GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Green Building News

LEDs Could Get A Lot Cheaper

As researchers develop new materials to replace rare-earth elements, steep cost reductions are possible

A new generation of LEDs could use a different class of phosphors, dramatically reducing costs, according to researchers.
Image Credit: Team EarthLED under license to Flickr

Researchers at Rutgers University have found a way to replace rare-earth elements in light-emitting diodes with much cheaper and more abundant materials. The cost of the new phosphor materials may be only 10% of the cost of rare-earth phosphors, according to a report published at Inside Climate News.

LED lamps use 10% to 20% of the energy of incandescent bulbs, and between 50% and 85% of the energy of compact fluorescent lamps. But LEDs also are more expensive because of the rare-earth elements used to manufacture them.

Over the past decade, rare-earth elements have increased in cost by as much as 1600% because of increased demand and fewer exports from China, which produces more than 90% of the world’s supply.

So researchers at Rutgers looked for alternatives. According to the report, they developed new combinations of materials to create new classes of phosphors which not only are free of rare-earth elements but also free of toxic materials such as cadmium and selenium.

“The prices of these rare earth elements have gone through the roof,” Zhichao Hu of Rutgers University told Inside Climate News. “That makes us think, ‘Is there a way we can bypass these rare-earth elements? Is there a way that we can use earth abundant elements to make phosphors that have the same properties, emit the same light, but are cheaper?’”

The findings were presented at the American Chemical Society’s annual conference in Boston last month.

The American Chemical Society said in a press release that LEDs produce the soft, white light consumers favor in a roundabout way. LEDs typically have a single semiconductor chip that produces light (usually blue) that must be shifted to white with the help of a yellow-emitting phosphor coating. That phosphor is now made with rare-earth elements, and that’s the problem the Rutgers team has been working on.

Huge drop in power use, or maybe not

A switch to LED lighting could cut the amount of electricity used for lighting in the U.S. by 50% by 2030, according to the U.S. Department of Energy.

But the report suggests that’s unlikely because of the “rebound effect”: As the cost of lighting falls, people use more of it.

The report, for example, cites a 2010 study by researchers at Sandia National Laboratories who found a direct connection between lower cost and higher consumption. Another study, this one published earlier this year, said that energy used for residential lighting will probably fall in the short term but energy savings will decline, or even be eliminated, as the amount of lighting increases over time.

“The multifunctionality of LED lighting may cause consumers to use significantly more light, creating the potential for both rebound and backfire to occur,” the study in the Journal of Industrial Ecology said. “The results indicate that the adoption of CFL and LED lighting will decrease residential energy consumption if consumers continue to use the same amount or slightly more light; however, when an expansion of lit spaces is included or a large increase in lighting usage occurs, energy consumption will increase and, over time, reduce or completely erode energy savings.”

And even with a projected 90% drop in the cost of phosphor materials, Inside Climate News suggested retail prices of LEDs wouldn’t automatically follow suit.

Ram Sephardi of the University of California Santa Barbara said that the new technology looks promising, but the “proof of the pudding” would be whether manufacturers adopt it.


  1. GBA Editor
    Scott Gibson | | #1

    Thanks for your correction.
    You're right. I should have been more careful with the wording at the top of the article. As you note, the potential savings for the phosphors doesn't necessarily translate into an equal reduction in retail cost of the LED. The article makes this point later ("And even with a projected 90% drop in the cost of phosphor materials, Inside Climate News suggested retail prices of LEDs wouldn't automatically follow suit.") but the first paragraph is misleading.

    As to the relative efficiencies of the different types of bulbs, Inside Climate News said: "An LED that uses 6 watts of electricity, for example, produces the same amount of light as a 14-watt CFL bulb or a 60-watt incandescent bulb." The report didn't list the specific lamps it used for the comparison.

    One Department of Energy report I found says the output of LEDs ranges from 78 to 90 lumens per watt, depending on the type of lamp, similar to what you found at the home center. Cree, however, says it developed an LED that produces 303 lumens per watt. A company called Nanoleaf advertises a 1800 lumen LED that uses 12 watts of electricity (150 lumens per watt). And elsewhere, a Department of Energy fact sheet shows a range of outputs for incandescent, CFL and LED lamps, not a single value for each.

    The Inside Climate News report may have used generalizations about the relative efficiencies of these bulbs that aren't reflected in what you can buy at your local hardware store or home center. Thanks for clarifying this.

  2. GBA Editor
    Martin Holladay | | #2

    The article has been corrected
    I'd like to thank Charlie Sullivan for his astute comments, and I'd like to thank Scott Gibson for his clarifications. We appreciate corrections from sharp-eyed readers.

    I have edited the article to address Charlie's legitimate points. Again, thanks.

  3. charlie_sullivan | | #3

    Hype Alert!
    [Edit: The errors I pointed out below were promptly corrected. Kudos to Scott and GBA for prompt and accurate corrections. I'm leaving the details here as footnotes to help document why the GBA report is now more accurate than some other reports on this advance.]

    I'm excited to hear about this research and I agree that there are great things that could be done with improved, low-cost phosphors, but I'm afraid that the hype has gotten compounded in the string of reporting here, to the point that the numbers here don't have much connection to reality.

