Pattison noted that although light source efficiency is a focus of the DOE SSL program, there’s tremendous untapped potential in all four areas. He observed that source efficiency for phosphor-converted LEDs is plateauing, but that the direct-emitter approach offers promise. Pattison explained that spectral efficiency involves optimizing for visibility, color rendering/highlighting, ambiance, human physiological response, roadway safety, wildlife response, skyglow impact, and plant growth properties. He emphasized that the value lies not just in generating photons, but also in organizing and controlling them.

Kevin Houser of Penn State University discussed efficacy in applied lighting, explaining that efficacy is a ratio of benefit to cost, but that lumens is not an appropriate proxy for benefit, and that watts is an imperfect proxy for cost because lighting also has other benefits, especially related to health, and all aspects of a life cycle analysis are cost factors. He said a family of efficacies – including such things as lumen hours per lifetime environmental cost, and circadian stimulus per watt – could cover common situations, but we need to move beyond a ratio to consider the larger picture. When lighting for people, Houser said, the benefits should be defined by human needs and the costs by the detrimental effects on our planet, but lumens is too blunt a metric to characterize optical radiation.

Robert Soler of BIOS Lighting talked about rethinking lighting application efficiency. He explained that room cavity ratio (RCR) is essentially the wall-to-floor ratio, and that fixture type can increase vertical illuminance, but only for certain RCRs. Emphasizing the need to understand multiple lighting parameters, including timing, spatial aspects, intensity, and spectrum, Soler said the best way to save energy is to focus light at what you want people to see, and he described two new lighting metrics: mean room surface exitance and task/ambient illumination ratio. He observed that whereas classic lighting was well suited for lighting the horizontal task, the task is now becoming self-illuminated and/or vertical. Noting that lighting floors first loses 80% of the lumens at first bounce, Soler considered whether spillover would be enough to light such things.

EFFICACY AND SAFETY WITH SSL

Monica Hansen of LED Lighting Advisors moderated a panel that looked at how lighting affects safety and where more research is needed. Joe Cheung of the U.S. Department of Transportation’s Federal Highway Administration discussed safety in roadway lighting. He noted that while studies have shown lighting can reduce nighttime fatal crashes by up to 60%, roadway lighting also has negative impacts, including health impacts, sky glow, glare, and impacts on flora and fauna. The keys for minimizing these factors, Cheung said, include reducing the amount of uplight and high-angle light, reducing the amount of disability and discomfort glare, and limiting the amount of light trespass. He discussed adaptive lighting as a potential means to reduce roadway lighting’s negative impacts and save energy, and said that adaptive lighting criteria are working their way into exterior lighting recommendations.

Don McLean of DMD & Associates discussed lighting safety, focusing on three areas related to pedestrians and cyclists: crosswalks, bike lanes, and security. He noted that lack of lighting is one of the primary causes of pedestrian fatalities, nearly three-quarters of which occur at night. McLean observed that the spectral power distribution of light sources is not considered when defining lighting levels but is an important factor. He noted that pedestrian collisions are mitigated by season, age, weather, speed, and other factors, and that lighting standards involving bicycles are currently very unclear and don’t deal with conflict points, with some applying vertical sidewalk levels and others simply applying roadway levels to bike lanes. Good visibility in crosswalks needs contrast, including improving the color contrast in the vertical lux.

Max Martell of the National Institute for Occupational Safety and Health (NIOSH) discussed human factors and lighting for underground mining. He noted that underground mines present unique lighting challenges, including low ceilings and very low reflectivity, a constantly changing environment, strict lighting regulations, unlit areas, an aging mining workforce, and the negative health effects of shift work and darkness. Martell observed that insufficient light has multiple impacts on safety, including decreased alertness, productivity, reaction time, and cognitive function, and increased accidents, health impacts, sensitivity to light, and fatigue. He discussed the use of LED lighting in mines and described a NIOSH study of human-centric lighting to mitigate circadian disruption and reduce accidents in mineworkers.

UNDERSTANDING AND DESIGNING EFFICIENT LIGHTING FOR VISION AND WELL-BEING

Bob Davis of PNNL moderated a panel that examined the generally agreed-upon scientific findings related to lighting for vision and well-being, including the potential for positive and negative outcomes. The panel then considered how existing knowledge can be embodied into design standards and examined the challenges in designing, installing, and evaluating lighting systems for vision and well-being, in order to identify the best practices to employ today and the future research needs to ensure that lighting is optimized for both energy efficiency and human response.

