Light emitting diode (LED) test apparatus and method of manufacture

Abstract

Embodiments relate to functional test methods useful for fabricating products containing Light Emitting Diode (LED) structures. In particular, LED arrays are functionally tested by injecting current via a displacement current coupling device using a field plate comprising of an electrode and insulator placed in close proximity to the LED array. A controlled voltage waveform is then applied to the field plate electrode to excite the LED devices in parallel for high-throughput. A camera records the individual light emission resulting from the electrical excitation to yield a function test of a plurality of LED devices. Changing the voltage conditions can excite the LEDs at differing current density levels to functionally measure external quantum efficiency and other important device functional parameters.

Claims

1. A method of manufacturing an optical device, the method comprising: providing a light-emitting device structure, the light-emitting device structure having a plurality of LED devices to be formed, disposed on a support substrate having a first contact layer accessible from a surface and a second contact layer underlying the light-emitting device structure; coupling a field plate device to the light-emitting device structure, the field plate device having a first face and a second face opposing the first face, the second face comprising a conductive layer, the second face being positioned in close proximity to at least a portion of the first contact layer of the light-emitting device structure such that a spatial gap is formed between a surface region of the second face and the first contact layer of the light-emitting device structure, and the spatial gap is filled with a liquid; generating a time-varying voltage waveform from a voltage source to form a voltage potential between the second face of the field plate device and the light-emitting device structure to inject current to each of the LED devices in the light-emitting device structure to cause the light-emitting device structure to emit electromagnetic radiation in a pattern; and capturing, using a detector device coupled to the light-emitting device structure, an image of the electromagnetic radiation in the pattern derived from the light-emitting device structure.

2. The method according to claim 1 wherein the field plate device is transmissive to the electromagnetic radiation and the electromagnetic radiation passes through the field plate.

3. The method according to claim 1 wherein the surface of a substrate under test is transmissive to the electromagnetic radiation and the electromagnetic radiation passes through the support substrate.

4. The method according to claim 1 wherein the detector device comprises imaging the electromagnetic radiation to produce an observable map of the pattern of electromagnetic radiation as a function of position over the light-emitting device structure of the support substrate.

5. The method according to claim 4 wherein detector device comprises a camera.

6. The method according to claim 5 wherein the field plate is transmissive and the camera is mounted to image the light-emitting device structure to collect the electromagnetic radiation through the field plate.

7. The method according to claim 5 wherein the support substrate is transmissive and the camera is mounted to image the light-emitting device structure to collect the electromagnetic radiation through the support substrate.

8. The method according to claim 1 wherein the time-varying voltage waveform is a voltage ramp from a first voltage potential to a second voltage potential to forward bias the light-emitting device structure at a selected current density during the measurement phase.

9. The method according to claim 5 wherein the camera integrates the electromagnetic radiation over the time-varying voltage waveform to produce a spatial map of total electromagnetic radiation produced over the light-emitting device structure.

10. The method according to claim 9 wherein the spatial map of integrated electromagnetic radiation is processed using image processing device to perform one or more of the following functions: signal averaging, thresholding and binning to develop a spatially-dependent functional test result of the light-emitting device structure.

11. The method according to claim 1 wherein the time-varying voltage waveform after the measurement phase is returned from the second voltage potential to the first voltage potential selected to use the light-emitting device structure reverse bias leakage current density and avoid exceeding potentially damaging reverse bias voltage.

12. The method according to claim 1 wherein the first contact layer of the light-emitting device structure is isolated using a material removal process to realize a plurality of individually addressable light-emitting devices.

13. The method of claim 1 wherein the first and second contact layers of the light-emitting device structure are isolated using a material removal process to realize a plurality of individually addressable light-emitting devices.

14. The method according to claim 1 wherein the spatial gap being no larger than a lateral distance of an LED device to be formed from the light-emitting device structure.

15. The method according to claim 1 wherein the spatial gap being no larger than ten times a lateral distance of an LED device to be formed from the light-emitting device structure.

16. The method according to claim 14 wherein the liquid within the spatial gap exhibits a sufficient dielectric property to couple a desired displacement current to excite the light-emitting device structure.

17. The method according to claim 15 wherein the liquid within the spatial gap exhibits a sufficient dielectric property to couple a desired displacement current to excite the light-emitting device structure.

18. The method according to claim 14 wherein the second face of the field plate device comprises a dielectric layer overlying the conductive layer, the dielectric layer positioned in close proximity to the at least a portion of the first contact layer of the light-emitting device structure such that the spatial gap is formed between the surface region of the dielectric layer and the first contact layer of the light-emitting device structure.

19. The method according to claim 15 wherein the second face of the field plate device comprises a dielectric layer overlying the conductive layer, the dielectric layer positioned in close proximity to the at least a portion of the first contact layer of the light-emitting device structure such that the spatial gap is formed between the surface region of the dielectric layer and the first contact layer of the light-emitting device structure.

20. The method according to claim 18 wherein the liquid within the spatial gap exhibits a dielectric property and a limited resistivity to couple a desired bias current to excite the light-emitting device structure.

21. The method according to claim 19 wherein the liquid within the spatial gap exhibits a dielectric property and a limited resistivity to couple a desired bias current to excite the light-emitting device structure.

22. The method according to claim 14 wherein the liquid within the spatial gap exhibiting a sufficient dielectric property and a limited resistivity to couple a desired displacement and DC bias current to excite the light-emitting device structure.

23. The method according to claim 15 wherein the liquid within the spatial gap exhibiting a sufficient dielectric property and a limited resistivity is selected to couple a desired displacement and DC bias current to excite the light-emitting device structure.

24. The method according to claim 20 wherein the liquid is deionized water with a resistivity between 0.01 and 1 megohm-cm.

25. The method according to claim 21 wherein the liquid is deionized water with a resistivity between 0.01 and 1 megohm-cm.

26. The method according to claim 22 wherein the liquid is deionized water with a resistivity between 0.01 and 1 megohm-cm.

27. The method according to claim 23 wherein the liquid is deionized water with a resistivity between 0.01 and 1 megohm-cm.

28. The method according to claim 20 wherein the liquid is deionized water with a resistivity between 1 megohm-cm and 18 megohm-cm.

29. The method according to claim 21 wherein the liquid is deionized water with a resistivity between 1 megohm-cm and 18 megohm-cm.

30. The method according to claim 22 wherein the liquid is deionized water with a resistivity between 1 megohm-cm and 18 megohm-cm.

31. The method according to claim 23 wherein the liquid is deionized water with a resistivity between 1 megohm-cm and 18 megohm-cm.

32. The method according to claim 15 wherein the field plate device dielectric layer is selected from silicon dioxide, silicon nitride or alumina.

33. The method according to claim 1 further comprising selecting at least one of the LED devices, and packaging the LED device.

34. The method according to claim 1 further comprising selecting at least one of the LED devices, releasing said LED device onto a member substrate.

35. The method according to claim 1, wherein each of the LED devices is a micro-LED characterized by a side dimension of 10 μm or less.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a simplified cross-section of an LED structure.

(2) FIG. 2 shows an LED support substrate containing LED device structures within an LED mass-production process.

(3) FIGS. 3A-B shows a top view (A) and a cross-sectional view (B) of a LED support substrate with singulated LED devices isolated by streets.

(4) FIG. 4 shows the LED support substrate with a non-singulated LED device structure with the top contact layer having sufficiently high sheet resistance to allow current injection functional test in the presence of adjacent shorts.

(5) FIG. 5A shows an embodiment of a field plate in close proximity to a portion of an LED device layer containing 4 LED devices on a support substrate.