    1. "A step that could lower costs for LEDs by as much as 90%." Actually, the researcher only claimed that he could reduce the cost of the phosphor by nearly 90%, not the cost of the LED. The phosphor is just one material in the LED, the cost of the LED includes the cost of a semiconductor manufacturing process as well as the cost of the materials. And the LEDs, in turn, are but one component of the cost of an LED bulb, which also includes a power supply and usually a heat sink. For example, the one pictured in this article has a massive an expensive aluminum heat sink.

    The blog post above suggests that the retail price might not drop as much as 90%, but implies that's a question of whether the manufacturers pass the savings on to the customer. But the 90% cost savings is only on a component of a component of the cost of the bulb, and even if the savings were fully passed on to the consumer, they might not even be noticeable.

    Source: the press conference youtube video:
    For the "nearly 90%" reduction in phosphor cost, start listening at 6:00.

    2. "LEDs use 10% of the energy of incandescent bulbs, and less than half the energy of compact fluorescent lamps.

    Those are both exaggerations. LEDs and LED bulbs are available with a wide range of efficacies. For comparison, I looked at best selling 60-W equivalent LED and CFL bulbs, in stock at big-orange-box home centers. A standard 60 W, 840 lumen incandescent provides 14 lm/W. A $8 Cree LED bulb provides 74 lm/W. That's fabulous and a great price and you should go buy lots of them, but 10X14 = 140, not 74. And the best-selling CFL, from ecosmart, provides 64 lm/W. That's 15% worse than the LED, not 50% worse.

    Footnotes: I chose 60 W equivalent nominal, because LEDs do better there than at higher wattages. And if you are picky about color rendering and get a high color rendering LED, that could also make it worse. This LED has CRI = 80, whereas the CFL is a little better with CRI = 82. High efficacy high CRI LEDs exist but they are hard to find.

    Despite the mistakes in the reporting (including mistakes from the source this post was based on), I'm excited to hear about this...I'll explain more about why in another comment.

  4. charlie_sullivan | | #4

    60 lm/W, but I still like the technology
    A few more details on evaluating the technology, and then why I'm excited about it even though the cost savings are limited.

    The press conference includes, around 17:00, a report of the lm/W achieved with the new phosphors: 60 lm/W. That's worse than a typical CFL, and was likely measured at more favorable conditions than in a real bulb (e.g. at low temperature, and without power supply losses--those are part of the reason Cree can legitimately report that they have lab results achieving 303 lm/W, while their bulb products only achieve 1/4 of that). Likely performance of the new phosphor and systems using it can be improved with more research, but as of now, the reported performance is worse than CFLs.

    And here's an estimate of the cost savings if the cost of the phosphor is reduced by 90%: Rare earth phosphor costs are about $1500/kg. That sounds astronomical, but since only about 10 mg is needed in an LED bulb, that's about 1.5 cents per LED bulb. Even if I'm off by a factor of 10, that's only 13.5 cents savings per bulb.

    I also tried looking at quantity prices for Cree blue LED and for the same type of LEDs coated with a phosphor to make white light, in order to tease out the cost added by the phosphor material (and the coating process). But the blue LEDs were actually more expensive than the white ones. That proves that the phosphor can't cost very much, but it doesn't nail down the actual cost--the blue ones are just more expensive because they are sold in smaller volume.

    But I'm still excited about the research. One reason is that some of the highest performance LED bulbs, like the unfortunately discontinued Philips L-prize bulb, use a large-area phosphor mounted remotely from the LED, rather than applying the phosphor to the surface of the LED. That means the heat generated by the phosphor doesn't heat the LED, which in turn makes the LED more efficient and/or makes it cheaper to provide heatsinking for the LED. It also means that the phosphor lasts longer, because it runs cooler and doesn't have as intense optical radiation hitting it.

    And even if the savings is only $0.01 per LED bulb, we need to make some tens of billions of them, so that is actually a fabulous payoff for the government's investment in the research.

  5. vensonata | | #5

    law of diminishing returns
    Led's, which I have had throughout the house for 5 years were a reasonable savings in energy from our former, quite pricey cfl's, the leds were pricey too, but worth it, considering off grid electricity prices at the time. The big improvement was light quality, most of the bulbs are 8 watt bulbs. But I asked myself, what if they came out with 4 watt bulbs with equivalent lumens? A 50% savings! But 50% of what? In winter, lighting a 10,000sq ft building for about 1 kwh day. So .5 kwh saving for 150 days, 75 kwh/year. that is worth $9 year. Hmmm, 30 new bulbs at $5 = $150 so a 15 year payback. Or a 150 watt pv panel to add to the system. Production 225kwh year. Three times as good an investment and a longer life than the bulbs, so perhaps 4 times the value. So at this point any improvements in led bulbs cannot compete with Pv "additional" panels. (Price of PV per watt should not always be calculated by whole system costs. One can add just panels sometimes for less than $1 watt to an already existing system.)
    So basically the 85% improvement of led over incandescents was great, but the fun is over. Could we call our present leds "pretty good bulbs"?

Log in or create an account to post a comment.


Recent Questions and Replies

  • |
  • |
  • |
  • |