Congress, recognizing the potential for energy savings in the use of general lighting for illumination, requested in the Energy Independence and Security Act of 2007 (EISA 2007) that the Department of Energy (DOE) contract with the National Research Council (NRC) to conduct a study to assess the status of solid-state lighting (SSL) as a technology. The contract requested that the National Academies provide an objective and independent assessment of the current state of solid-state lighting and its future potential for accommodating the new minimum efficiency standards for lighting. The NRC established the Committee on Assessment of Solid State Lighting (Appendix A) composed of diverse experts in the fields of solid-state physics, electronics, lighting design, human perception of light, industrial commercialization, and policy. The statement of work directed the committee to review the development and future impacts of SSL, including projections of cost and research and development (R&D) necessary to overcome barriers to widespread adoption and the potential for unintended consequences of deployment.
Solid-state lighting consists of two technologies—the inorganic semiconductor-based light-emitting diode (LED) and the organic polymeric-based light-emitting diode (OLED). Both technologies are the subject of active research worldwide. The LED technology is currently in the early stages of commercial deployment while OLEDs are in the demonstration phase. All LED-based luminaires1 require optics to distribute the unidirectional light emitted by the LED and a large heat sink to maintain the LED temperature within limits. Furthermore, like fluorescent lamps,2 both LED and OLEDs require they be supplied through an electronic circuit to provide them with the proper form of electric power. Both LEDs and OLEDs and the luminaires based on them are discussed below.
The committee’s main findings and its key recommendations for the Department of Energy are listed in Boxes S.1 and S.2, respectively.
LED-BASED SOLID-STATE LIGHTING
Technology and Lighting Products
Because a single LED emits light that is monochromatic, devices that emit white light must do so by combining the emissions of individual red, green, and blue (RGB) LEDs or by using a single blue LED whose emission excites a phosphor, which in turn emits white light. This latter design is the more common. The efficacy3 of white LEDs has been increasing rapidly and is expected to approach 200 lumens/watt (lm/W) by 2020, greatly exceeding all other general illumination technologies. The committee found that luminaires and lamps based on LEDs will be able to support the lumen output standards Congress required to be promulgated by DOE in Section 321 of EISA 2007.
LED-based lighting products currently are available in two forms. The first consists of lamps that can replace, one-for-one, the incumbent lamp (i.e., the lamp that is currently in use) without modification to the original fixture. The LED, optics, heat sink, and electronic drivers are all packaged in the replacement lamp, typified by the screw-in replacement—now offered by Cree, Philips, OSRAM Sylvania, and others—for the familiar household incandescent bulb (the “A-19”). A 60 W A-19 lamp produce about 850 lumens, for an efficacy of 14 lm/W. The high-quality commercial LED equivalent produces 93 lm/W. For comparison, the equivalent
1 A luminaire consists of, minimally, a lamp holder, commonly called a socket, and the way to connect the socket to the electrical supply. Most fixtures also contain optical elements that distribute the light as desired, such as a reflector, lens, shade, or globe. When needed, fixtures and luminaires contain a ballast or a driver.
2 The term lamp is equivalent to the term light bulb in every-day usage, i.e., the source of light that attaches to the luminaire by means of a screwbase or pins.
3Efficacy is a measure of the efficiency with which a lamp or luminaire converts electricity to useful light. It is defined as the ratio of the luminous flux to the total electrical power consumed and has units of lumens per watt.
Finding 1: Luminaires and lamps based on LEDs will be able to support the standards for lumen output Congress required to be promulgated by DOE in Section 321 of the Energy Independence and Security Act of 2007.
Finding 2: Cost is the biggest obstacle to the widespread deployment of SSL based on LEDs.
Finding 3: The Bayh-Dole waiver is discouraging some universities and small companies from participating in the DOE program.
Finding 4: On a lifecycle basis, warm and cool white LEDs are already cheaper than incandescent lighting and will likely be comparable to that of fluorescent lighting technologies in the near future.
NOTE: The full text of all findings and recommendations in the report appear in Chapter 7.
Recommendations to the Department of the Department of Energy
Recommendation 1: The Department of Energy should continue to make investments in LED core technology, aimed at increasing yields, and in fundamental emitter research to increase efficacy, including improvements in the controlled growth and performance of the emitter material.
