Detection method for potential electric leakage defect of LED device and LED sorting equipment

Abstract

The invention discloses a method for detecting potential electric leakage defects of an LED device, and relates to the technical field of semiconductor photoelectric device detection. Aiming at the technical pain point that recessive microdefects (such as a metal migration channel caused by a V-shaped pit and dislocation) cannot be identified by the existing single spectroscopic detection, the invention provides a detection logic of reference parameter acquisition, defect stress excitation, parameter acquisition after excitation and differential analysis and judgment. Specific stress is applied to primarily screened qualified products, ion migration at potential defects is excited in an oriented mode, and hidden electric leakage is dominated. Potential defective products are accurately rejected by calculating parameter drift distances (VFL, IR) before and after stress. The method is not only suitable for interactive screens, but also suitable for high-reliability scenes such as vehicle-mounted display and high-end illumination, and can effectively intercept hidden lamp beads causing'caterpillar 'failure or early light attenuation, thereby remarkably improving the product yield and the terminal reliability.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor optoelectronic device testing technology, specifically relating to a method for detecting potential leakage defects in LED devices and an LED sorting device. Background Technology

[0002] Light-emitting diodes (LEDs) have become core components in modern lighting and display fields due to their advantages such as high luminous efficiency, long lifespan, small size, and fast response speed. As end products develop towards miniaturization (Mini / Micro LED), high-density integration, and intelligence, the market demands "zero defects" in the reliability of LED devices. In applications such as small-pitch displays and interactive screens, tens of thousands of LED beads are densely arranged in a single screen. The failure of any single LED bead can lead to display abnormalities such as "caterpillar" patterns (constantly lit or dark spots), dead LEDs, or signal crosstalk, severely affecting the display effect and incurring high after-sales maintenance costs due to the need for complete screen disassembly for repairs.

[0003] LED chips are mainly composed of a P-type semiconductor layer, an active layer (MQW), and an N-type semiconductor layer. During epitaxial growth and chip manufacturing processes, microscopic defects such as lattice dislocations, V-pits, sidewall damage, or metal ion migration channels inevitably occur in the epitaxial layer due to factors such as substrate lattice mismatch, differences in thermal expansion coefficients, and process fluctuations. These defects often exhibit "latent" characteristics in their initial state: the defect channels are in a high-resistivity state or temporarily covered by a surface passivation layer, making them difficult to identify with conventional electrical tests.

[0004] In existing routine production and testing processes, the industry primarily relies on "single-shot spectral testing" as a screening method to identify defective products. This involves measuring the startup voltage (VFL) under a specific forward microcurrent and detecting the leakage current (IR) under reverse bias. However, this method is essentially a static, instantaneous detection, reflecting only the electrical parameters at the moment of testing, and has significant limitations. Many chips containing microscopic defects pass all parameters during factory testing and are deemed "zero-moment good products." However, during subsequent SMT reflow soldering (a high-temperature process above 200°C) or long-term use in end-user devices, driven by both electric and thermal fields, the electrode metals (such as Ag, Al, Au, etc.) undergo electrochemical migration along the aforementioned microscopic defect channels, gradually forming conductive filaments or deep-level recombination centers. This leads to a sharp decrease in the parallel resistance of the PN junction, causing a surge in leakage current and ultimately resulting in functional failure. The aforementioned failure process exhibits significant time lag and stress dependence. Existing detection processes are unable to stimulate the dynamic evolution of defects or transform latent risks into identifiable explicit parameter changes, resulting in a large number of "zero-time good products" randomly failing in end products. This not only degrades display quality but also becomes a key bottleneck restricting the mass production and application of high-reliability Mini / MicroLEDs.

[0005] Therefore, there is an urgent need for a universal detection method that can simulate the stress of actual applications, accelerate the potential failure process, and transform latent defects into explicit parameter changes, so as to accurately intercept "latent" defects at the manufacturing stage. Summary of the Invention

[0006] The purpose of this invention is to provide a method for detecting potential leakage defects in LED devices and an LED sorting device, which solves the problem that a single spectral test in the prior art cannot identify potential leakage defects in LED devices, and achieves non-destructive and efficient screening of potential defective products.

[0007] The technical solution of the present invention is as follows: In a first aspect, the present invention provides a method for detecting potential leakage defects in LED devices, comprising the following steps: S1. Obtain the LED device under forward test current I. F The initial positive voltage V FL1 and in reverse test voltage V R The initial reverse leakage current I under R1 ; S2. Apply a preset stress to the LED device to induce physical degradation at the micro-defects; the stress parameter value is greater than the safe operating boundary value of the LED device and less than the failure threshold of the LED device. S3, under forward test current I F and reverse test voltage V RUnder these conditions, the forward voltage V of the LED device after stress treatment is obtained. FL2 With reverse leakage current I R2 ; S4. Calculate the change in forward voltage ΔV FL =|V FL2 -V FL1 | and / or the change in reverse leakage current ΔI R =|I R2 -I R1 |, if △V FL and / or △I R If the threshold value is exceeded, the LED device is determined to have a potential leakage risk.