(6) FIG. 5B shows the corresponding equivalent electrical circuit of the embodiment of FIG. 5A.

(7) FIG. 6A shows the main Capacitive Current Injection (C.sup.2I) measurement sequence: Current Injection/Measurement (I), Hold (II), Discharge/Reset (III) phases.

(8) FIG. 6B shows the corresponding LED current injected by the Capacitive Current Injection (C.sup.2I) measurement sequence of FIG. 6A.

(9) FIGS. 7A-B shows two embodiments of a field plate with camera looking through the field plate (A) and through the LED device support substrate (B).

(10) FIG. 8 shows the expected current density (A/cm2) versus dV/dT voltage ramp of a preferred embodiment.

(11) FIG. 9 shows a substrate scale method to attach the field plate onto a support substrate containing LED device structures using vacuum developed in the space between the field plate and the support substrate.

(12) FIG. 10 show a smaller field plate & camera optical system in a step/repeat mechanical configuration.

(13) FIG. 11 show a circuit model used to simulate the C.sup.2I functional test method according to an embodiment.

(14) FIGS. 12A-D the detailed sequence of the current injection/measurement phase I of an embodiment.

(15) FIGS. 13A-D the detailed sequence of the current injection/measurement phase III of an embodiment.

(16) FIGS. 14 A-D shows a longer time axis (200 msec) showing 4 measurement sequences of an embodiment.

(17) FIG. 15A shows an embodiment of a field plate in close proximity to a portion of a LED device layer containing 4 LED devices on a support substrate with a buried common contact and optional dielectric layer (i.e., “leaky” dielectric layer) and coupling gap medium to allow DC bias functional test.

(18) FIG. 15B shows the corresponding equivalent electrical circuit of the embodiment of FIG. 15A.

(19) FIG. 16 shows an embodiment of a field plate without a dielectric in close proximity to a portion of a LED device layer containing 4 LED devices on a support substrate with a buried common contact. The configuration using an external load resistor and coupling capacitor and DI water gap medium allows DC bias and AC pulsed functional test.

(20) FIG. 17A shows an embodiment of a field plate in close proximity to a portion of a LED device layer containing 4 LED devices on a support substrate with a buried common contact and dielectric layer.

(21) FIG. 17B shows the corresponding equivalent electrical circuit of the embodiment of FIG. 17A.

(22) FIG. 18 shows an embodiment of a field plate in close proximity to a portion of a LED device layer containing 4 LED devices on a support substrate that is used as a dielectric layer for capacitively coupling the second electrode.

(23) FIG. 19 shows a histogram plot of several LED devices falling within small ranges of Data values (called channels or bins) on the vertical scale as a function of Data in the horizontal scale.

DETAILED DESCRIPTION

(24) A further explanation of LEDs is found throughout the present specification and more particularly below. In an example, one type of LED is an organic light emitting diode (OLED) in which the emissive layer of the diode is formed of an organic compound. One advantage of OLEDs is the ability to print the organic emissive layer on flexible substrates. OLEDs have been integrated into thin, flexible displays and are often used to make the displays for portable electronic devices such as cell phones and digital cameras.

(25) Another type of LED is a semiconductor-based LED in which the emissive layer of the diode includes one or more semiconductor-based quantum well layers sandwiched between thicker semiconductor-based cladding layers. Some advantages of semiconductor-based LEDs compared to OLEDs can include increased efficiency and longer lifespan. High luminous efficacy, expressed in lumens per watt (lm/W), is one of the main advantages of semiconductor-based LED lighting, allowing lower energy or power usage compared to other light sources. Luminance (brightness) is the amount of light emitted per unit area of the light source in a given direction and is measured in candela per square meter (cd/m.sup.2) and is also commonly referred to as a Nit (nt). Luminance increases with increasing operating current, yet the luminous efficacy is dependent on the current density (A/cm.sup.2), increasing initially as current density increases, reaching a maximum and then decreasing due to a phenomenon known as “efficiency droop.” Many factors contribute to the luminous efficacy of an LED device, including the ability to internally generate photons, known as internal quantum efficiency (IQE). Internal quantum efficiency is a function of the quality and structure of the LED device. External quantum efficiency (EQE) is defined as the number of photons emitted divided by the number of electrons injected. EQE is a function of IQE and the light extraction efficiency of the LED device. At low operating current density (also called injection current density, or forward current density) the IQE and EQE of an LED device initially increases as operating current density is increased, then begins to tail off as the operating current density is increased in the phenomenon known as the efficiency droop. At low current density, the efficiency is low due to the strong effect of defects or other processes by which electrons and holes recombine without the generation of light, called non-radiative recombination. As those defects become saturated radiative recombination dominates and efficiency increases. An “efficiency droop” or gradual decrease in efficiency begins as the injection-current density surpasses a low value, typically between 1.0 and 10 A/cm.sup.2.

(26) Semiconductor-based LEDs are commonly found in a variety of applications, including low-power LEDs used as indicators and signage, medium-power LEDs such as for light panels and automotive tail lights, and high-power LEDs such as for solid-state lighting and liquid crystal display (LCD) backlighting. In one application, high-powered semiconductor-based LED lighting devices may commonly operate at 400-1,500 mA, and may exhibit a luminance of greater than 1,000,000 cd/m.sup.2. High-powered semiconductor-based LED lighting devices typically operate at current densities well to the right of peak efficiency on the efficiency curve characteristic of the LED device. Low-powered semiconductor-based LED indicator and signage applications often exhibit a luminance of approximately 100 cd/m.sup.2 at operating currents of approximately 20-100 mA. Low-powered semiconductor-based LED lighting devices typically operate at current densities at or to the right of the peak efficiency on the efficiency curve characteristic of the LED device. To provide increased light emission, LED die sizes have been increased, with a 1 mm.sup.2 die becoming a fairly common size. Larger LED die sizes can result in reduced current density, which in turn may allow for use of higher currents from hundreds of mA to more than an ampere, thereby lessening the effect of the efficiency droop associated with the LED die at these higher currents.

(27) LEDs have been used in portable devices such as watches, smartphones and laptops as well as computer monitors and TV displays for many years however only indirectly as an alternative white light source for Liquid-Crystal Display (LCD) display technologies. These were called “LED” TVs and the like, but the actual LEDs were predominantly GaN-based white LEDs to illuminate the backlight in lieu of the cold fluorescent lamp (CFL) backlight sources used before. The color pixel generation continued to be based on LCD technology that worked by a light subtraction process where colors are generated by blocking other colors using an intervening color filter. For example, a red pixel would be generated by a red color filter that blocked the green and blue portion of the backlight LED white spectrum. Grey scale (light intensity of the pixel) occurred by modulating light polarization through a liquid-crystal cell placed between two crossed polarizers along the light path.

(28) Although the LED backlight driven LCD display technology was more efficient and reliable than the CFL backlit version, the technology is still not power efficient. The reason is simple: although the LED white backlight devices can be fairly efficient in terms of external quantum efficiency (photons emitted per electrical carriers injected into the LED device), there are numerous inefficiencies in the rest of this LCD display technology. The first polarizer will cut a little half of the non-polarized white backlight, then each pixel is colorized by subtracting ⅔ of the remaining light (R without GB for red, G without RB for green and B without RG for blue). Other losses include pixel fill factor and film/LCD cell absorption and scattering. The total light output is therefore less than about ⅙ of the white LED backlight intensity.