Shadab Rahman of Harvard Medical School talked about the physiologic effects of light. He explained that the light characteristics that modulate its nonvisual effects include not only its intensity and duration, but also its wavelength, timing, pattern, and photic history. Rahman observed that there’s a diverse range of physiologic responses to light, that they follow a dose response, and that the system involved is complex and dynamic. He noted that the short-wavelength-sensitive melanopsin response of the intrinsically photosensitive retinal ganglion cells (ipRGCs) dominates in long-duration, bright-light conditions; that visual photoreceptors modulate physiologic response to light; and that there’s significant variability from one person to another in how they respond.

David Sliney, a consulting medical physicist, discussed what’s known and not known about blue light hazard. He said despite the fact that many studies point to the potential role of light – especially short-wavelength light – as a cause or promoter of age-related macular degeneration (AMD), only limited epidemiological evidence supports that link. Sliney stated that the weight of scientific evidence indicates that intense sunlight accelerates retinal aging, that a subset of people endure a much higher exposure than most, and that the area of retina exposed impacts the ipRGC response. He pointed out that good lighting design blocks out most of the high-intensity light that is overhead, and that indoor LED lighting, if richer in short wavelengths, may hardly impact retinal exposure because of low blue-violet reflectance of viewed surfaces and also because it’s vastly less than the exposure we get when outdoors. “Comfort is probably a very good indicator of what’s really safe,” Sliney said.

Brian Liebel of the Illuminating Engineering Society spoke about how lighting standards evolve, starting with discovery and proceeding to consensus. He observed that the discovery of the ipRGCs unlocked previous mysteries involving visual metrics, circadian entrainment, and the impact of light on alertness, health, and well-being. Liebel noted that in addition to saving energy, LEDs offer potential for greater control of light than is offered by other lighting technologies – not just instant-on and dimming, but spectral tuning, wireless communication, and additional processing and feedback functions. This, he said, in turn confers the ability to be more precise and more flexible. The question, Liebel said, is knowing the right way to utilize these benefits, and how to ensure a proper balance between energy efficiency and quality of light and life.

Ron Gibbons of Virginia Tech Transportation Institute (VTTI) discussed the challenges for data collection in the outdoor environment, using as an example an ongoing DOE-supported VTTI study investigating the health impacts of outdoor lighting. He noted that the impact of roadway lighting on melatonin levels has never been measured in a naturalistic lighting environment, and that the ongoing study is looking at drivers, pedestrians, and those living nearby. Gibbons recounted how he and his colleagues screened more than 300 people to come up with the 37 who were recruited for 30 needed slots, but attrition reduced the final number to 29. Participants were required to come in once a week, he noted, plus as many as 12 staff members were required per night. Gibbons said the results should be complete by September.

Bob Davis spoke on behalf of Gena Glickman of the University of California at San Diego and the Uniformed Services University about a DOE-supported study she’s conducting to develop and test a novel lighting strategy for optimizing sleep, circadian health, and alertness in hospital night shift workers. Stating that light is probably the best tool we have to help shift workers, but that how to best do it remains an open question, Davis explained that Glickman’s study involves the use of actigraphs, from which sleep patterns are inferred based on activity, and that it combines two evidence-based lighting interventions that address two different responses. He noted that the circadian phase is synchronized by maximizing light input during the desired daytime and minimizing input just before and during desired bedtime, and acute alertness is engendered by using a special light on an individual basis prior to task performance.

Davis discussed the effects on energy use and human response of tunable LED lighting in real-world settings. He noted that at a nursing home in Sacramento, CA, the implementation of such a lighting system resulted in 68% energy savings in corridors, with residents sleeping through the night; a reduction in the use of psychotropic and sleep medications; and a 41% reduction in negative behaviors. In classroom settings in TX and CA, Davis said, the energy savings were 45%-60% based only on power reduction, with even deeper savings (>70%) due to dimming, and color consistency between fixtures and over dimming was very good. A medical behavioral health center in Seattle, he related, achieved 41% energy savings compared to a circadian non-tunable base case, with manual controls resulting in even greater savings; the biophilic design goals were supported; and nurses no longer needed to pester patients about going to bed.

POSTER SESSION

The evening of Day 1, researchers representing more than 30 projects presented their work at a poster session that provided attendees with additional opportunities for discussion, information exchange, and potential partnering. The projects were funded not only by the DOE SSL program, but also through DOE under the Small Business Innovation Research (SBIR) grants and the National Science Foundation (NSF) Established Program to Stimulate Competitive Research (EPSCoR). Two posters representing the SSL student competition included winners from the University of Houston and the University of Central Florida.