Recommendation 2: The Department of Energy and lamp manufacturers and retailers should work together to ensure that consumers are educated about the characteristics and metrics of these new technology options.
Recommendation 3: The Department of Energy should support research to understand the fundamental nature of efficiency droop at high currents in OLEDs and to seek means to mitigate this effect through materials and device architectural designs.
Recommendation 4: The Department of Energy should focus on efforts that result in significant light outcoupling enhancements for OLED that are low cost to implement and are independent of both wavelength and viewing angle.
Recommendation 5: The Department of Energy SSL program should be maintained and, if possible, increased.
Recommendation 6: The Department of Energy should seek to obtain 50 percent cost sharing for manufacturing R&D projects, as was done with the projects funded by the American Recovery and Reinvestment Act.
Recommendation 7: The committee recommends that the Department of Energy consider ending its waiver of Bayh-Dole for SSL funding.
NOTE: All the findings and recommendations presented in the report are collected in Chapter 7, where the recommendations are double-numbered to indicate the chapter in the main text where they appear in context.
spiral tube compact fluorescent light (CFL) has an efficacy of 63 lm/W. LED lamps have also been developed as dropin replacements for lamps with other form factors, such as 4-foot linear fluorescents, although the total light output is lower.
The second product form is the retrofit luminaire, which is similar to many existing non-SSL products and requires complete removal and replacement of the incumbent luminaire— recessed troffers, high-bay fixtures, track lighting, and pendant lights, for example. Two further applications in which LED-based luminaires have performed well are downlighting,
including recessed cans, where the directional quality of the emitted light is important, and roadway lighting, in which a premium is placed on durability and low maintenance. In 2010 LED luminaires had achieved a 4.3 percent penetration in this latter application. SSL products for downlighting have efficacies of 35 to 85 lm/W compared to 10 to 30 lm/W for fluorescent and halogen luminaires in the same application.
Role of Standards and Testing
Because of the different spectral, electrical, and thermal characteristics of LEDs, OLEDs, and SSL products, existing standards to measure the photometric properties (i.e., measures of perceived light intensity) and colorimetric properties (i.e., measures of perceived color characteristics) of other lighting technologies frequently cannot effectively be employed. A number of standards development organizations are involved in recommending test procedures for the measurement of LEDs, OLEDs, and SSL products.4 The United States has taken early leadership on several influential standards, such as IES LM-79-08 “Electrical and Photometric Measurements of Solid-State Lighting Products,” which specifies the procedures for measuring total luminous flux, electrical power, luminous efficacy, and chromaticity of SSL lamps and luminaires. Despite rapid progress, a number of important test and measurement standards still need to be developed for SSL to be successful. For example, there is currently no way to measure or estimate the lifetime of SSL luminaires.
The committee found that cost is the biggest obstacle to the widespread deployment of SSL based on LEDs. The high cost relative to conventional light sources is due to a combination of costs associated with the LED device, heat sink, electronics, and packaging, each of which is the subject of substantial R&D activity. All categories of cost will need to be addressed along the value chain to improve the value proposition of higher-quality light, longer product life, and overall lower life-cycle cost compared to current lighting products on the market. Thermal management is particularly challenging because the LED chip must be kept at a temperature below 200°C. The small size of the chip means that even a watt or two of dissipation will raise its temperature well beyond this limit if adequate heat sinking is not provided.
Cost of the LED device is primarily driven by two issues: (1) the mismatch in thermal expansion between the sapphire substrate on which the LED is grown5 and the nitride LED material, resulting in thermal stresses that decrease device yield; and (2) because of process variability, the variability in the emission characteristics (color) among individual LEDs, which necessitates they be sorted (i.e., “binned”) and grouped for consistency. The efficacy of the LED is limited by both physical mechanisms within the semiconductor material and the limited ability to access light trapped in the substrate and emissive layers (i.e., improved outcoupling). Increasing efficacy not only improves energy savings, but also has a strong leveraging effect on the cost of LED lamps and luminaires because, as less heat is generated, smaller and less complicated thermal management and packaging systems are required. The committee recommends that the Department of Energy continue to make investments in LED core technology, aimed at increasing yields, and in fundamental emitter research to increase efficacy, including improvements in the controlled growth and performance of the emitter material.