[0008] The detection method provided by this invention applies a specific intensity of excitation stress to LED devices, inducing physical degradation at microscopic defects. This transforms latent microscopic defects into detectable electrical differences, directly targeting leakage risks caused by physical degradation and effectively filling the blind spots of traditional static testing. The specific mechanism is as follows: For LED devices with microscopic defects, stress treatment generates energy at the defect location. This energy is sufficient to drive unstable metal ions (such as Ag+) to undergo minute displacement or overcome potential barriers to form micro-leakage channels, resulting in a decrease in the forward voltage (VFL) or an increase in the reverse leakage current (IR) of the LED device with microscopic defects. By controlling the stress intensity within the aforementioned range, high-quality LED devices with good lattice quality will not undergo physical degradation under stress, while LED devices with microscopic defects will experience significant shifts in the forward voltage (VFL) and / or reverse leakage current (IR) under stress. Furthermore, by analyzing the changes, the influence of batch-to-batch process fluctuations is eliminated, focusing on the relative changes of individual devices before and after stress, enabling the capture of minute changes in physical properties.

[0009] As an optional solution to the detection method of the present invention, in step S4, △V FL The preset threshold is 0.05V~0.07V, △I R The preset threshold is 0.01µA to 0.05µA.

[0010] As an optional scheme of the detection method of the present invention, in step S2, physical degradation includes at least one of metal ion migration, conductive filament formation, micro-breakdown of insulating passivation layer or expansion of lattice defects.

[0011] As an optional solution of the detection method of the present invention, in step S2, the stress is one or more of electrical stress, thermal stress, mechanical stress and environmental stress.

[0012] As a preferred option among the above-mentioned alternatives, the electrical stress is selected from at least one of forward high current impact, reverse high voltage impact, and pulse current impact; the thermal stress is selected from at least one of high temperature impact, low temperature freezing, and thermal cycling impact; the mechanical stress is ultrasonic vibration and / or mechanical pressure; and the environmental stress is high humidity environmental stress and / or high pressure environmental stress.

[0013] As a preferred embodiment of the detection method of the present invention, the stress is a positive high current impact stress, and the current intensity of the positive high current impact is 3 to 5 times the rated operating current of the LED device.

[0014] As an optional scheme of the detection method of the present invention, the range of the forward test current IF is set to 0.1μA~50μA, and the test time under the forward test current is 1ms~50ms; the range of the reverse test voltage VR is set to 3V~15V, and the test time under the reverse test voltage is 1ms~50ms.

[0015] As an optional solution to the detection method of the present invention, steps S1 to S4 are integrated into the mass production line and executed sequentially as a continuous online detection process.

[0016] Secondly, the present invention also provides an LED sorting device for performing the above-described detection method, comprising: The testing station is used to perform steps S1 and S3. The stress application station is used to perform step S2; The sorting actuator is used to perform step S4 and shun LED devices that are determined to have a potential leakage risk.

[0017] As an optional embodiment of the LED sorting equipment of the present invention: the stress application station is equipped with at least one of a programmable power supply, a temperature control device, or a mechanical vibration device.

[0018] The LED device testing method provided by this invention has broad applicability, applicable to various LED devices fabricated using semiconductor materials such as gallium nitride (GaN), aluminum gallium indium phosphide (AlGaInP), and gallium arsenide (GaAs), and with packaging structures covering upright, flip-chip, and vertical structures. The testing wavelengths include yellow, red, blue, green, and ultraviolet light. It should be clarified that the applicability of this method is not limited to specific examples of the aforementioned materials, structures, or wavelengths. Its core lies in addressing the common failure mechanisms caused by microscopic defects and physical degradation. All LED devices affected by these two types of failure mechanisms, regardless of their material system, packaging form, or luminous characteristics, can be effectively detected and evaluated using the method of this invention.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, some of which will become clear as the description proceeds, and others will be learned by practicing the invention. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the detection method for defective LED beads in the embodiments of this application.

[0022] Figure 2 This is a FIB-SEM image of the potentially defective LED chips screened in Embodiment 3 of this application (where the right image is a partial enlarged view of the area in the red box in the left image).

[0023] Figure 3 This is a scatter plot of ΔVFL after a high current surge on the yellow light chip in Embodiment 4 of this application.

