As president of LED Lighting Technologies, Dr.Nisa Khan consults in the solid-state lighting industry and educates consumers about LED lighting. She has a bachelor’s degree in physics and mathematics, and master’s and Ph.D. degrees in electrical engineering. Email her at email@example.com
The 2010 LED market appears to be exploding beyond projections – it’s now forecast to reach $9.1 billion, instead of $8.4 billion that Strategies Unlimited researchers projected 14 months ago. Much of the growth stems from the evolving use of LED-based backlight systems in LCD screens, but, interestingly, that specialty’s growth rate is predicted to slow, while the high-brightness LEDs (HBLED) market could double over the next four years, to reach $18.4 billion.
Although current LED-lamp systems easily integrate into signage, display backlighting and various other markets, HBLED technology needs to progress before it can meet (or exceed) broad-use lighting-market projections. The “needs attention” areas are:
- The internal quantum efficiency (IQE) improvement at the LED-chip level, to (possibly) overcome the “droop” challenge that frustrates the industry. “Droop” defines the efficiency-drop phenomenon that occurs in LED lamps as electrical-power input is increased.
- To overcome the directional nature of LEDs. Significantly more uniform light distribution is required for general lighting applications.
- To increase LED lifespan through better thermal management.
- To identify critical cost factors and reduce LED production costs.
I’ve focused on three and four in previous columns and will revisit them with new information in the future, and, I’ll get into the directional problem within the next few months. Most importantly, in this column, I want to discuss the first subject –“droop” — and propose some new reasons for it, with verification ideas.
Higher LED chip efficiency
White LEDs are most widely used for signage, display illumination and lighting applications. The most commonly used technology for white LEDs comprises nitride-based, i.e., gallium nitride (GaN), indium gallium nitride (InGaN) or aluminum indium gallium nitride (AlInGaN), or some combination of such compounds in the active, light-emitting region. Nitride-based LEDs show a significant reduction, or “droop,” in IQE at higher injection currents.
Remarkably, the world’s smartest LED scientists are perplexed by droop behavior and vigorously strive to solve this industry-impeding puzzle. And, although researchers have recently provided various, efficiency-droop explanations, counter arguments and experiments show none are widely accepted.Advertisement
Switching on an LED diode requires forward-biasing of the p-n semiconductor regions that allows negatively charged electrons to recombine with holes (positively charged particles) to both gain and subsequently release energy in the form of light. This recombination process can be either radiative or non-radiative, but only the former type leads to electroluminescence or light emission. Various non-radiative processes, including non-radiative recombination, significantly reduce the IQE, which subsequently hampers the LED’s overall efficacy.
In particular, nitride-based LEDs’ non-radiative processes become more dominant at higher currents, and this, according to E. Fred Schubert et al of Rensselaer Polytechnic Institute, when observed at injection current densities exceeding values that range between 0.1 and 10 A/cm2, decreases efficiency.
Schubert’s group, and others who have analyzed the droop problem, claim the droop phenomenon, beyond these electric-current ranges, can only be comprehensively explained if the current leaking beyond the active region is included.
Although their research has identified several useful mechanisms for current leakage outside the active region, no one has included current leakage, due to the non-ideal, three-dimensional device structure that creates the undesirable current distribution inside and outside the active region.
Many state-of-the-art LEDs’ active regions contain quantum well (QW) structures that allow for enhanced carrier recombination, to boost efficiency. Schubert’s group has, thus far, identified the leakage mechanisms and recombinations outside of QWs, which includes a lack of electron capture into QWs; electron escape from QWs; the spacer-electron, blocking-layer interface’s electron-attracting properties, plus this layer’s p-type doping properties; and the GaN’s asymmetric electron and hole-transport properties.
A new mechanism for efficiency droop
In an LED chip, the current flows from the p-contact (anode) to the n-contact (cathode), and the current distribution is determined by the applied current magnitude, or forward voltage, as well as the three-dimensional chip structure, which includes the electrodes’ size, shape and positions, relative to the active regions.
The dotted-line arrows (Fig. 1) show a typical LED-chip structure where the current distribution in the active region is very non-uniform. Further, a good deal of injected current is wasted beyond the active regions because the electrodes are asymmetric. This non-uniformity and waste contribute to the efficiency drop; it may also become more significant at higher currents, because of stronger curvatures or bending in the electric field distribution near the p-electrode.
A leakage-current verification experiment
My analysis indicates current leakage may be a significant part of the efficiency droop. Overcoming such leakage will require device-structure optimization in all three dimensions. Here, then, I present an experiment that may reveal that this type of structural leakage causes the efficiency to drop at higher current densities.
One optimized structure identified to produce uniform current distribution in the active region is the superluminescent diode (SLD) — an edge-light-emitting LED that usually produces much higher brightness when compared to surface-emitting LEDs, due to the superluminescence action. Its geometric structure is very similar to that of a laser diode, but it has low coherence, like conventional LEDs, and also lacks such lasing properties as stimulated emission and optical feedback.
I propose the following experiment: Acquire an SLD (Fig. 2), with such dimensions as W = 5 micron (1 micron = 10-4 cm), L = 5mm, and apply one mA DC current to it, then measure the light coming out of its edges. The resulting structure’s sheet-current density would be 40A/cm2 – significantly higher than the range specified by Schubert, et al.
Alternatively, with a conventional LED, extend the p-contact over the entire top surface of 1mm X 1mm; apply 300mA to it, and measure the light emitting from its edges – treating it, thus, like an edge-emitting LED. The resulting current density of this structure is 30A/cm2 – also higher than the Schubert’s range beyond where droop is observed.
Because the current distribution is uniformly strong over the active region (Fig. 2), and because little or none of the applied current will be wasted outside the active region in such proposed structures, I believe the efficiency won’t degrade even at much-higher current densities, beyond the range where others have observed significant drop in efficiency in conventional LEDs with structures similar to Fig 1.Advertisement
This proposed experiment with SLDs, or regular edge-emitting LEDs, may show that droop can be reduced. But even if such a discovery is proven, the edge-emitting diode isn’t useful for large-space illumination; however, it could be ideal for edge-illuminated signs and displays.
Nevertheless, regardless of whether such an experiment can prove the main mechanism for efficiency droop, the problem still remains: How do we get a droop-free, white LED structure that also provides broad-space illumination and doesn’t produce narrow, linear illumination like edge-emitters? As you might expect, I plan to visit this topic again.
Find it here! ISA Sign Expo 2023
Whether you're looking to grow your business or advance your career, ISA International Sign Expo® 2023 is the only place to find everything you need to be successful in the sign, graphics, print and visual communications industry.
Photo Gallery4 hours ago
15 Super Wraps for the Super Bowl
Business Management3 hours ago
16 Tips to Become a Better Sign Business Negotiator
Photo Gallery2 days ago
15 Historical Photos of Signs in Color
News2 weeks ago
Buc-ee’s: Typo or Marketing Ploy?
News6 days ago
2024 Women in Signs Awards Nominations Now Open
Benchmarks1 day ago
5 Vehicle Wraps At Your Service
Product Buying + Technology1 week ago
Mark Kissling2 days ago
LED for Neon Retrofit for a 50-Year-Old Rooftop Sign