(29) The trend is for more power efficient and bright display technologies, especially with portable, battery operated devices where battery life is a key factor. Micro-LED is a promising technology for achieving higher power efficiencies. In a micro-LED display, a small LED device placed within a pixel area is directly driven to generate light in a direct emissive manner. Color can be generated either by (i) utilizing UV-LEDs (i.e. GaN-based) with color phosphors to generate the pixel colors by photon down conversion and/or (ii) by using LEDs that generate the color directly (i.e. AlGaAs, GaAsP, AlGaInP, GaP for red, GaP, AlGaInP, AlGaP for green and ZnSe, InGaN, SiC for blue). In either case, the direct emission/direct view of the micro-LED display promises a sixfold improvement or more in power efficiency.

(30) Although the basic technology to realize micro-LED based displays is well known, numerous manufacturing and quality control challenges exist. One of these is functionally testing millions of micro-LED devices within the manufacturing process in a cost-effective and efficient manner before the pixels are committed to use. It is therefore desired to enable functional testing without direct electrical contact and in a manner compatible with micro-LED large-scale manufacturing processes. Further details of the present invention can be found throughout the present specification and more particularly below.

(31) Embodiments of the present invention describe LED device fabrication processes and manners of functionally testing LED devices without direct electrical contact. In particular, some embodiments of the present invention may relate to manners of functionally testing high-brightness LED, medium power LED, low-power LED and micro LED devices.

(32) In various embodiments, description is made with reference to figures. However, certain embodiments may be practiced without one or more of these specific details, or in combination with other known methods and configurations. In the following description, numerous specific details are set forth, such as specific configurations, dimensions, and processes, etc., in order to provide a thorough understanding of the present invention. In other instances, well-known semiconductor processes and manufacturing techniques have not been described in particular detail in order to not unnecessarily obscure the present invention. Reference throughout this specification to “one embodiment” means that a feature, structure, configuration, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of the phrase “in one embodiment” in various places throughout this specification are not necessarily referring to the same embodiment of the invention. Furthermore, the features, structures, configurations, or characteristics may be combined in any suitable manner in one or more embodiments.

(33) The terms “spanning”, “over”, “to”, “between” and “on” as used herein may refer to a relative position of one layer with respect to other layers. One layer “spanning,” “over” or “on” another layer or bonded “to” or in “contact” with another layer may be directly in contact with the other layer or may have one or more intervening layers. One layer “between” layers may be directly in contact with the layers or may have one or more intervening layers.

(34) Certain embodiments of the invention describe an LED device assembly in which an LED device structure layer is transferred from a support substrate and bonded to a pickup plate assembly before further processing. In accordance with embodiments of the present invention, the C.sup.2I functional testing step can be applied either before the transfer or after one or more transfers. For the purposes of simplifying the various possible configurations wherein the plurality of the LED structures is transferred and possibly bonded onto a different substrate, the target substrate shall be called a support substrate in each case. For example, the substrate that supported the LED structures during MOCVD growth is also called a support substrate, however after release and attachment to a pickup plate, such a plate and any other substrate or plate used to mechanically support the LED device layer will also be called a support substrate. If a pickup plate is used, common electrical contact can be accomplished using an electrically conducting material film between the transferred LED device structures and the rest of the pickup plate. As further described below, the common contact can also be accomplished using a second dielectric layer and optional voltage waveform source. In some cases, the pickup plate material would also have a degree of controllable tackiness to allow LED device pickup and transfer in production. The term support substrate will be generally used to connotate its role as mechanical support and will be the substrate described as part of (C.sup.2I) functional testing apparatus throughout this description.

(35) Depending on the specific embodiment of this invention and the point in the manufacturing process C.sup.2I functional testing is made, the support substrate can be transparent and have additional coatings. These either directly support the test process or exist as part of the requirements of the specific LED manufacturing process step as will be described in more detail below.

(36) Referring to FIG. 1, a representative LED 104 comprises of deposited layers that form a n-type cathode layer 100, an active layer (usually a Multi-Quantum Well or MQW series of sub-layers) 101 and a p-type layer 102 and p-contact 103. This LED structure is simplified and many additional layers such a buffer layers, blocking layers, n-contact layer(s) and the like are not shown for simplicity. Electrically, the LED would be contacted through layer 103 (or contact 106) as the anode and through layer 100 (or contact 105) as the cathode. Passing current through the LED device using a forward (positive voltage) bias from anode to cathode would generate light from radiative recombination processes from carriers flowing through the active region. The design of the active layer 101 is optimized for maximizing radiative recombination processes that emit light. Reverse biasing the LED structure will not generate light. Limiting reverse bias voltage is important with LEDs to avoid damaging or destroying the device through a process called breakdown. Within a safe reverse bias region, small leakage currents flow through the device.

(37) In LED manufacturing, the LED devices are made in mass-production using methods similar to substrate-based mass-production processes common in the semiconductor industry. Referring to FIG. 2, the LED structure described in FIG. 1 is deposited onto a suitable growth or support substrate 201 to make an LED substrate 200. Depending on the type, quality and color of the LED desired, different substrate material types can be used. Examples are GaP, GaAs, GaN substrates or heteroepitaxial growth substrates such as sapphire and silicon carbide (SiC) are also possible. Layer-transferred semiconductor layered template substrates are yet another type of growth substrate. The LED structure is then growth to develop a lower contact 202 (n-type or cathode in this example), active region 203 and upper contact 204 (p-type or anode in this example).

(38) The LED substrate of FIG. 2 contains multiple, non-singulated LED structures. Isolation of individual LED devices of the desired size and function can be made within the LED manufacturing sequence using process steps such as etching, lithography, passivation and deposition. Referring to FIGS. 3A and 3B, the desired LED devices can be isolated while residing on support substrate 301 using processes such as etching to form for example a trench 308. If these etch structures (sometimes called “streets”) are made over the substrate to form individually isolated structures such as square devices, a high number of LED devices 309 are electrically isolated and available for release and packaging. In this example, the trench 308 does not etch through the bottom common contact layer 302 and can thus be connected to a common potential 310. Each LED device 309 can thus be individually contacted using a voltage source 306 to the p-layer 304 and p-contact layer 305. Light 307 can then be measured from the contacted device to evaluate its functionality. In this example, a top emitting LED structure is shown where the top contact 305 could be a transparent electrode such as Indium Tin Oxide (ITO). Other structures are possible such as a bottom emitting structure. In this case, the support structure would be preferably transparent and the p-contact layer would be a light reflecting layer such as a metal layer. The LED would thus be tested by measuring the light escaping from the support substrate. Although the above was described as preferred embodiments to maximize the light capture, it would be possible to measure indirectly scattered or reflected light from LEDs even if the light measurement was done, for example, above the LED in a bottom emitting LED structure. Of course, there can be other variations, modifications, and alternatives.

(39) FIG. 4 shows a support substrate 401 where the LED devices are still not isolated. If the top contact layer 405 has a limited conductivity (such as an ITO layer with a relatively high film sheet resistivity), functional testing could still be accomplished despite a short 408 present nearby. Contacting a point on the surface using a voltage source 406 would develop a current through the top contact 405, p-layer 404, active layer 403, n-layer 403 to common contact 402. The relatively high resistance to the neighboring short 408 could allow light emission 407 to occur. Using a field plate in lieu of this direct contact example according to an embodiment of this invention would allow non-singulated LED layer testing. Dark (non-emissive) or weakly emissive areas would be an indicator of LED layer functional yield at an early stage of the LED manufacturing process. The efficacy and spatial resolution of this alternative embodiment would be a function of top layer sheet resistivity.

(40) There is therefore a need to inject a current to excite individual LED devices or LED areas such as described in FIGS. 3 & 4 in a manner that can support large-scale manufacturing.