[0024] Figure 4 This is a FIB-SEM image of the failed LED in Comparative Example 1 of this application (the right image is a magnified view of the area within the red box in the left image). Detailed Implementation

[0025] To make the technical problems solved, the technical solutions, and the beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0026] This application provides a method for detecting potential leakage defects in LED devices, comprising the following steps: S1. Obtain the LED device under forward test current I. F The initial positive voltage V FL1 and in reverse test voltage V R The initial reverse leakage current I under R1 It should be noted that at this stage, obviously defective products with parameters that significantly exceed the specifications can be removed, and the products that pass the initial screening can be retained. S2. Apply a preset stress to the LED device to induce physical degradation at the micro-defects; the stress parameter value is greater than the safe operating boundary value of the LED device and less than the failure threshold of the LED device; it should be noted that the stress is applied to accelerate the physical degradation (e.g., material degradation or ion migration process) at the potential micro-defects by using physical mechanisms (such as Joule heating, strong electric field), so that the latent defects are converted into detectable electrical signal differences, while controlling the stress dosage to ensure that normal good devices are not damaged. S3, under forward test current I F and reverse test voltage V R Under these conditions, the forward voltage V of the LED device after stress treatment is obtained. FL2 With reverse leakage current I R2 It should be noted that, under the same test conditions as in step S1, the photoelectric parameters after stress treatment were tested, controlling a single variable for subsequent comparative analysis. S4. Calculate the change in forward voltage ΔV FL =|V FL2 -V FL1 | and / or the change in reverse leakage current ΔI R =|I R2 -I R1 |, if △V FL and / or △I R If the value exceeds a preset threshold, the LED device is deemed to have a potential leakage risk. It should be noted that the parameter drift of a single LED device before and after stress application is calculated, and the device is judged based on the preset threshold. If ΔV FL or △I R If the value exceeds a preset threshold, the device is deemed to have a potential leakage risk and is rejected; otherwise, it is considered a qualified product. By analyzing the changes, the influence of batch-to-batch process fluctuations is eliminated, focusing on the relative changes of individual devices before and after stress, and capturing minute changes in physical properties.

[0027] As an optional solution to the detection method of the present invention, in step S4, △V FL The preset threshold is 0.05V~0.07V, △I R The preset threshold is 0.01µA to 0.05µA. It should be noted that this preset threshold is set based on statistical data from a large number of well-made devices regarding thermal relaxation characteristics and metal migration mechanisms. After being subjected to a very short-term stress shock, well-made devices mainly produce reversible thermistor voltage fluctuations, with ΔV... FL Typically, this is within 0.04V; however, in defective devices that have undergone physical degradation, the internal parallel resistance will decrease substantially, leading to a significant increase in ΔV. FL A significant voltage jump of 0.05V or higher occurs. For LED devices with suboptimal material properties, ΔV FLThe threshold value can be flexibly adjusted up to 0.07V to balance defect detection sensitivity and yield retention rate. Similarly, for the change in reverse leakage current ΔIR, the reverse blocking characteristics of good devices are extremely stable. Extremely short-term stress shocks will not cause substantial damage to their PN junction depletion layer, and the natural fluctuation of their leakage current is usually within the background noise level of the test instrument (usually less than 0.005µA). However, once physical degradation such as metal ion migration occurs at the defect (such as a V-shaped pit), forming microscopic conductive filaments, it will directly lead to the opening of the reverse micro-short circuit channel, causing a nonlinear surge of ΔIR of more than 0.01µA. Setting the threshold value of ΔIR in the range of 0.01µA to 0.05µA is to flexibly adapt to LED chips of different sizes and specifications. Specifically, for miniaturized chips (such as Mini / MicroLEDs) with extremely low substrate leakage current and stringent reliability requirements, a strict threshold of 0.01µA can be used to ensure zero missed defects. For large-size or high-power chips with larger junction areas and slightly higher surface leakage background, the threshold can be appropriately relaxed to 0.05µA. This threshold range can accurately distinguish between normal thermal fluctuations and actual physical degradation, ensuring a high detection rate for potential defects while effectively avoiding false positives on good products, thus achieving an optimized balance between detection reliability and production yield.

[0028] As an optional embodiment of the detection method of this invention, in step S2, physical degradation includes at least one of metal ion migration, conductive filament formation, micro-breakdown of the insulating passivation layer, or lattice defect propagation. It should be noted that these forms of physical degradation are the core causes of LED devices exhibiting "caterpillar-like" constant brightness, dark spots, or leakage failure after long-term use (especially since flip-chip electrodes are highly susceptible to electrochemical migration of silver ions forming conductive filaments). This invention specifically targets these physical characteristics with directional excitation, directly addressing the failure points and ensuring the scientific validity and effectiveness of the detection method.

[0029] As an optional embodiment of the detection method of this invention, in step S2, the stress is one or more of electrical stress, thermal stress, mechanical stress, and environmental stress. It should be noted that LED devices with different packaging structures (upright / flip / vertical) and different material systems have different most sensitive failure mechanisms. Providing multiple stress options and their combinations makes this detection method highly universal. Testers can flexibly configure the stress type according to the specific characteristics of the device under test, thereby most efficiently activating its specific type of potential defects.

