stress monitoring table, it can provide early warning of local failure risks and provide an accurate data basis for dynamic safe operating condition domain mapping; by obtaining a reference thermal boundary normalization result summary table, it eliminates the impact of thermal boundary condition differences on loss assessment and improves the consistency and comparability of power loss assessment results under different operating conditions.

[0113] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for evaluating the power loss of gallium oxide semiconductor materials, characterized in that: include, A metrology reference table for gallium oxide semiconductor materials is collected, and the metrology reference table is processed to achieve metrology chain consistency, resulting in a test metrology chain. This chain is then summarized with the metrology reference table to obtain the metrology reference master file. The main metrology reference file is used to quickly predict the turn-off and commutation electric fields, identify weak medium zones, unify medium reliability indicators, and summarize them into a medium zone stress monitoring table. Implement online monitoring of zoned stress in the medium zoned stress monitoring table, determine the upper limit of the field strength before failure, map the safe operating condition domain, and generate a dynamic safe operating condition upper limit table. A heat stack matrix is ​​built based on the dynamic safety operating condition upper limit table, temperature measurement calibration specifications are formulated and sampling is performed according to the test and measurement chain, and a statistical table of synchronous electrothermal measurement data is output. Based on the statistical table of synchronous electrothermal measurement data, an equivalent thermal resistance and thermal capacity network is constructed, and the temperature rise drift coefficient is calculated while loss is converted to obtain a general table of reference thermal boundary normalization results. In the equivalent thermal resistance and thermal capacity network, the surface heat dissipation contact layers are classified and sorted according to the junction temperature target. Adjacent electrothermal synchronous data records are extracted to calculate the temperature change increment and loss increment, and the temperature rise drift coefficient of each surface heat dissipation contact layer is obtained and summarized into a set of temperature rise drift coefficients. In the electrothermal synchronous measurement data table, find the electrothermal synchronous data record with the smallest temperature change after operation. Determine the junction temperature target, surface heat dissipation contact layer, loss, equivalent thermal resistance set, and equivalent heat capacity set corresponding to this record as the reference thermal boundary conversion benchmark. For electrothermal synchronous data records with a junction temperature target higher than the reference thermal boundary conversion benchmark under the same surface heat dissipation contact layer, read the temperature rise drift coefficient of the electrothermal synchronous data record that matches the surface heat dissipation contact layer and corresponding junction temperature target from the temperature rise drift coefficient set. Then, compare the loss of the electrothermal synchronous data record with the temperature rise drift coefficient of the electrothermal synchronous data record. The temperature rise drift coefficients are multiplied to obtain the same equivalent loss. For the electrothermal synchronous data records of the surface heat dissipation contact layer that are different from the reference thermal boundary conversion benchmark, the temperature rise drift coefficients corresponding to each junction temperature target registered in the temperature rise drift coefficient set of the surface heat dissipation contact layer are read sequentially. The loss of these electrothermal synchronous data records is multiplied with the corresponding temperature rise drift coefficient to obtain different equivalent loss. Finally, all the same equivalent loss and different equivalent loss are registered together with their respective junction temperature targets, surface heat dissipation contact layers, equivalent thermal resistance sets and equivalent heat capacity sets to obtain the reference thermal boundary normalization result summary table. The reference thermal boundary normalization result summary table is time-mapped and periodically summed to generate a gallium oxide power loss assessment report.

2. The power loss assessment method for gallium oxide semiconductor materials as described in claim 1, characterized in that: The structure of the gallium oxide semiconductor material includes a gallium oxide functional epitaxial layer, a dielectric overlay stack, a metallization layer, and a surface heat dissipation contact layer. The surface heat dissipation contact layer includes the material and thickness of the surface heat dissipation contact layer; The metrological reference table includes the thickness and doping parameters, crystallographic orientation, and step characteristic parameters of the gallium oxide functional epitaxial layer.

3. The power loss assessment method for gallium oxide semiconductor materials as described in claim 2, characterized in that: The steps to obtain the master file of the metrological standard are as follows: Voltage measurement points, current measurement points, gate signal measurement points, and temperature measurement points are connected to the metallization layer to obtain voltage channels, current channels, gate channels, and temperature channels. Zero-point calibration, phase calibration, and bandwidth calibration are performed on the voltage channels, current channels, gate channels, and temperature channels. The sampling time of the voltage channel, current channel, gate channel and temperature channel is calibrated and summarized into a test measurement chain; The metrology reference table and the test metrology chain are merged and registered to obtain the metrology reference master file.

4. The power loss assessment method for gallium oxide semiconductor materials as described in claim 3, characterized in that: The steps for identifying vulnerable media partitions are as follows: Based on the main metrological reference file, the gallium oxide functional epitaxial layer is layered, and the voltage change generated by the metallization layer is distributed according to the layer ratio to obtain the electric field value. The step transition point is determined based on the crystallographic orientation and step characteristic parameters, and the electric field concentration point is identified. The start and end points of the dielectric cover are extracted from the dielectric cover stack. By comparing the electric field values ​​at the step transition point, electric field concentration point, and dielectric coverage start and end point with the surrounding locations, the sidewall passivation transition region, gate oxide critical region, and terminal inflection point are obtained. The sidewall passivation transition region, gate oxide critical region, and terminal inflection point are uniformly registered as a list of dielectric-sensitive locations. The electric field values ​​of each location in the list of dielectric-sensitive locations are compared one by one with the electric field values ​​of other locations outside the list of dielectric-sensitive locations to obtain the weak dielectric partitions of gallium oxide semiconductor materials.