(41) The invention has as its current injection device a field plate comprising of 3 elements: a mechanical support plate, an electrode and a dielectric layer. Referring to FIG. 5A, the field plate 501 comprises of a field plate support (top), electrode layer 502 connected to a voltage source 503 and adjacent to one face of dielectric layer 504. The mechanical support plate can also be electrically conductive and only require a dielectric layer. Of course, there can be other variations, modifications, and alternatives.

(42) The field plate electrode would be connected to voltage source 503 and the open face of the dielectric layer 504 would form a capacitance per unit area of:
C′.sub.FP=ε.sub.o×ε.sub.r/t.sub.d  (1)
Where C′dFP is the capacitance per unit area of the field plate (F/cm.sup.2) ε.sub.o is vacuum permittivity (8.854×10.sup.−14 F/cm) ε.sub.r is the relative permittivity of the dielectric layer (dimensionless) t.sub.d is the dielectric layer thickness (cm)
In an example, important material characteristics of the dielectric layer material includes dielectric constant, dielectric breakdown strength, resistivity, and optical transmissivity. For capacitive coupled configurations, readily deposited dielectrics such as silicon dioxide, silicon nitride and alumina (Al.sub.2O.sub.3) are of particular interest. If a DC test configuration is desired, a dielectric having limited current leakage would allow DC biasing if coupled to the device using an appropriate gap medium also having limited resistivity. In such a configuration, the field plate dielectric can be optional where the field plate voltage can now be directly coupled to the LED devices through the gap medium. Of course, there can be other variations, modifications, and alternatives.

(43) Again referring to FIG. 5A, the field plate 501 would be placed sufficiently near LED support structure 505 with an n-contact bottom electrode 506 connected to a common contact 507 and a plurality of p-contact top electrodes 508. Although the voltage across each LED device is shown in this description as being developed using a voltage source 503 and common contact 507, the voltage source can be alternatively connected to the bottom or two voltage sources can be connected, one to each of contacts 503 and 507. The effective LED device drive voltage would be the voltage difference between contacts 503 and 507 for all voltage source configurations. For this invention, the term “close proximity” shall mean the open face of the field plate dielectric layer 504 is placed in sufficient proximity to the open face of the LED structure contact surface 508 to allow a desired electrical coupling between the voltage source 503 to the top LED electrode surface 508. In FIG. 5A, this gap is shown as 509 and can be minimal with a limited little or no gap. Gap 509 should be small enough to allow sufficient capacitive coupling (for optimizing the current injection efficiency) and to minimize spatially defocusing the current injection effect. For the rest of this description, gap 509 will be assumed to be zero and thus C′.sub.EFF will be made equal to C′.sub.FP.

(44) The electrical analogue of the structure made by assembly 500 is shown in FIG. 5B. Voltage source 510 (503 in FIG. 5A) is connected to an effective capacitor C.sub.EFF 511 connected to LED device 512 having a top surface area A.sub.EFF. A voltage change will impress a current I.sub.LED onto LED device 512. For this example, isolation of LED devices with a common bottom contact is assumed. The effective capacitance C.sub.EFF is simply the series capacitance of the field plate dielectric layer with the capacitance of gap 509, both of area A.sub.EFF:
C′.sub.gap=ε.sub.o×ε.sub.r/t.sub.gap  (2)
Where C′.sub.gap is the capacitance per unit area of the gap (F/cm.sup.2) ε.sub.o is vacuum permittivity (8.854×10.sup.−14 F/cm) ε.sub.r is the relative permittivity of the gap medium (dimensionless) t.sub.gap is the gap thickness (cm)
and
C.sub.EFF=A.sub.EFF×(C′.sub.FP×C′.sub.gap)/(C′.sub.FP+C′.sub.gap)  (3)
C′.sub.EFF=(C′.sub.FP×C′.sub.gap)/(C′.sub.FP+C′.sub.gap)  (4)
Where C.sub.EFF is the effective LED device coupling capacitance (F) C′.sub.EFF is the effective LED device coupling capacitance per unit area (F/cm.sup.2) A.sub.EFF is the effective LED device area (cm.sup.2)
For the rest of this description, gap 509 will be assumed to be zero and thus C′.sub.EFF will be made equal to C′.sub.FP.

(45) The current I.sub.LED 513 and current density J.sub.LED are readily calculated as:
I.sub.LED=C.sub.EFF×dV/dt  (5)
J.sub.LED=C′.sub.EFF×dV/dt  (6)
Where dV/dt is the voltage rate of change between the voltage source 510 and the common electrode 507 in FIG. 5A (or the cathode contact in FIG. 5B). For this particular embodiment, LED 512 is connected anode to cathode however cathode to anode injection is possible by reversing all voltage polarities.

(46) FIGS. 6A and 6B shows the voltage and current waveforms that would form a measurement sequence according to a preferred embodiment of this invention. There are at least 2 phases in the measurement sequence, the current injection phase I (at time t.sub.0 to t.sub.1) and the discharge phase III (at time t.sub.2 to t.sub.3). A voltage holding phase II has been added to allow for enough time for the camera integration window to close before phase III commences however this can be very short and may not be necessary. The phases are described in more detail below and assumes null voltage at all point before t.sub.0.

(47) Referring to FIG. 6A, a voltage source-time plot 600 shows the voltage source waveform. At time t.sub.0, phase I begins with a positive ramp dV/dt|.sub.1 from 0 to V.sub.1 from time t.sub.0 to t.sub.1. This ramp will inject a current I.sub.LED into a LED of area A.sub.EFF according to equation (5) and a corresponding current density J.sub.LED according to equation 6. At time t.sub.1, the voltage is held at this voltage V.sub.1 until time t.sub.2. From time t.sub.2 to t.sub.3, the voltage is lowered back to a zero-voltage state using a negative ramp dV/dt|.sub.2. At time t.sub.3, another measurement sequence can then commence.

(48) FIG. 6B shows the corresponding current I.sub.LED waveform during the measurement sequence from the driving waveform of FIG. 6A. During phase I, a near constant I.sub.LED current will flow through the LED device according to equation (5). Light 602 will be emitted during Phase I. During Phase II, I.sub.LED and light emission will drop to 0. To measure the light output of a particular LED device among a plurality of devices simultaneously being excited by the field plate, an integrating camera that can capture the light from one or more LED devices is used. Image processing can generate a value proportional to specific LED devices located within the camera's field of view. This value would be in turn proportional to the optical energy integrated over phase I. It is therefore desirable to start an image sensor integrating time period from a point slightly before t.sub.0 to an end point slightly after t.sub.1. This will ensure the camera integrating sensor will capture the complete light pulse emanating from the LED structures during phase I.

(49) FIGS. 7A and 7B shows two embodiments of the invention. The figures show top and bottom camera placement that can intercept at least a portion of the plurality of LED devices being excited by the measurement sequence through the field plate. Referring to FIG. 7A, a test configuration 700 is shown with a transparent field plate assembly 702 including a field plate electrode 703 and dielectric layer 704. Electrode 703 is connected to a voltage source 705. This assembly is placed in close proximity to the LED device support substrate 701 supporting a plurality of LED device 707 connected to a common contact 706. A camera 708 is placed above the field plate assembly 702 to perform the functional test. FIG. 7B shows an alternative test configuration 709 where the camera is placed below the support substrate. In this configuration, the support substrate and intermediate layers to the LED device structures must be transparent to allow the light to reach the camera.