[0030] As a preferred option among the above-mentioned alternatives, the electrical stress is selected from at least one of forward high-current impact, reverse high-voltage impact, and pulsed current impact; the thermal stress is selected from at least one of high-temperature impact, low-temperature freezing, and thermal cycling impact; the mechanical stress is ultrasonic vibration and / or mechanical pressure; and the environmental stress is high-humidity environmental stress and / or high-pressure environmental stress. It should be noted that further refinement of the specific stress form can provide more precise excitation methods. For example, forward high current can provide dual driving forces of Joule heating and electric field, accelerating ion migration; thermal cycling can effectively expose the potential for insulation layer cracking caused by thermal expansion coefficient (CTE) mismatch; and ultrasonic vibration can quickly expose weak solder joint problems in the bonding wires. These specific stress methods ensure the controllability and high reproducibility of the excitation process.

[0031] In a preferred embodiment of the detection method of this invention, the stress is a positive high-current impact stress, and the current intensity of the positive high-current impact is 3 to 5 times the rated operating current of the LED device. It should be noted that this parameter range is crucial for achieving a balance between efficient excitation and non-destructive screening. If the impact current is less than 3 times the rated current, the provided activation energy is insufficient to excite significant ion migration within a millisecond-level detection cycle, leading to missed detections. If it is more than 5 times, the generated transient heat may exceed the chip's heat dissipation limit, causing irreversible thermal breakdown in good products, resulting in false negatives. A current intensity of 3 to 5 times the rated current can both generate a high-intensity field at the defect tip to drive degradation and ensure the safety of good products.

[0032] As an optional scheme of the detection method of the present invention, the forward test current I... F The range of the forward test current is 0.1μA to 50μA, and the test time under the forward test current is 1ms to 50ms; the range of the reverse test voltage VR is 3V to 15V, and the test time under the reverse test voltage is 1ms to 50ms. It should be noted that the specific test parameters can be flexibly selected to meet the testing needs of different specifications of LED devices and diverse production lines.

[0033] As an optional embodiment of the detection method of this invention, steps S1 to S4 are integrated into the mass production line and executed sequentially as a continuous online detection process. It should be noted that by integrating testing into production, quality control is moved upstream, effectively intercepting defects at the source of production, significantly reducing defect rates, rework costs, and quality losses, while simultaneously improving product consistency and overall production line efficiency.

[0034] Secondly, the present invention also provides an LED sorting device for performing the above-described detection method, comprising: The testing station is used to perform steps S1 and S3. The stress application station is used to perform step S2; The sorting actuator is used to perform step S4 and shun LED devices identified as having potential leakage risks. It should be noted that this equipment materializes the above-mentioned theoretical detection method into automated hardware. Through the reasonable flow setting of physical workstations, it achieves a seamless completion of steps S1 to S4, meeting the stringent requirements of large-scale industrial production for automation and high output. It is the best platform for the implementation of the method of this invention.

[0035] As an optional embodiment of the LED sorting equipment of this invention, the stress application station is equipped with at least one of a programmable power supply, a temperature control device, or a mechanical vibration device. It should be noted that, as mentioned above, LED devices with different material systems and packaging structures have different potential failure mechanisms. Therefore, providing various types of stress application hardware allows users to freely select or combine appropriate equipment based on the specific failure mechanism of the LED device to be tested.

[0036] The following are some embodiments of this application, and the embodiments of the present invention will further describe the technical solution of the present invention in detail. Example 1

[0037] This embodiment uses a Y3030 flip-chip yellow LED bead as the test object. Because yellow light materials (such as AlGaInP or GaN with special structures) are prone to forming surface defects such as V-shaped pits during epitaxial growth, and silver (Ag) with high reflectivity is often used as the reflective electrode, they are sensitive to potential leakage electrodes.

[0038] The detection method in this embodiment, such as Figure 1 As shown, it includes the following steps: Step S1, Reference Parameter Acquisition and Initial Screening: 25,085 yellow LED beads from the same production batch were selected as the test samples. A high-precision automated spectrometer was used for the first round of testing to obtain the yellow LED beads at the forward test current I. F The initial positive voltage V FL1 and in reverse test voltage V R The initial reverse leakage current I under R1 According to the specifications, qualified standards are set, obviously defective products are rejected, and products that pass the initial screening are retained.

[0039] The test conditions are: forward current I F =1µA, test duration 5ms; reverse voltage V R =5V, test duration 10ms. The initial screening pass standard is set according to the product specifications: 1.5V ≤ V FL1 ≤1.9V and I R1<0.05µA. After three rounds of testing, 19 obviously defective LED beads with open circuits, short circuits, or parameters exceeding the standard were removed, leaving 25,066 yellow LED beads that passed the initial screening and were all "qualified at the first moment".

