The life and Times of an LED product

The LED is the runaway success story of modern electronics. It’s not just the lighting industry – LEDs are taking over automotive lighting and flat-screen TVs, not forgetting smartphones, computers, light-up Santas and flashing Madonnas. It’s difficult to find figures for LED lighting, but I came across these figures for TVs: the average LED TV has around 1,500 LEDs inside it. It has been estimated that the global market for LED TVs is around 100 million units a year, and growing. Do the sums and that’s a material weight of 34 million kilos of LEDs – per year. And the architectural lighting market will dwarf those figures. 

 

If you take at face value the vague claims made about LEDs being ‘environmentally friendly’, thenthis seems like great news. Sadly it’s not that simple. The working end of LED technology is a very strange world indeed. The growth of semiconductor technology (not just LEDs) has driven a massive increase in the extraction and processing of all sorts of odd metals to meet the demands of this new industry. But when it gets to the final product, there’s barely anything there. To make all these little LEDs, we’re ripping raw materials out of the earth at a head-spinning rate. One of the most common versions of LED chip used in lighting is the InGaN chip. 

 

This uses the usual gallium nitride structure, but it also contains indium. And while gallium is reported to exist in untold reserves around the planet, indium is already on the ‘at risk’ register, with 20 years’ worth of reserves. So – yes – LEDs are environmentally friendly should you happen to swallow one, but just throwing them away to landfill could make the LED revolution short-lived. And let’s not forget that most LED units contain lead, copper and nickel – all contaminants. It’s reckoned that 95 per cent of an LED chip can be recycled and put back into the materials supply chain. But it takes a bit of effort. Straight recycling involves crushing all those tiny LED chips and then separating out the various materials through a series of increasingly chemical processes. 

 

The extraction process produces gallium and indium with around 99 per cent purity, but that’s not good enough for the semiconductor industry. Further refining is needed, and that involves immersion in sulphuric acid and other tasty solutions. Of course, one of the issues that we need to sort out is what kind of quantity of used LED chips are needed for these recycling processes to make commercial sense. How many shovelfuls of LEDs do you need to chuck into the ‘cleansing fire’ for a worthwhile outcome? Are we talking alchemists’ crucibles here, or vast fiery furnaces? We can’t afford to wait until the raw materials become commercially scarce and therefore costsensitive – we need to get these processes sorted out now.

 

Let’s take a closer look at what goes on inside an LED luminaire. The Quartet downlight from High Technology Lighting, designed by my good friend Thomas Holgeth, is designed and built in the UK. It uses a Xicato LED module and sound commercial reasoning has determined that some of the components have been made overseas, though to a UK design specification. Apart from its excellent lighting performance, it’s an elegant piece of design – and you can take it apart with a set of Allen keys, which is what I’ve done. Xicato, like a number of other companies, uses ‘remote phosphor’ technology to produce its topquality illumination. It means that there is far less wastage in chip production, which suits me fine. If you were to take a hacksaw to the LED module (not recommended), this is what you would find. It’s a closed system – each component is reliant on every other component to deliver the desired performance. It all happens in a circular aluminium housing. Inside, there’s a circular chamber, topped by a phosphor disc. At the bottom of the chamber, mounted on a dinky circuit board, is a ring of LED chips (it’s not unlike a technical version of a matryoshka doll).

 

The complete downlight weighs in at 500g, but the LED module weighs just 54g (I’ve always wondered whether the LED is actually drawing its energy from a parallel universe – now I’m sure of it). And I’ve just read that 40 per cent of the energy used by LEDs is lost as heat. Let’s take a look at this phosphor disc in more detail. The light from the LED chips knocks around inside what Xicato calls the internal optical mixing cavity before hitting the ceiling of the chamber – which is the phosphor disc. Xicato’s discs are made of glass, though other transparent materials, such as plastics, are in common use. The transparent disc is coated with a phosphor mix, which determines the colour properties of the module. Remote phosphor technology (RPT) means that the ring of blue LED chips sitting in the bottom of the chamber don’t need to be so sophisticated because the heavy lifting is happening at the phosphor interface. 

 

Remote phosphor isn’t everyone’s cup of tea; questions are raised about the impact on luminous efficacy, and the stability of the phosphor and disc material over time. But I always prefer something that can be stripped down and reduced to its constituent parts. The material content of an individual LED module is tiny – one might be tempted to say insignificant (of the 54g Xicato module, most is taken up with the aluminium housing). But the manufacture of LED chips is counted in billions, and the demand for arsenic, gallium, indium, and the rare-earth elements cerium, europium, gadolinium, lanthanum, terbium and yttrium, has accelerated at an unprecedented rate – to the extent that the global reserves of some of these elements is already a concern. Organisations such as Recolight have plans for the collection and recycling of LEDs, which is good news. But I still have the sense that we’re running very quickly down a one-way street towards materials scarcity. And there’s only an empty hole in the ground at the end of it.

 

There are two chief reasons why we need to learn how to recycle LED content effectively. First, we can’t rely on the continued extraction of these metals, because they may not be there to extract, or countries that have them may impose export bans so they can hold on to them for domestic markets. Second, although the nasty stuff (arsenic, lead) tends to be embedded into the LED construction, they will, in many years’ time, eventually leach into the soil and groundwater. We’re unlikely to be around by then, but someone will. And let me float another idea here. Convincing arguments for the switch to LEDs are usually built around energy payback times and lifetime guarantees. But I’m more interested in what happens when a building refurbishment removes luminaires well before the end of their LED module’s useful life.

 

Do we really have to send all that ‘light-in-potential’ into the recycling merry-go-round? I’d like a more dynamic approach to the LED – making new use of old modules that still have, say, 50 per cent of their active life. It would be an interesting take on the Aladdin story: ‘Old lamps for new!’

 

Coming back to that Quartet downlight, let’s see what else is in there. It turns out that almost half of the 500g weight is taken up by the heatsink. Most commercial heatsinks are made of aluminium; a friendly metal which can be worked easily, recycled and reworked repeatedly. It’s reckoned that 75 percent of the aluminium produced since the 19th century is still in productive use, though with demand set to increase from 40 million tonnes to 70 million tonnes by 2020, over two-thirds of that demand will have to come from primary sources, and that will mean an increase in bauxite mining. This is not a good thing, as you will know if your garden overlooks a bauxite mine.

 

As an industry, we can be rightly proud of our record on materials recycling. But the LED is a game-changer, and we need a more interventionist procedure. I find the idea of smelting down an engineered block of aluminium after 10 or 20 years to be profligate. If heatsinks are available off-the-shelf, why do they have to be newly minted? Why can’t we simply remove the many thousands of used heatsinks that will come on-stream from around 2020 onwards and just re-use them? They may be bigger than needed, but the world won’t be any smaller and light fittings won’t need to be any smaller than they are at present. Just a thought.