5. The power loss assessment method for gallium oxide semiconductor materials as described in claim 4, characterized in that: The summary is compiled into a medium-zone stress monitoring table, and the steps are as follows. Leakage current, threshold drift, small signal capacitance change, and breakdown event records are read for the gate oxide critical region, sidewall passivation transition region, and terminal inflection point in the weak medium partition, and the medium is judged to be in an unacceptable state or an acceptable state. In the critical region of gate oxide, the sidewall passivation transition region and the terminal inflection point, records of media failure and media satisfaction are recorded separately according to whether the media is in an unpermitted state or in a permitted state, and then merged into a unified media reliability index. By summarizing the unified medium reliability index and the list of medium-sensitive locations, a medium zoning stress monitoring table is obtained.

6. The power loss assessment method for gallium oxide semiconductor materials as described in claim 5, characterized in that: The steps for generating the dynamic safety condition upper limit table are as follows: The dielectric partition stress monitoring table is turned on in the metallization layer, and voltage change, current change, gate signal change and temperature change are acquired simultaneously. At the same time, the corresponding leakage current, threshold drift, small signal capacitance change and breakdown event record are obtained. The leakage current, threshold drift, small signal capacitance change and breakdown event records at each detection location are compared with the unified medium reliability index to confirm the upper limit of the pre-failure field strength and the allowable value of the upper limit of the pre-failure field strength at the detection location. The upper limit of the field strength before failure and the allowable value of the upper limit of the field strength before failure are converted into the bus voltage range, gate bias range, voltage change time range and junction temperature range respectively, and then merged into a safe operating condition domain. By summarizing all safety operating condition domains, a dynamic safety operating condition upper limit table is obtained.

7. The power loss assessment method for gallium oxide semiconductor materials as described in claim 6, characterized in that: The steps for compiling the output electrothermal synchronous measurement data are as follows: The junction temperature ranges of each detection location in the dynamic safety condition upper limit table are classified and junction temperature targets are extracted. For the junction temperature targets, surface heat dissipation contact layers that can make temperature changes fall within the junction temperature range are selected. After registration, the junction temperature targets are sorted from low to high to form a heat stack matrix. According to the junction temperature target, the recording order of the temperature before operation, the starting temperature of operation, the temperature during operation, and the temperature after operation is specified, and zero-point calibration is performed for each, forming a temperature measurement calibration specification. The voltage, current, gate, and temperature channels of the test metering chain are invoked to perform a turn-on operation on the metallization layer and perform zero-point calibration to obtain electrothermal synchronous data records corresponding to the junction temperature target and the surface heat dissipation contact layer. The electrothermal synchronization data were sorted according to the junction temperature target, and the surface heat dissipation contact layer was labeled one by one. The data were then summarized with the temperature measurement calibration specifications to obtain a summary table of electrothermal synchronization measurement data.

8. The power loss assessment method for gallium oxide semiconductor materials as described in claim 7, characterized in that: The construction of the equivalent thermal resistance and thermal capacity network specifically refers to calculating the loss amount by compiling the electrothermal synchronous measurement data, performing zero-point correction on the post-operation temperature and the operation process temperature respectively, and forming the equivalent thermal resistance set and the equivalent thermal capacity set with the ratio of the loss amount, which are then registered as the equivalent thermal resistance and thermal capacity network.

9. The power loss assessment method for gallium oxide semiconductor materials as described in claim 8, characterized in that: The steps to obtain the summary table of reference thermal boundary normalization results are as follows: In the equivalent thermal resistance and thermal capacity network, the surface heat dissipation contact layers are classified and sorted according to the junction temperature target. Adjacent electrothermal synchronous data records are extracted to calculate the temperature change increment and loss increment, and the temperature rise drift coefficient of each surface heat dissipation contact layer is obtained and summarized into a set of temperature rise drift coefficients. In the electrothermal synchronous measurement data table, a reference thermal boundary conversion benchmark is selected, and the temperature rise drift coefficient set is called to convert the loss of each record of electrothermal synchronous data, generating a reference thermal boundary normalization result summary table.

10. The power loss assessment method for gallium oxide semiconductor materials as described in claim 9, characterized in that: The steps for generating the gallium oxide power loss assessment report are as follows: By performing time mapping between the reference thermal boundary normalization result summary table and the electrothermal synchronization data record, a reference thermal boundary normalization result summary table with time sequence is obtained. The operation group is obtained from the reference thermal boundary normalization result summary table with time sequence information. The number of times the operation group is registered is counted, multiplied with the loss amount, and summed to obtain the total converted loss amount. This total loss amount is then registered with the reference thermal boundary conversion benchmark to form a gallium oxide power loss assessment report.

Images