[0040] Step S2, High-Current Stress Excitation: A transient high-current shock is applied to the yellow LED beads that passed the initial screening in Step S1. The specific parameters are: forward DC current 300mA, duration 40ms. In this embodiment, the maximum operating current of the yellow LED beads is approximately 100mA. 300mA (3 times the maximum operating current) is selected as the excitation stress. This stress is above the safe operating zone boundary and below the physical damage threshold. It can instantly generate a high-intensity field and Joule heating at the defect tip, directionally accelerating the migration of potential Ag ions. Simultaneously, due to the extremely short duration (40ms), it will not cause thermal damage to the intact crystal lattice of the good product.

[0041] Step S3, Parameter Acquisition After Stress Application: After the stress is applied, the forward voltage V of each yellow LED bead is measured again under the same test conditions as in Step S1. FL2 With reverse leakage current I R2 .

[0042] Step S4: Calculate the change in forward voltage ΔV for each yellow LED bead. FL =|V FL2 -V FL1 | and the change in reverse leakage current ΔI R =|I R2 -I R1 |

[0043] △V of LED beads that pass the initial screening FL and △I R The distribution data were statistically analyzed, and the results are shown in Table 1 below.

[0044] Table 1 ΔV before and after impact FL and △I R Distribution statistics table <![CDATA[△V FL (V)]]> Quantity (pieces) <![CDATA[△I R (µA)]]> Quantity (pieces) <![CDATA[△V FL <0.02]]> 28 <![CDATA[△I R <0.005]]> 24980 <![CDATA[0.02≤△V FL <0.04]]> 21315 <![CDATA[0.005≤△I R <0.01]]> 75 <![CDATA[0.04≤△V FL <0.06]]> 3689 <![CDATA[0.01≤△I R <0.05]]> 6 <![CDATA[0.06≤△V FL <0.07]]> 19 <![CDATA[△I R ≥0.05]]> 5 <![CDATA[0.07≤△V FL <0.10]]> 5 <![CDATA[0.10≤△V FL ]]> 10 total 25066 25066 Step S5, Threshold Determination and Rejection: In this embodiment, the screening criterion is set as △V. FL ≥0.07V or ΔI R Products with an impact strength ≥0.01µA are considered potentially defective. The statistical results after the impact test are as follows: (The following is a list of products that meet the ΔV requirement.) FL There are 15 LED beads with a voltage of ≥0.07V, △I R There are 11 LEDs with an A ≥ 0.01µA voltage, of which 11 have a ΔI value. R All the LED beads that exceed the standard simultaneously meet △V FL Exceeding the standard (i.e., △I) R The sample exceeding the standard is △VFL (A subset of samples exceeding the standard). Therefore, based on this standard, a total of 15 "potentially defective" items were identified after the impact. Example 2

[0045] The main difference between this embodiment and embodiment 1 is that the type of stress applied in step S2 is different. This embodiment uses reflow soldering high temperature simulation as the thermal stress excitation source.

[0046] Step S1, Reference Parameter Acquisition and Initial Screening: Using a batch of yellow LED beads of the same model as in Example 1 and the same test conditions, the reference parameters of the yellow LED beads at the forward test current I are obtained. F The initial positive voltage V FL1 and in reverse test voltage V R The initial reverse leakage current I under R1 And retain V as required by the specifications. FL1 and I R1 There are 24,124 qualified yellow LED beads, all within the normal range.

[0047] Step S2, Thermal Stress Induction: Place the 24,124 yellow LED beads into a reflow oven. Set the parameters according to the standard surface mount manufacturer's oven temperature profile (peak temperature 260℃, isothermal duration matching actual production) to simulate the ultimate thermal stress during the actual surface mount process. This thermal stress is sufficient to cause physical degradation of microscopic defects (such as microcracks in the passivation layer) due to mismatched coefficients of thermal expansion.

[0048] Step S3, Parameter Acquisition After Stress Application: After the LED beads have cooled to room temperature, the forward voltage V of each yellow LED bead is measured again under the same test conditions as in Step S1. FL2 With reverse leakage current I R2 .

[0049] Step S4, Differential Calculation and Judgment: Calculate the change in forward voltage ΔV for each yellow LED bead. FL =|V FL2 -V FL1 | and the change in reverse leakage current ΔI R =|I R2 -I R1 In this embodiment, the screening criterion is also set as △V. FL ≥0.07V or ΔI R Products with an A value ≥0.01µA are considered potentially defective. Based on this standard, a total of 10 potentially defective LEDs were identified. The specific parameters of these 10 defective LEDs are shown in Table 2.