(50) Optical power generated by an LED is related to electrical power flowing through the LED by the external quantum efficiency η.sub.EXT or P.sub.opt=η.sub.EXT×P.sub.elec. The parameter η.sub.EXT is in turn quite sensitive to the current density and other device characteristics such as light extraction efficiency. The optical power of an LED device is thus related to electrical power as:
P.sub.opt=η.sub.EXT×P.sub.elec=η.sub.EXT×I.sub.LED×V.sub.F=η.sub.EXT×C.sub.EFF×dV/dt×V.sub.F  (7)
Where P.sub.opt=LED optical power (W) η.sub.EXT=LED external quantum efficiency V.sub.F=LED forward voltage drop (V)
Over a time period Δt=t.sub.1−t.sub.0 (phase I):
E.sub.opt=η.sub.EXT×C.sub.EFF×dV/dt×V.sub.F×Δt  (8)
Where E.sub.opt=LED optical energy emitted during phase 1 (J)

(51) According to equation 8, the integrating camera will measure a value proportional to each measured LED's external quantum efficiency. Changing the voltage ramp value will have the effect of selecting a different current density according to equation 6. By plotting Eopt as a function of ramp value to V.sub.1, a plot of light energy (related to η.sub.EXT) as a function of J.sub.LED can be generated. This capability can be particularly useful to measure low current density performance of micro-LED devices. Micro-LED devices are typically driven at very low values of 0.001-1 A/cm.sup.2 and are more sensitive to drop in external quantum efficiency at these low levels due to non-radiative recombination processes.

(52) During phase III, a negative dV/dt ramp allows the voltages to be returned to zero to reset the system for another measurement. During this phase, the LEDs will be reverse biased and will discharge C.sub.EFF using reverse bias leakage current. So as not to reverse bias the LEDs to a voltage level that can cause damage, the negative voltage ramp must be sufficiently slow to keep all devices within a safe reverse bias voltage range. Such a range can be selected depending on the type and design of the LEDs to be tested. As merely an example, reverse bias leakage current density for GaInN LEDs can be estimated using a paper entitled “Transport mechanism analysis of the reverse leakage current in GaInN light-emitting diodes”, Q. Shan & al., Applied Physics Letter 99, 253506 (2011). FIG. 2 shows a −5V reverse bias leakage current of approximately 1.5×10.sup.−7 A on a 1 mm.sup.2 LED device at room-temperature. This corresponds to 15 μA/cm.sup.2. This reverse bias leakage current density will be used to calculate values and parameters for specific C.sup.2I examples described below.

(53) Suitable integrating cameras must satisfy the following criteria: a. Pixel sensitivity and dynamic range (allows LEDs to be accurately measured through the operating range of interest without excessive dark noise and signal saturation). b. High pixel density and frame rate (increases throughput and parallel LED measurement). c. Global shutter and flexible triggering (all pixels must be triggered and integrate in over the same time period).
One example of an industrial camera meeting these criteria is model GS3-U3-23 S6M-C from PointGrey Research Inc., Richmond, BC, Canada. The camera is a 2.3 megapixel (1920×1200) monochromatic camera with global shutter, 5 μsec to 31.9 sec exposure range, over 160 frames per second rate, 1/1.2″ sensor format, 12-bit digitization, 5.86 μm pixel size, 72 dB dynamic range, 76% quantum efficiency (525 nm), an e-saturation capacity of about 32,000 electrons and temporal dark noise of about 7 e−. Used singly or in a matrix arrangement where nxm cameras would be used to measure a larger field plate area simultaneously, the camera will have the capability to measure numerous LED devices with the requisite accuracy.

(54) For the following examples, a field plate with a 3 μm silicon dioxide dielectric layer is assumed (ε.sub.r=3.9). This dielectric material is commonly used and can be sputtered, grown or deposited on numerous materials. The thickness was selected to be sufficiently thin to allow testing of micro-LED devices down to 10 μm×10 μm or less and can support a voltage exceeding about 1500 volts before breakdown. C′.sub.FP would be 1.15 nF/cm.sup.2.

(55) A value for V.sub.1 of 500V is assumed (see FIG. 6A). With these assumptions and parameter selections, the light pulse energy per LED can be simplified as:
E.sub.opt=η.sub.EXT×C.sub.EFF×ΔV×V.sub.F  (9)

(56) For the parameters selected, FIG. 8 shows the current density selected for a voltage ramp time period. For example, the LED would be driven at 0.01 A/cm.sup.2 if the field plate voltage was driven from zero to +500 Volts in approximately 60 μsec (phase I). The camera shutter would be opened slightly before the ramp begins (for example, 10-50 μsec before t.sub.0) and would be opened slightly after the end of phase I (for example, 10-50 μsec after t.sub.1). Apart from ensuring that the phase I LED light pulse is fully integrated within the camera shutter time window, excessive integration time should be avoided as this will tend to raise the noise floor of the camera. Phase II can be selected to end just as the integrating shutter closes.

(57) To safely recover during phase III, equation 6 is utilized with the current density selected to be approximately equal to the leakage current density. For example, utilizing a target leakage current density of 10 μA/cm.sup.2 (a little lower than the expected leakage of 15 μA/cm.sup.2) and ΔV=500V, equation 6 predicts a minimum Δt of almost 60 msec. This corresponds to a measurement repetition rate of about 16 frames per seconds for injection current densities of 0.0005 A/cm.sup.2 or above.

(58) To estimate the signal achievable with this measurement approach and the area covered by one camera, the following additional parameters are assumed: a. GaN LED (about 410 nm emission & 65% camera quantum efficiency) b. V.sub.F about 3V c. E.sub.opt=170 nJ/cm2 (η.sub.EXT˜10%)

(59) At about 3 eV per photon, approximately 3.5×10.sup.11 photons/cm.sup.2 are emitted during phase I. The number of corresponding photo-electrons that can be generated within the camera would be 0.65×3.5×10.sup.11 photons/cm.sup.2 or 2.3×10.sup.11 photo-electrons/cm.sup.2 (assuming sensor to field plate 1:1 magnification). At this magnification, a 5.86 μm pixel size would still capture over 78,000 electrons, more than twice the pixel saturation capacity. A lower V.sub.1 voltage could be selected if a lower integrated photo-electron count per camera pixel is desired.

(60) The imaging of the of the field plate to the camera sensor area is thus less a function of the available signal than the number of pixels allocated per LED device. For a larger LED device measuring 250 μm on a side, less magnification is necessary. Assuming a 2×2 pixel area to cover each LED device for accurate metrology, one camera could measure 960×600 LED devices, 240 mm×150 mm field plate area or more than the area of a 6″ support substrate. In this example, V.sub.1 could be reduced to less than 100V and perhaps lower while still maintaining excellent signal to noise ratio. If the light pulse energy is too high, a neutral density filter or other absorption filter can be placed between the emitting surface and the camera to avoid camera saturation.

(61) For a micro-LED application with 10 μm×10 μm LED device size, the same 960×600 LED devices per sensor would be measured or about 9.6 mm×6 mm field plate area. A step and repeat system with approximately 16×25 steps would allow testing of a 6″ micro-LED substrate containing over 170 million devices. If a single measurement per LED device is sufficient, a synchronized image capture with a moving camera or cameras could decrease test time to less than 1 minute or even a few seconds. For example, a 16 frame per second capture rate would allow a full 6″ substrate to be functionally tested in about 25 seconds. That would correspond to over 9 million LED devices tested per second, far faster than probe cards and individual test methods.