OPEN DISCUSSION SESSIONS

Day 2 concluded with four concurrent informal sessions that involved open discussion and Q&A on emerging areas of lighting science:

• Lighting application efficiency – Attendees explored ideas for developing a common framework for characterizing sub-efficiencies for different lighting applications and systems, and for developing ideal target spectrum and intensity levels for various lighted activities.
• Lighting for vision and well-being – Attendees examined the current understanding and disconnects between physiologists, technology developers, and implementers, and identified best practices and future research needs.
• Connected lighting – Attendees shared lessons learned related to data collection, energy reporting, interoperability, and value assessment.
• The DOE R&D funding mechanisms – Attendees had an opportunity to ask questions regarding the DOE SSL FOA process, the SBIR grant process, and the collaborative OLED testing opportunity.

TRACK SESSIONS

On Day 1 and Day 2 of the workshop, attendees split into separate LED and OLED track sessions, in order to take a deeper dive into technical discussions on a wide range of topics. Expert speakers kicked off technical discussions that examined current research efforts and future research needs, with an eye toward informing DOE SSL program planning.

LED Track Sessions

On Day 1, Monica Hansen moderated an expert panel that led a technical discussion on LED chip and materials advances, focusing on emitters, phosphors, and quantum dots (QDs). The speakers focused on state-of-the-art performance results and innovations that have led to improvements in their areas of materials advancements. The two main thrusts were development in green LED materials, and the challenges in improving performance in downconversion materials such as phosphors and QDs.

The opportunities identified by the panel included continuing to drive the importance of the high quality of light and to develop better lighting metrics and make the use of SPDs mainstream; conducting research into thermal dissipation approaches as the trend for high lumen density continues to increase; and improving multichannel drivers and sensors to enable color tunability performance in luminaires.

Morgan Pattison moderated an expert panel that covered a range of LED luminaire subtopics. Bob Davis discussed a PNNL study of very-high-efficacy lighting products, which sought to identify efficacy-performance tradeoffs that were apparent with such products. Arun Dutta of LEDVANCE covered the architecture of filament lighting products. LED filament lights require a specific type of LED package and have the benefit of looking like old-fashioned incandescent lights. They can have high efficacy but can also have issues with flicker. Dutta also discussed some additional new concepts being developed at LEDVANCE, including a night-day luminaire that shifts SPD and light level based on time of day for physiological impacts and horticultural lighting concepts.

Peter Kozodoy of Glint Photonics talked about the technology he developed that provides optical configurability or dynamic beam steering of lighting products. He discussed the motivation behind optical configurability or steerability, including form factor, optical performance, and sustainability benefits. Kozodoy also addressed the underlying technical approach. Marcelo Schupbach of Wolfspeed discussed the use of wide-bandgap (WBG) electronics for LED power supplies. He discussed the size and efficiency benefits of using WBGs but also the challenges of using them for lower power applications such as SSL.

This panel discussion highlighted the range of efforts for advancing SSL, including very high efficacy, alternative package and lighting-product form factors, dynamic beam steering, and new opportunities for smaller and more-efficient power supplies.

OLED Track Sessions

On Day 1, Lisa Pattison of SSLS Inc. moderated an expert panel that led a technical discussion on OLED materials and stack advances. Mark Thompson of the University of Southern California gave a brief overview of the challenge with blue emitters and various approaches to harvesting singlets and triplets for high-efficiency devices with long lifetimes. Currently, panel manufacturers use hybrid devices in which a blue fluorescent is stacked with red and green phosphorescent devices. Despite the low-efficiency fluorescent blue emitter, reasonable efficiency can be achieved, as there is not a large component of blue light in white light. Another hybrid-device approach, which Thompson is studying with DOE support, is to split the excitons into two populations, collecting the singlets on the blue fluorescent and passing the triplets to red and green, making 100% efficiency possible. Thompson also addressed oriented emitters and metal-centered dopants, and talked about recent work on air-stable copper compounds that show promise in place of iridium-based phosphorescent materials.

Christian Kasparek of Cynora discussed thermally activated delayed fluorescence molecules (TADF), which have seen a renewed interest in recent years as a possible path to high-efficiency, stable blue. In the TADF process, the singlet-triplet energies are close enough that thermal energy can upconvert triplets to the singlet state for up to 100% quantum efficiency. When the TADF molecule acts as the dopant emitter, this is standard TADF. In another approach (often referred to as hyperfluorescence), the TADF molecule is accompanied by a second emitter. Cynora uses the co-emitter approach and, today, can achieve 22% external quantum efficiency (EQE), 466 nm wavelength, and L95 of 20 h. These results are good, but a lifetime boost is needed. Much longer lifetimes are possible with longer-wavelength molecules. However, Cynora calculated that when using sky-blue materials in white OLED devices, the CRI and efficiency take a hit. The company expects to begin mass production of blue TADF materials in 2020.