[0050] Table 2 Specific parameters of defective LED chips after reflow soldering Serial Number <![CDATA[I R1 (uA)]]> <![CDATA[V FL1 (V)]]> <![CDATA[I R2 (uA)]]> <![CDATA[V FL2 (V)]]> <![CDATA[△I R (µA)]]> <![CDATA[△V FL (V)]]> 1 0.079 1.81 0.082 1.818 0.003 0.008 2 0.361 1.75 0.247 1.75 0.114 0 3 0.002 1.495 0.0004 1.494 0.0016 0.001 4 0.124 1.751 0.1602 1.757 0.0362 0.006 5 0.002 1.488 0.0005 1.468 0.0015 0.02 6 0.021 1.314 0.0092 1.322 0.0118 0.008 7 0.325 1.807 0.2919 1.812 0.0331 0.005 8 0.002 1.476 0.0017 1.457 0.0003 0.019 9 0.001 1.469 0.0008 1.455 0.0002 0.014 10 0.174 1.743 0.0161 1.699 0.1579 0.044 Example 3

[0051] To verify whether the "potentially defective products" intercepted by the detection method of this application have truly solved the terminal "caterpillar" failure problem, this embodiment classifies the samples in Embodiment 1 and Embodiment 2 and performs accelerated lifetime verification and microphysical analysis.

[0052] High-temperature aging accelerated life test: Three groups of samples were selected for aging tests at 85℃ and 50mA for 24 hours.

[0053] Group A consists of potential defective products detected by electrical stress: 12 potentially defective LED chips rejected in Example 1 were randomly selected. Group B consists of absolutely good products that passed the method: 6 good LED chips judged as qualified in Example 1 were randomly selected. Group C (potential defective products detected by thermal stress): 2 defective LED chips that failed reflow soldering in Example 2 were randomly selected.

[0054] Group A LED beads underwent a high-temperature aging test at 85℃. The parameter changes before and after 24 hours of aging are shown in Table 3 below.

[0055] (Note: "Light decay" in the table refers to the percentage decrease in brightness; △V) FL The change in forward voltage after the aging test compared to before aging; △I R The change in reverse leakage current after the aging test compared to before aging; ΔV F2 The change in forward switching voltage after the aging test compared to before aging; △W LD (Refers to the change in the dominant wavelength of the LED after the aging test compared to before aging) Table 3. Statistical table of parameter changes of Group A LED beads after 24 hours of aging. Serial Number 24H50mA optical decay <![CDATA[△I R ]]> <![CDATA[△V FL ]]> <![CDATA[△V F2 ]]> <![CDATA[△W LD ]]> 1 0.74% 0.0001 0.0000 0.00 -0.1 2 3.55% -0.7193 0.5796 -0.01 0.1 3 2.26% -0.0001 -0.0080 0.00 0.0 4 2.36% 0.0000 0.1470 0.00 0.0 5 1.27% -0.0002 -0.0050 0.00 0.0 6 3.87% 0.0000 -0.0020 0.00 -0.1 7 2.71% 0.0033 0.0620 0.00 -0.1 8 3.92% 0.0068 0.8813 0.00 0.1 9 2.35% 0.0078 0.0330 0.00 -0.1 10 1.56% -0.0002 0.0430 0.00 -0.1 11 1.81% 0.0000 -0.0020 0.00 0.0 12 1.35% 0.0001 0.0320 0.00 -0.1 average value 2.65% -0.06 0.15 0.00 -0.02 Group A LEDs all exhibited varying degrees of parameter degradation after aging, with an average light decay of 2.65% and ΔV. FL The average value is 0.15V, ΔI R The average value was 0.06uA, which verified that the defective LEDs screened in Example 1 did indeed have potential problems.

[0056] Group B LED beads underwent a high-temperature aging test at 85℃. The parameter changes before and after 24 hours of aging are shown in Table 4 below.

[0057] Table 4. Statistics on parameter changes of Group B LED beads after 24 hours of aging. Serial Number 24H50mA optical decay <![CDATA[△I R ]]> <![CDATA[△V FL ]]> <![CDATA[△V F2 ]]> <![CDATA[△W LD ]]> 1 0.00% 0.0001 0.0060 -0.02 0.0 2 0.00% 0.0001 0.0020 -0.02 0.0 3 0.00% -0.0003 0.0030 -0.01 0.0 4 0.00% -0.0003 0.0000 -0.01 0.0 5 0.00% -0.0001 0.0010 -0.01 0.0 6 0.00% 0.0001 -0.0030 -0.01 0.0 average value 0.00% 0.00 0.00 -0.01 0.00 After the group B LED beads age, △V FL , △I R , △V F2 and △W LDThere were no obvious changes, confirming that the good quality LED beads were normal.