(62) In a preferred embodiment, FIG. 9 shows a substrate-sized field plate can be attached to a support substrate using vacuum to make an assembly 900 suitable for functional test. The field plate 901 is placed on the LED device support substrate 902 with a compliant vacuum seal 903 placed on the outside peripheral area to maintain a level of vacuum between the field plate and the LED device support substrate. The air is then evacuated in the space between the plates using vacuum port 904. The plates would be pressed together at up to atmospheric pressure to minimize the gap in a uniform manner, thus optimizing the effective field plate coupling capacitance C.sub.EFF. A support substrate exchange mechanism could exchange substrates to be tested under the field plate by cycling port 904 between vacuum and vent conditions. A camera 905 that measures above the field plate is shown in this embodiment. Of course, there can be other variations, modifications, and alternatives.

(63) In another embodiment, FIG. 10 shows an assembly comprising of a smaller field plate 1000 and camera 1001 placed above an LED device support substrate 1002. The field plate/camera assembly is moved in successive move/measure steps 1003 to measure the complete substrate 1002. Of course, there can be other variations, modifications, and alternatives.

(64) Electrical simulations of the measurements sequence showing the main phase 1 and 3 waveforms are shown in FIGS. 11-14. The system being simulated is as follows: 1. Field plate: 3 μm silicon dioxide, C′.sub.EFF=1.15 nF/cm2 2. 10 μm×10 μm LED device size: 1.15 fF C.sub.EFF, 15 pA reverse leakage current 3. 0.01 A/cm.sup.2 current density test point 4. V.sub.1=500V (60 μsec ramp time to achieve 0.01 A/cm.sup.2 current density injection) 5. 60 msec measurement repetition rate 6. LED device is a standard diode capable of reverse leakage current of about 10 pA
The program used is a SPICE circuit simulator called Micro-Cap version 11 from Spectrum Software (Sunnyvale, Calif.). One 10 μm×10 μm LED device was simulated under the above conditions. FIG. 11 shows the circuit diagram where C.sub.EFF=1.15 fF driven by a voltage generator V2. This generator was programmed to ramp from 0 to +500V in 60 μsec followed by a 60 msec ramp down from +500V to 0V. Voltage source V3 is unconnected but was programmed to show an example of a desired camera shutter window. In this example, the shutter is opened 10 μsec before phase I and closes 10 μsec after phase I.

(65) FIG. 12 A-D shows the Phase 1 waveforms with the voltage source V2 (FIG. 12A), LED device forward bias (FIG. 12B), LED device forward current (FIG. 12C) and the camera shutter control signal from voltage source V3 (FIG. 12D). Referring to FIG. 12D, the camera integrator shutter opens 10 μsec before the start of the voltage (at time+10 μsec on the time axis). At time +20 μsec on the time axis, the voltage source starts to ramp towards+500V (time to). During this phase I until time +80 μsec, the LED device is biased at +10 nA (FIG. 12C) at a forward bias of approximately +250 mV (FIG. 12B). This corresponds to 0.01 A/cm.sup.2 current density as desired. After time+80 μsec, the voltage ramp stops and the LED current drops to zero. At time+90 μsec, the camera shutter closes, completing its integration of the light pulse that was generated during phase I. The voltage source will now start a slow discharge at a target leakage current of −10 pA. FIG. 13 A-D shows the same voltage and current points during the phase III discharge lasting about 60 msec. FIG. 13C shows a −10 pA discharge current that allows C.sub.EFF to discharge over 60 msec from +500V to 0V safely. After the voltage source is returned to zero at about +60 msec, a new measurement sequence is initiated. FIG. 14 A-D shows a longer time axis (200 msec) showing 4 measurement sequences.

(66) A direct common contact also allows DC biasing and functional testing configurations. FIG. 15 shows an embodiment where the LED devices can be biased in a DC mode only or in combination with a time varying AC mode. Referring to FIG. 15A, the field plate 1501 comprises of a field plate support (top), electrode layer 1502 connected to a voltage source 1503 and adjacent to one face of an optional “leaky” dielectric layer 1504. The mechanical support plate can also be electrically conductive and only require the optional “leaky” dielectric layer 1504. Of course, there can be other variations, modifications, and alternatives.

(67) The field plate electrode is connected to voltage source 1503 and the open face of the optional “leaky” dielectric layer 1504 form a capacitance per unit area of:
C′.sub.FP=ε.sub.o×ε.sub.r/t.sub.d  (10)
Where C′.sub.FP is the capacitance per unit area of the field plate (F/cm.sup.2) ε.sub.o is vacuum permittivity (8.854×10.sup.−14 F/cm) ε.sub.r is the relative permittivity of the dielectric layer (dimensionless) t.sub.d is the dielectric layer thickness (cm)

(68) The dielectric would have a resistivity of ρ.sub.d, selected to allow the desired biasing of the LED devices in a DC bias configuration. The time constant driving the bias response time is ε.sub.o×ε.sub.r×ρ.sub.d. The effective resistance can be calculated as follows:
R′.sub.FP=ρ.sub.d×t.sub.d(ohms-cm.sup.2)  (11)
Where R′.sub.FP is the resistance for a unit area of the field plate (ohms-cm.sup.2) ρ.sub.d is the resistivity of the field plate dielectric layer (ohm-cm) t.sub.d is the dielectric layer thickness (cm)
In an example, the leaky dielectric layer can be generally described as a layer with a reasonably high relative dielectric constant, resistivity on the order of 1 Mohm-cm or higher and a sufficiently high dielectric breakdown field strength. Type II DI (deionized) water meets these criteria with a dielectric constant of 81, resistivity of 1 Mohm-cm and a breakdown field strength exceeding 13 MV/cm. In other examples, the layer can be a slightly conductive doped glass/ceramic, plastic or the like. If a small relative dielectric constant of about 1 is acceptable, an air layer with a voltage within a gap could become slightly conductive by weak ionization to accomplish the function of a “leaky” dielectric layer.

(69) Again referring to FIG. 15A, the field plate 1501 would be placed sufficiently near LED support structure 1505 with an n-contact bottom electrode 1506 connected to a common voltage source 1507 and a plurality of p-contact top electrodes 1508. Although the voltage across each LED device is shown in this description as being developed using a voltage source 1503 and common voltage source 1507, a voltage source can be alternatively connected to either contact 1502 or 1506. The effective LED device drive voltage would be the voltage difference between contacts 1503 and 1507 for all voltage source configurations. For this configuration, the open face of the field plate dielectric layer 1504 or the electrode contact 1502 is placed in sufficient proximity to the open face of the LED structure contact surface 1508 to allow a desired electrical coupling between the voltage source 1503 to the top LED electrode surface 1508. In FIG. 15A, this gap is shown as 1509 and can be minimal with a limited gap. Gap 1509 should be small enough to allow sufficient capacitive and resistive coupling (for optimizing the current injection efficiency) and to minimize spatially defocusing the current injection effect for the selected gap medium.