Bernard Kippelen of Georgia Tech delved deeper into TADF, rethinking the photophysics of organic compounds. He pointed out that the science we use to understand OLED materials is based on flat, all-carbon conjugated molecules. We’re moving away from such molecules with the addition of heteroatoms, high-Z metals (Ir, Pt) for strong spin-orbit coupling, twisting of molecules due to steric hindrance, and many charge-transfer interactions and intersystem crossing between molecules, so that we are no longer limited to pi – pi* transitions. In their DOE-funded project, Kippelen and his team are looking at blue emitters and have also explored neat TADF films, where the TADF is a single emitting layer rather than a dopant in a host. Kippelen concluded by urging the community to rethink the basics and go beyond conventional wisdom in order to overcome the blue-emitter challenge.

After these three talks, the discussion centered largely around lifetime, with the question of whether TADF or hyperfluorescence will be able to achieve reasonable stability. In discussion of R&D priorities, the group felt that the stable blue material task,  as described in DOE’s R&D Opportunities document, was still a priority R&D topic, but felt that a color metric should be included. It was also recommended to incorporate a ∆u'v' metric as a function of device lifetime.

Lisa Pattison then moderated a second panel on advances in OLED light extraction. Steve Forrest gave a brief overview of the problem with OLED light outcoupling. Even if the materials are 100% efficient, due to waveguiding and surface plasmon modes, only 20% to 25% of the light escapes a planar OLED device. To boost that, approaches need to be found that are low-cost, suitable for large-scale manufacture, and effective. Forrest described four approaches he and his team explored with the help of DOE funding, noting that those approaches mostly focus on the plasmon modes. He advised those developing light-extraction technology to find solutions that are wavelength- and viewing-angle-independent, low-cost, and adaptable to both top- and bottom-emitting structures. Forrest stated that the practical limit for outcoupling is 70% to 80%.

Franky So of North Carolina State University said that he and his team have also been working under DOE funding on the topic of light extraction, particularly with the use of corrugated substrates to reduce waveguided and surface plasmon losses. There has been concern that corrugated substrates are not suitable for manufacturing because they could be prone to reduced yields, but So is confident that the approach has promise. He also addressed the effects of varying the refractive index of various layers within the OLED stack. In particular, So looked at the electron transfer layer (ETL). He showed how the index of refraction can be influenced by deposition conditions – in particular, oblique angle deposition. Using these deposition techniques and modifying the refractive index of the ETL, his team observed 25% extraction enhancement.

Selina Monickam of Pixelligent talked about high-refractive-index materials in OLED lighting applications. Pixelligent has received multiple rounds of DOE funding to develop light-extraction technology for OLEDs. Monickam and her team have been quite successful with the most market-ready approach available and have developed materials and technology for internal light-extraction films for the major global OLED lighting manufacturers. Pixelligent’s key technology is high-refractive-index materials formed from ZrO2 nanocrystals in a transparent polymer matrix that can be incorporated as a smoothing layer in OLED stacks to provide index matching for light outcoupling and to serve as a host for scatterers. For flexible substrates, Pixelligent has had to reformulate the high index matrix. Monickam and her team are working with OLEDWorks to optimize. Manufacturers currently incorporate internal light extraction only in rigid panels, due to a lack of solutions for flexible devices.

The ensuing discussion focused on DOE metrics. It was agreed that light extraction remains a priority R&D challenge for OLED lighting. The group felt that the metrics and task description as found in the DOE R&D Opportunities document described the challenge well. However, it was felt that the driver efficiency targets could be increased from current values of 85% (2018) and 90% (2020) to values of 92% to 93% in 2020. This would increase the targets for overall luminaire efficacy. It was also pointed out that lighting panel size and functionality (beam-shaping capability, flexibility) could have defined targets.