[0058] Group C LED beads underwent a high-temperature aging test at 85℃. The parameter changes before and after 24 hours of aging are shown in Table 5 below. Table 5. Statistics on parameter changes of Group C LED beads after 24 hours of aging. Serial Number 24H50mA optical decay <![CDATA[△I R ]]> <![CDATA[△V FL ]]> <![CDATA[△V F2 ]]> <![CDATA[△W LD ]]> 1 3.03% 0.0021 0.1330 0.00 -0.1 2 0.09% 0.0002 0.0240 0.00 -0.1 average value 1.56% 0.001 0.08 0.00 -0.10 After aging, all group C LEDs exhibited varying degrees of parameter degradation, with an average light decay of 1.56% and ΔV. FL The average value is 0.08V, ΔI R The average value was 0.001uA, verifying that the defective LEDs screened in Example 2 did indeed have potential problems. Based on the comparative results of aging tests on the three groups of LEDs: samples judged as defective by the detection method of this application all showed performance degradation or failure after aging, while samples judged as good remained stable without abnormal behavior. This result fully verifies the ability of the detection method provided in this application to identify potential defects in LEDs, possessing high predictive accuracy and reliability.

[0059] The failed LED beads in group A were cut by focused ion beam (FIB) and observed by scanning electron microscopy (SEM).

[0060] like Figure 2 As shown, unfilled V-shaped pit defects were observed in the epitaxial layer structure of the LED chip. More importantly, high-brightness metallic Ag filler was observed deep within and on the sidewalls of the V-shaped pits. Example 4

[0061] To verify the universality of this testing method on different types of LED devices, this embodiment selected a yellow LED chip of another specification (higher rated current) for batch verification.

[0062] The detection method in this embodiment is strictly performed according to the following steps: Step S1, Reference Parameter Acquisition and Initial Screening: 30561 large-size yellow LED chips (Y3535) from the same production batch were selected as the samples to be tested. Under forward test current I... F =1µA (test duration 5ms), reverse test voltage V R Under the condition of 5V (test duration 10ms), obtain the initial forward voltage V of each chip. FL1 and reverse leakage current I R1 Based on the requirements of this product specification, the initial screening pass standard is set as: V FL1 ≥1.5V and I R1 ≤0.05µA. The initial spectroscopic defects are shown in Table 6 below.

[0063] Table 6 Initial Spectroscopy Defects of Yellow Light Chip Yellow light chip defect standard quantity percentage <![CDATA[I R1 >0.05uA]]> 3 0.010% <![CDATA[V FL1 <1.5V]]> 33 0.108% After step S1, 36 obviously defective products were removed, and the products that passed the initial screening were retained.

[0064] Step S2, High-Current Stress Excitation: From the qualified products in the initial screening, 28,839 chips were randomly selected for stress excitation testing. Since the rated operating current and ultimate withstand current of this chip model are both higher than those of the Y3030 chip used in Example 1, the forward high-current impact parameters were adjusted to: current intensity 350mA, duration 40ms. These parameters were precisely set within the safe operating area (SOA) of this large-size chip, sufficient to excite potential defects without damaging the good product lattice.

[0065] Step S3, Parameter Acquisition After Stress Application: After the stress impact ends, under the same micro-current / low-voltage test conditions as in Step S1, the forward voltage V of these 28839 chips is acquired again. FL2 With reverse leakage current I R2 .

[0066] Step S4, Differential Calculation and Statistics: Calculate the change ΔV of a single chip. FL =|V FL2 -V FL1 The data before and after excitation were compared, and the results are shown in Table 7.

[0067] Table 7 High Current Impact Failure of Yellow Light Chip Yellow light chip defect standard quantity percentage <![CDATA[I R2 >0.05uA]]> 0 0.000% <![CDATA[V FL2 <1.5V]]> 3 0.010% <![CDATA[0.05V≤△V FL <0.07V]]> 13 0.045% <![CDATA[0.07V≤△V FL ]]> 2 0.007% Step S5, Threshold Determination and Depth Analysis: Based on the above data, three chips showed V after the impact. FL2 It directly fell below the acceptable lower limit of 1.5V, and there were also 2 chips with ΔV. FL Five chips with a voltage of ≥0.07V were identified as "potentially leaky defective products" by the system and were removed.

[0068] Reliability Verification and Guiding Significance: To further verify the overall quality of this batch of chips and the guiding role of this invention, the chips were subjected to high current impact and tested according to conventional standards (ΔV). FL Of the chips deemed good (<0.07V), 22,054 were randomly selected and subjected to the reflow soldering thermal stress test described in Example 2. After reflow soldering, the forward voltage V of the yellow light chip was tested again under the same conditions as in Example 1. FL3 The adverse situations are shown in Table 8 below.