(70) The electrical analogue of the structure made by assembly 1500 is shown in FIG. 15B. Voltage source 1510 (1503 in FIG. 15A) is connected to each LED device 1512 having a top surface area A.sub.EFF through two effective capacitors, one representing the optional “leaky” field plate dielectric 1504 (C.sub.FP 1511) and one representing the gap dielectric medium 1509 (C.sub.GAP 1513). Each capacitor is shunted by a resistor representing the leakage path through the field plate dielectric R.sub.FP 1514 and gap medium R.sub.GAP 1515. A voltage source 1516 (1507 in FIG. 15A) connected to the bottom common contact 1506 completes the electrical circuit. A voltage change and level will impress a current I.sub.LED 1517 onto LED device 1512. The effective capacitance C.sub.EFF is simply the series capacitance of the field plate dielectric layer with the capacitance of gap 1509, both of area A.sub.EFF:
C′.sub.gap=ε.sub.o×ε.sub.r/t.sub.gap  (12)
Where C′.sub.gap is the capacitance per unit area of the gap (F/cm.sup.2) ε.sub.o is vacuum permittivity (8.854×10.sup.−14 F/cm) ε.sub.r is the relative permittivity of the gap medium (dimensionless) t.sub.gap is the gap thickness (cm)
and
C.sub.EFF=A.sub.EFF×(C′.sub.FP×C′.sub.gap)/(C′.sub.FP+C′.sub.gap)  (13)
C′.sub.EFF=(C′.sub.FP×C′.sub.gap)/(C′.sub.FP+C′.sub.gap)  (14)
Where C.sub.EFF is the effective LED device coupling capacitance (F) C′.sub.EFF is the effective LED device coupling capacitance per unit area (F/cm.sup.2) A.sub.EFF is the effective LED device area (cm.sup.2)

(71) The gap medium shunt resistor is calculated as:
R′.sub.GAP=ρ.sub.gap×t.sub.gap(ohms-cm.sup.2)  (15)
Where R′.sub.GAP is the resistance for a unit area of the gap medium (ohms-cm.sup.2) ρ.sub.gap is the resistivity of the gap layer (ohm-cm) t.sub.gap is the gap layer thickness (cm)
and the effective shunt resistor is calculated as:
R.sub.EFF=(R′.sub.FP+R′.sub.GAP)/A.sub.EFF  (16)
Where R.sub.EFF is the effective coupling shunt resistance (Ohms)

(72) An example of the DC injection functional test mode is explained here using the structure of FIG. 16. The test configuration is as follows: 1. Device size: 25 μm×25 μm 2. Field plate has no dielectric layer 3. Gap is 25 μm with type II DI water (>1 Mohm-cm) 4. Target DC bias of 10 mA/cm.sup.2 set by bottom electrode voltage source and bias load resistor R.sub.L 5. AC (pulsed) operation driven by the top field plate voltage source through an external coupling capacitor C.sub.C 6. Test area 5 cm.sup.2
The field plate 1601 is placed sufficiently near LED support structure 1602 with an n-contact bottom electrode 1603 connected to a common ground contact 1604 and a plurality of p-contact top electrodes 1605. For each LED device, the gap medium coupling capacitance is 18 fF and its shunt resistance is 400 Mohm. The relatively large coupling capacitance is made possible using DI water in gap 1606 which has a high relative dielectric constant of 81. The AC and DC components of the injected current within each LED device is denoted by current 1617.

(73) The DC bias is set by adjusting voltage source 1607 to a preset positive potential V.sub.pos to bias the LED devices through the load resistor R.sub.L 1608, field plate contact 1609, gap medium 1606, through the LED devices to bottom contact 1604. To achieve the DC bias point of 10 mA/cm.sup.2, a total current of 50 mA must flow through R.sub.L. If R.sub.L is chosen as 100 kohm to allow an efficient coupling of an AC pulse through coupling capacitor C.sub.C 1610 roughly 5 kV positive bias V.sub.pos is needed. The voltage drop through the DI water is about 25V while the LED requires 2.5-3.5V to turn on. Note that by varying the bias level, a graph of the output light level versus DC current density bias can be measured. Pulsing the DC bias synchronously with the camera capture time would allow signal averaging and multiple device bias set points to be measured.

(74) A pulsed signal response can be combined by pulsing source 1611 through coupling capacitor 1610. Assuming a C.sub.L of 1 nF, a fast pulse can couple charge into the LED devices before R.sub.L×C.sub.C relaxation can lower the charge coupling efficiency. For 100 kohm and 1 nF device values, the system relaxation time constant is 100 μsec, a value sufficiently long to ensure that the charge can be transferred to the LED devices and converted to a measurable pulse of light. In this example, the total DI water coupling capacitance is about 14 nF. If a 100 mA/cm.sup.2 bias is injected for 5 μsec, a total charge of about 3 pC per LED device or 2.4 μC for the test area must be delivered. Since the DI coupling capacitor is about 14 nF, the effective coupling capacitance would still be about 1 nF. The voltage pulse magnitude required of source 1611 would be 2400 volts ramped in 5 μsec. This way, both DC biasing and AC pulsed functional test can be made of an LED device substrate with a bottom common contact. Of course, there can be other bias and pulsing configurations, variations, modifications, and alternatives.

(75) Although this invention has been described with a common contact existing under the LED devices, other configurations for injecting current are possible. FIG. 17A shows another embodiment 1700 where an analogue to the field plate 1701 is present within support substrate 1702 below the plurality of LED device structures such as LED device 1703. Under the lowest LED device structure layer (the n-layer in examples described in this invention), a dielectric layer 1704 and electrode 1705 completes the support substrate capacitive coupling device. Electrode 1705 is connected to a voltage source 1706. The field plate is connected to a separate voltage source 1707 and field plate electrode 1708. In this example, a camera 1709 is placed above the field plate to capture the light emission response of the plurality of LED devices under test. In this example, the isolation between devices is shown to be complete, however this method would still function with or without full isolation of the n-layer. FIG. 17B shows the equivalent circuit 1711 of this capacitively coupled support substrate configuration. The only change is the insertion of a second coupling capacitor C.sub.EFF2 below each LED device cathode. The resulting circuit can be made to operate equivalently and be effective in performing C.sup.2I functional test. For example, assuming a support substrate dielectric layer 1704 identical to dielectric layer 1710 within the field plate, voltage source 1706 driven identically but negatively to voltage source 1707 (0 to −500V for source 1706 and 0 to +500V for source 1707), the measurement system 1700 would perform essentially identically to a common contact support substrate configuration.

(76) In yet another embodiment, C.sup.2I functional testing could also be applied to a modification of the test configuration of FIG. 17A that eliminates the need for a buried electrode within the support substrate. In this embodiment, the dielectric property of the support substrate itself is used to inject current through the LED devices. For example, a quartz, sapphire or plastic support substrate could serve as dielectric 1704 in FIG. 17A. FIG. 18 shows a specific embodiment 1800 of this configuration. A support substrate 1801 having adequate dielectric properties and thickness containing a plurality of LED devices on its surface such as LED device 1802 is placed on top of an electrode 1803 connected to a voltage source 1806. Field plate 1804 having a dielectric layer 1805 and electrode 1806 connected to a second voltage source 1807 completes the C.sup.2I functional test circuit. A camera 1808 placed above field plate 1804 is shown in this embodiment. The equivalent electrical circuit would be similar to FIG. 17B except that the value C.sub.EFF2 is likely substantially smaller due to the thickness of the support substrate. For example, a support substrate made of sapphire (ε.sub.r˜10) of 500 μm in thickness, C′.sub.EFF2 will be approximately 18 pF/cm.sup.2, about 65 times smaller than C.sub.EFF1. A faster voltage ramp and/or a larger voltage value for V.sub.1 could compensate for this loss of coupling efficiency. For example, field plate voltage source 1807 could be driven from 0 to +500V while the support substrate voltage source 1804 could be driven at 0 to −32.5 kV (−500V×65=−32.5 kV). The electric field strength within the sapphire support substrate would be 0.65 MV/cm, well below its breakdown strength of approximately 1 MV/cm. Driven in this fashion, the LED devices would be driven substantially equivalently and allow C.sup.2I functional test without a buried contact within the LED device support substrate. High-voltage waveform generators to drive electrode 1803 can be realized using IGBT, MOSFET, or thyristor devices. High-voltage switches capable of switching up to 36 kV are model number HTS-361-01-C (36 kV, 12A max current) and model number HTS-361-200-FI (36 kV, 2000 A max current) from Belke Electronic GMBH (Kronberg, Germany). Programmable waveform shaping circuits could slow the fast voltage change to a voltage ramp meeting the desired C.sup.2I functional test properties. For a 6″ substrate, the total capacitance would be about 3.2 nF and at 16 measurements per seconds, the ½ CV.sup.2f power would be about 27 Watts and the average current would be 830 μA, safely within normal operating specifications for commercially available high-voltage switches. For the HTS-361-200-FI 2000 A capable switch, current density C.sup.2I measurements as high as 11 A/cm.sup.2 could be performed. Of course, there can be other variations, modifications, and alternatives.