On Day 2, Norman Bardsley of Bardsley Consulting moderated an expert panel that led a technical discussion on advances in OLED integrated substrates. Jeff Spindler of OLEDWorks gave an overview of the work that must be done in preparing a foundation for the deposition of the organic materials that convert electricity to light. R&D is needed in each of three major components: thin, conformable substrates that protect against ingress of oxygen and water vapor; internal light-extraction layers; and transparent conductors. New manufacturing methods are needed to integrate and pattern these layers at substantially lower cost. The short-term approach of OLEDWorks to producing conformable panels is to use ultra-thin glass, but Spindler and his colleagues also recognize the need for devices based upon plastic substrates.

Mark Taylor of Corning provided a status report on the collaboration between Corning Glass and OLEDWorks. Substantial progress has been made in edge healing and in the processes that are used to bond and then release the ultra-thin glass to a carrier during manufacture. However, further work is needed on packaging of flexible panels. Although OLEDWorks and Corning are strong supporters of the development of roll-to-roll manufacturing, they estimate the cost savings to be 15% to 25%. This could be a significant contribution, but much larger reductions are needed if OLED lighting panels are to succeed in many general lighting applications. On the other hand, applications to automobile lighting have shown that OLED luminaires can already be cost-competitive at the luminaire level.

Whitney Gaynor of Sinovia Technologies described the successful integration of Sinovia’s transparent conductors, based upon silver nanowires, with moisture barriers from Vitriflex on inexpensive plastic substrates. The team met targets for electrical conductivity, optical transmission, and smoothness, and came close to the goal for device lifetime, but reproducing this success in high-volume production will be challenging. Gaynor and her colleagues also demonstrated remarkable progress in patterning using flexographic printing in roll-to-roll mode. It seems likely that transparent conductors using ITO or silver nanowires will provide acceptable solutions for small and medium-size panels. In larger panels, the conducting layer may need to be supplemented by wire grids. Printing seems to be necessary to meet cost targets, but the conductivity of printable metal inks has been much less than that of the bulk metals.

Paul Leu of the University of Pittsburgh described a DOE-funded project in which particle-free silver inks from Electroninks have been used to form grids at temperatures below 120 °C, achieving conductivity up to 90% of that of bulk silver. This will enable devices to be formed on most plastics and on textiles. The combination of high optical transparency and electrical conductivity appears to yield results well in excess of any obtained with nanoparticles or flakes.

While it’s clear from hearing these panelists that DOE support has led to substantial progress by innovative companies in developing potential solutions to major challenges faced in OLED lighting, and that the DOE Collaborative OLED Testing Program has been valuable in checking this progress in a manufacturing environment, the refinement of these new materials and processes to support reliable, cost-effective manufacturing remains a major challenge.

Bardsley then moderated a second panel on advances in OLED encapsulation and back-end processes. Michael Boroson of OLEDWorks gave an overview of the progress to date in OLED lighting, stating that it has tracked that of LEDs with a five-year delay, and he forecast the potential for further development with continued DOE R&D funding. Boroson also outlined the manufacturing steps needed to produce an OLED luminaire, including 11 processes after the formation of the core OLED stack. He summarized the attributes that will be required to contribute to energy savings through substantial penetration of the general lighting market and the achievement of more-realistic cost targets, which have become possible through the experience OLEDWorks has gained on its production line. Major progress has been made by the manufacturers of OLED displays in the fabrication of flexible panels on plastic substrates and in encapsulation techniques to prevent damage due to the entry of oxygen and water vapor. Much of the equipment to accomplish this has come from U.S. suppliers, but it’s far too expensive for lighting applications.

Ofer Sneh of Sundew Technologies discussed the work he and his colleagues are doing to simplify the process of OLED encapsulation using atomic layer deposition. This approach is being tested at OLEDWorks and compared with the more traditional plasma-enhanced chemical vapor deposition (PECVD) and physical vapor deposition (PVD, also known as sputtering).

Larry Sadwick of InnoSys discussed the relative merits of LED and OLED sources in lighting and pointed out the advantage of hybrid devices. He stressed the potential advantage of OLEDs with respect to color and recommended further work on color tuning. Sadwick showed a block diagram for OLED power suppliers and drivers, suggesting several different configurations. His more detailed discussion of drivers and controls contained more questions than answers and clearly demonstrated the need for more research and innovation into the special characteristics of OLED drivers.

During the discussion of the integration of OLED panels into luminaires, the importance of DOE support for the development of innovative prototypes was stressed, along with the value of GATEWAY demonstrations in enabling such prototypes to be evaluated both by technical experts and potential adopters.

NEXT STEPS

Erika Gupta of DOE concluded the three-day workshop by thanking the attendees and speakers for their input and participation, which help to shape DOE SSL Program planning. The workshop presentations are available for download on the DOE SSL website.