In addition to cost, consumer acceptance of SSL will depend on an understanding of its unique characteristics and the new vernacular used to specify it. To this end DOE has created the Lighting Facts label for SSL lamps, which provides specifications for luminous output (lumens), power (watts), efficacy (lumens per watt), color temperature, and color rendering index (CRI). The Environmental Protection Agency (EPA) and ENERGY STAR® program are also engaged in developing informative labeling for SSL products. But consumers must be made aware of the significance of label parameters, and to this end EISA 2007 authorized $10 million a year to advance public awareness. This money has yet to be appropriated. The committee recommends that DOE and lamp manufacturers and retailers work together to ensure that consumers are educated about the characteristics and metrics of these new technology options.
Poor experience with spiral CFL lamps has made consumers skeptical of new lighting technologies. But unlike spiral CFLs, SSL turns on to full brightness instantly, is unaffected by low temperatures, has good color quality, and is inherently dimmable with properly designed lighting controls. However, a number of SSL performance characteristics may jeopardize consumer acceptance if not addressed. The most significant of these is the incompatibility of SSL lamps with many existing dimming controls, precluding a simple SSL retrofit, particularly in residential applications. Although unlike CFLs LEDs are in principal
4 These include, but are not limited to, the following: the Illuminating Engineering Society of North America (IES or IESNA), a professional organization; the International Electrotechnical Commission (IEC), an international consensus standards organization for electrotechnology; the International Commission on Illumination, an international standards body; the National Electrical Manufacturers Association; and the Underwriters Laboratory, which sets safety standards. The American National Standards Institute also provides accreditation and serves as the U.S. national member organization to the IEC.
5 Some devices are grown on a silicon carbide (SiC) substrate, which some manufacturers believe to be a better, albeit more expensive, alternative.
easily dimmed, their low current and driver electronics require special controls. National Electrical Manufacturers Association (NEMA) is working on a standard to address this issue (NEMA SSL 7-2012; Phase Cut Dimming for Solid State Lighting: Basic Compatibility). There is, at present, no standardized method for measuring the lifetime of SSL products, even though lifetime is a critical parameter in economically justifying SSL. Consequently, lifetime is a missing metric on the Lighting Facts label. DOE has instead recently incorporated a lumen maintenance metric, LM-80.6 This metric gives the number of hours of operation before the lumen output of the LED emitter degrades to 70 percent of its initial value (the so-called L70 point). This metric does not apply to product (luminaire) life, and if the durability of the balance-of-product does not match the expected 25,000-hour life of the LED emitter, the committee expects there will be negative consumer reactions.
The color of illuminated objects is also a key determinant of the perceived quality of lighting products, and in this regard the CRI of LEDs can be very high—comparable to high-CRI fluorescent lamps. There is consensus, however, that improved measures of color quality7 are needed to guide manufacturers, which, for SSL products, can be more numerous and much smaller in size compared to the incandescent lamp market. This diffuse supplier market compounds the problem of industry standardization.
ORGANIC LED-BASED SOLID-STATE LIGHTING
The OLED offers the possibility of unusual form factors by taking advantage of the inherent slim, flexible character of the device itself, and by leveraging its area-source characteristic to develop possible new applications. Although some OLED-based luminaires are commercially available, their present costs limit widespread adoption. The lifetime of an OLED is very sensitive to its exposure to both air and moisture, making the hermetic sealing of large, flat packages critically important. Both lifetime and efficacy are also negatively impacted by the high currents required to generate light of brightness sufficient for general purpose lighting, leading to the phenomena of current droop and thermal droop (a decrease in lumen output with increasing current or temperature). The committee recommends that the Department of Energy support research to understand the fundamental nature of efficiency droop at high currents in OLEDs and to seek means to mitigate this effect through materials and device architectural designs. Color consistency among OLED panels forming a luminaire is also a challenge. While OLEDs are employed extensively for displays, displays do not require the large area packages or high levels of illumination of general purpose lighting. Perhaps the largest efficiency gain that has yet to be achieved is improved outcoupling of light in OLEDs, made particularly difficult compared with LEDs by the large areal dimension and integrated form factor of the former. The committee recommends that the Department of Energy focus on efforts that result in significant light outcoupling enhancements that are low-cost to implement and are independent of both wavelength and viewing angle. While there is a manufacturing infrastructure for OLED displays, located almost exclusively in Asia, there is currently none for lighting products.