(77) Certain image processing methods can be utilized to improve the accuracy of the measured data corresponding to each LED device under test. Each imaged LED device onto the sensor would be imaged onto a specific area within the camera sensor array. One image processing method uses spatial information from the target image to generate a physical centroid (x,y) location for each LED device within the measured camera output data image. This correspondence of LED device centroid location on the support substrate to its corresponding centroid location on the camera sensor can be developed and possibly corrected using camera magnification, optical distortion correction, image capture to sense and locate the LED device matrix and the like. The resulting centroid matrix would therefore be the set of (x,y) location within the sensor image for each LED device. For example, referring to the previous example, a 960×600 LED device set imaged onto a 1920×1200 digital sensor matrix would have a centroid matrix as follows:
Centroid for LED(i,j)=Camera data location(x,y)
Where i, j are integers (i=1 to 960, j=1 to 600) for each measured LED while the camera location (x,y) is a floating point number within the sensor pixel area (0<x<1920, 0<y<1200). Once this centroid matrix is developed, image processing methods using weighted functions can take the digitized image and develop a set of data values that are extracted using a weighing function where more weight is given to sensor data imaged closest to the physical LED centroid location. Image processing systems can accomplish this convolution function in parallel and usually at frame rate speeds. The LED data values thus comprises of an output LED device (i,j) matrix of data values calculated using centroid weighted functions applied to the digitized camera data in a preferred embodiment.

(78) Additional image processing methods applied to the digitized camera output (proportional to the total integrated light emitted by the LED devices imaged onto the camera sensor(s) during phase I) can be utilized to develop a result indicative of LED device functionality. This functional data will be in the form of a matrix that contains one or more values derived from the measurement. For each LED at location (i,j), there will be a set of n data points Data.sub.n(i,j)=Value.sub.n (where n is an integer greater or equal to 1). Multiple independent Data.sub.n(i,j) values for each LED under test for example, could be values of light output at differing current density values measured using n measurement sequences taken with different phase I voltage ramp values. Each Data.sub.n(i,j) measurement data value can in turn be the average of multiple measurements to improve signal-to-noise ratio. Signal averaging is a well-known method where the standard deviation of a signal exhibiting stochastic noise would be reduced by sqrt(m) where m is the number of measurements points that are averaged. For example, if a data point exhibiting a stochastic noise standard deviation of z, averaged data points using the average of 100 data points would have a standard deviation of z/sqrt(100) or 10 times lower.

(79) Once the LED device (i,j) data values are collected, a threshold or set of test criteria can be applied to develop a determination of functionality, perhaps adding a Data.sub.n(i,j) value of 0 or 1 (0=bad device, 1=good device) for each LED being measured. For example, non-emitting or weakly emitting devices could be labeled as bad devices if a desired minimum threshold is applied to the data. Of course, multiple thresholds and other criteria applied to the set of data values or the pass/fail criteria could also be useful in functional test, repair strategies and process yield analysis (cause and correction). As merely an example, multiple thresholds could be applied to the LED device Data.sub.n(i,j) data to generate a bin number label for each LED device to match LEDs in functionality and drive a strategy of releasing devices with similar characteristics according to a criteria or set of criteria. Random-access laser lift-off or other individual LED device release methods could aggregate LED devices having similar bin numbers based on the bin label matrix value for each (i,j) LED device. This could be useful to limit display non-uniformity caused by using LED devices having excessively different functional characteristics. Multiple thresholds could also be utilized to develop statistics useful for yield and process control. For example, the standard deviation and other statistical analyses applied to bin data can be an indicator of yield and process stability. Sudden changes in these derived quantities can signal a process excursion. FIG. 19 shows a histogram plot 1900 of several LED devices falling within small ranges of Data.sub.n values (called channels or bins) on the vertical scale as a function of Data.sub.n in the horizontal scale. Most of the LED devices fall within a functionally acceptable range 1901 while LED devices below threshold 1902 or above threshold 1903 are considered rejects. The width 1904 of the LED device binning function can be useful for yield and process control. LED devices falling within similar bins 1905 could be later aggregated and used for their similar functional test results to improve display uniformity.

(80) If the functional test apparatus according to this invention images less than the desired area and requires a step and repeat function, the centroid matrix may need to be recalculated for each new LED device area to be measured. If the step system is sufficiently accurate to align the next set of LED devices to be measured however, the centroid matrix may be reused. Of course, there can be other variations, modifications, and alternatives.

(81) Generally, a field plate allows functional testing of a substrate containing LED devices to occur either by one or more cameras fixed or moved relative to the field plate. Test equipment cost, complexity, target LED device size and test throughput capability are some of the criteria that must be evaluated before a specific configuration is selected. Other design limitations and criteria must also be addressed to assure measurement functionality to desired specifications. One such design criteria is assuring the phase I voltage waveform across each LED device being tested is not significantly distorted due to contact resistance and parasitic capacitance. For example, a fast voltage ramp for Phase I desired to measure higher current density operation could cause a significant waveform distortion and voltage drop caused by RC low-pass filtering for LED devices situated in the middle of the field plate. This could occur if the field plate electrode or common contact resistance is too high. Mitigation of these effects could happen by lowering the effective contact sheet resistivity or attaching a lower resistivity layer before testing. Finally, a large field plate will require some power to charge & discharge the field plate capacitance CFP at the measurement repetition rate and may generate resistive heating within the contact layers. For example, a 6″ substrate field plate using a 3 μm silicon dioxide dielectric layer would have a total capacitance C.sub.FP of about 200 nF. If a 16 Hz capture rate and 500V ramp is assumed, the ½ CV.sup.2f power would be about 0.5 W. At this proposed operating point, small and manageable test power levels are generated, even with a full 6″ field plate configuration.

(82) While the above is a full description of the specific embodiments, various modifications, alternative constructions and equivalents may be used. Although the above has been described using a selected sequence of steps, any combination of any elements of steps described as well as others may be used. Additionally, certain steps may be combined and/or eliminated depending upon the embodiment. Furthermore, although the description and examples has been directed towards GaN LED devices on a planar surface, any planar or curved surface containing photon emitting devices could be functionally tested using the C.sup.2I method. For example, Vertical-Cavity Surface-Emitting Lasers (VCSELs), Organic LEDs (OLEDs), silicon photonics devices and other surface emitting devices could be tested using this invention. Additionally, in another example, II-VI semiconductor materials and associated devices can also be used. In an example, the LED or other device can have a variety of applications such as general or specialized lighting, a display, whether a large panel, mobile device, or projection, chemical treatment, vehicle lighting, medical, and others. In an example, the method can also include selecting one of the good devices, releasing said LED device onto a member substrate and packaging the device. A member substrate can include the final product substrate or a temporary substrate to receive the released LED device. The package can be a standard can, chip on board, or submount, or module device. After the device is packaged, it can be configured in one of a variety of applications. Of course, there can be other variations, modifications, and alternatives. Therefore, the above description and illustrations should not be taken as limiting the scope of the present invention which is defined by the appended claims.