[0069] Table 8. Reflow Soldering Defects of Yellow Optical Chips Yellow light chip defect standard quantity percentage <![CDATA[V FL3 <1.5V]]> 12 0.054% Results Analysis: Testing revealed that 12 chips still experienced forward voltage defects after reflow soldering. Combined with... Figure 3 (In this embodiment, the yellow light chip experiences a high current surge followed by a ΔV) FL In-depth analysis of the scatter plot shows that the large-size yellow light chip experienced a significant impact on the ΔV value after the impact. FL There are numerous scattered points with extremely dispersed distribution (13 points have a value of 0.05V≤ΔV). FL (Critical state sample <0.07V). This indicates that there are systematic quality fluctuations in this batch of chips during the epitaxial growth stage, the overall crystal lattice is extremely fragile, and the potential leakage current hazard is extremely large. Regarding... Figure 3 For high-risk batches with scattered and unconcentrated defects, producers can utilize the differential algorithm of this invention to flexibly tighten the interception threshold from the conventional 0.07V to 0.05V or even lower. By increasing the screening stringency, critical defective products like those in Table 8 that only reveal failure after reflow soldering can be completely intercepted, thereby achieving "zero-defect" delivery to end customers. Comparative Example 1

[0070] Some of the LED chips that were deemed "zero-time qualified products" in Example 1 were put into mass production and assembled into interactive screen modules using standard processes. However, some products exhibited a "caterpillar" display anomaly in actual applications. FIB-SEM microscopic analysis was performed on the failed LED chips (…). Figure 4 It was discovered that the epitaxial layer contained incompletely filled V-shaped pit defects. The deposited Ag metal within these pits formed deep-level defect centers, inducing tunneling current and causing chip leakage. This defect morphology is similar to that in Example 3. Figure 2 The features shown are consistent.

[0071] The above results strongly validate the detection mechanism of this invention: the high-current impact effectively "activates" the silver ion migration behavior at the V-shaped pit in the early stage, allowing potential defects to pass through ΔV. FL With △I R Significant changes in parameters become apparent, enabling precise screening of hidden defects before mass production.

[0072] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method of detecting potential leakage defects in an LED device, comprising: The method comprises the following steps: S1, obtaining initial forward voltage V F of the LED device at a forward test current I FL1 and initial reverse leakage current I R of the LED device at a reverse test voltage V R1 ; S2, applying a preset stress to the LED device to induce physical deterioration at the micro-defects; a parameter value of the stress is greater than a safe working boundary value of the LED device and less than a failure threshold value of the LED device; S3、in the forward test current I F and the reverse test voltage V R , obtain the forward voltage V FL2 and the reverse leakage current I R2 of the LED device after stress treatment; S4, calculating the variation of the forward voltage AV FL = |V FL2 FL1 | and / or the variation of the reverse leakage current AI R = |I R2 R2 |, if the AV FL and / or AI R exceeds a pre-set threshold, it is determined that the LED device has a potential risk of leakage.​​ 2. The method of claim 1, wherein the LED device is a white LED device. In step S4, the preset threshold value of the △V FL is 0.05V~0.07V, and the preset threshold value of the △I R is 0.01µA~0.05µA.

3. The method of claim 1, wherein the LED device is a white LED device. In step S2, the physical deterioration includes at least one of metal ion migration, conductive filament generation, insulating passivation layer micro-breakdown, or lattice defect expansion.

4. The method of claim 1, wherein the LED device is a white LED device. In step S2, the stress is one or more of electrical stress, thermal stress, mechanical stress, and environmental stress.

5. The method of claim 4, wherein the LED device is a flip chip LED device. The electrical stress is selected from at least one of forward large current impact stress, reverse high voltage impact stress, and pulse current impact stress; the thermal stress is selected from at least one of high temperature impact stress, low temperature freezing stress, and cold-hot cycle impact stress; the mechanical stress is ultrasonic vibration stress and / or mechanical pressure stress; and the environmental stress is high humidity environmental stress and / or high pressure environmental stress.

6. The method of claim 1, wherein the LED device is a white LED device. The stress is forward large current impact stress, and a current intensity of the forward large current impact is 3-5 times of a rated working current of the LED device.

7. The method of claim 1, wherein the LED device is a white LED device. The range of the forward test current IF is 0.1-50 μA, and the test time under the forward test current is 1-50 ms; the range of the reverse test voltage VR is 3-15 V, and the test time under the reverse test voltage is 1-50 ms.

8. The method of claim 1, wherein the LED device is a white LED device. The steps S1-S4 are integrated in a mass production line and sequentially executed as continuous online detection procedures.

9. An LED sorting apparatus, characterized by, A device for performing the detection method according to any one of claims 1-8, comprising: a test station for performing steps S1 and S3; a stress application station for performing step S2; a sorting execution mechanism for performing step S4 and diverting the LED device determined to have potential leakage risk.

10. The LED sorting device of claim 9, wherein: The stress application station is configured with at least one of a programmable power supply, a temperature control device, or a mechanical vibration device.

Images