DOE LIGHTING PROGRAM
Solid-State Lighting has been funded in recent years at roughly $25 million per year, of which roughly $9 million was directed toward R&D in FY2011, emphasizing three interrelated thrusts: (1) core technology research and product development, (2) manufacturing R&D, and (3) commercialization support. The SSL Manufacturing Initiative was added to the SSL R&D portfolio in 2009 with the aim of reducing costs of SSL sources and luminaires, improving product consistency and maintaining high-quality products, and encouraging a significant role for domestic U.S.-based manufacturing.8 The DOE Lighting program also addresses issues related to commercialization. It supports independent testing of SSL products, supports exploratory studies on market trends and helps to identify critical technology issues, supports workshops to foster collaboration on standards and test procedures, promotes a number of industry alliances and consortia, disseminates information, and supports a number of other initiatives. It also conducts technical, market, economic, and other analyses and provides incentives to the private sector to innovate.
DOE has done a remarkable job of helping to advance SSL R&D and manufacturing and educating the lighting community, and the committee recommends that the Department of Energy’s SSL program be maintained and, if possible, increased. However, the committee notes that the percentage of matching funds from R&D grant recipients has declined in the past few years. The committee recommends that the Department of Energy seek to obtain 50 percent cost sharing for manufacturing R&D projects, as was done with the projects funded by the American Recovery and Reinvestment Act. In addition, the committee found that the Bayh-Dole waiver is discouraging some universities and small companies from participating in the program. The
6 Illuminating Engineering Society, IES LM-80-2008, Approved Method for Measuring Lumen Depreciation of LED Light Sources.
7 At present, the color rendering index, managed by the International Commission on Illumination (CIE), is the internationally accepted metric for the evaluation of a light source’s color rendering abilities and was developed in response to the advent of fluorescent lamps.
8 U.S. Department of Energy. 2011. Solid-State Lighting Research and Development: Manufacturing Roadmap. Washington, D.C.
BENEFITS OF DEPLOYMENT OF SSL PRODUCTS
The committee estimated the prospective benefits of reduced energy consumption from the deployment of SSL lighting products. The committee calculated benefits in two scenarios, each measured against a counter-factual baseline in which there were no impacts from EISA 2007. The first scenario calculated the savings that would accrue based on the lamp efficacy standards in EISA 2007 Section 321, and it is estimated that electricity consumption for lighting would be reduced by 514 terawatt hours (TWh9) in the residential sector and 60 TWh in the commercial, cumulative from 2012 to 2020. In a scenario with more aggressive assumptions on the improvements in the efficacies of LED luminaires, cumulative savings in the residential sector over the same time period were 939 TWh and in the commercial sector 771 TWh.10
The committee prepared a first-order comparison of the consumer life-cycle costs of lighting consumption in a further two scenarios for daily usage of lights: 3 hours per day (h/day) and 10 h/day.11 These two scenarios are representative of average daily usages in the residential and commercial sectors, and the results are found to be very sensitive to the number of hours of use. The committee found that on a life-cycle basis, warm and cool white LEDs are already cheaper than incandescent lighting and will likely be comparable to that of fluorescent lighting technologies in the near future. For applications where the daily usage is larger than 10 h/day, cool, white LEDs now have a similar consumer life-cycle cost to that of CFLs or T12 linear fluorescent tubes.
With continued U.S. government support and funding and DOE leadership, the promise of low-cost and very efficient solid-state lighting could be realized, lowering U.S. energy needs and allowing the United States to be a significant solid-state lighting manufacturer and technology provider.
9 The assumed lamp efficacies were as follows: 96 lm/W in 2010; 141 lm/W in 2012; 202 lm/W in 2015; and 253 lm/W in 2020.
10 A typically sized electric power plant of 500 megawatt capacity, operating 5,000 hours, would generate 2.5 TWh in a year.
11 The following assumptions were used: a retail electricity price of 0.11 $/kWh and a 10 percent discount rate, reflecting the implicit discount rate of the consumer. It is further assumed that a 60 W incandescent light bulb would be replaced by another lighting technology providing the same energy service (approximately 850 lumens).