The case for nanoceramic PCBs in UV LEDs
Advances in technology have seen the rise of UV LEDs as an alternative to HIDs. LEDs use less power, run considerably cooler and operate at a specific wavelength and power cycle instantly. These characteristics are transforming the UV curing industry, from driving efficiency to offering new solutions such as hand-held curing applications that would not have been conceivable with HID lamps.
Yet LEDs are still relatively inefficient. Only around 40% of the power that goes into a UVA LED is converted into light with the remaining 60% converted into waste heat. This heat needs removing from the LED as quickly as possible to prevent the LED from overheating and exceeding its maximum operating temperature. Failure to remove this heat can lead to the light quality deteriorating and ultimately to catastrophic failure of the LED.
Unlike HID lamps, the small surface area and relatively low temperature of LEDs mean the amount of heat they can convect and radiate away is negligible. The only way to remove the heat is to conduct it out the back of the die, through the PCB, to a heatsink and to the ambient atmosphere. To keep the LED die under its maximum operating temperature, the substrate material used for the PCB needs to be highly thermally efficient. With UVA applications, the module substrate tends to be either a thermally effective metal-clad PCB (MCPCB) or electronics-grade ceramics, particularly high-performance aluminium nitride (AlN). In higher-power density applications epoxy-based MCPCBs simply don’t have the requisite thermal performance (<100W/mK vs 170W/mK for AlN) making AlN the substrate of choice where heat is a significant issue.
However, aluminium nitride is not without its challenges. The material itself is expensive and tricky to manufacturer into circuits because of its brittleness. Aluminium nitride comes in a 4×4-inch tiles (occasionally 7×5-inch). Any larger and the brittleness of the material makes for such a low yield as to be untenable. With 4×4-inch tiles, a yield loss of 20% is not uncommon. The other issue with brittleness is when mounting the finished module onto its heatsink. In an ideal world, the circuit should be attached as firmly as possible to reduce any air gaps between the circuit and the heatsink. Screwing a brittle aluminium nitride module to a heatsink is likely to fracture it if too much pressure is applied.
Clearly an alternative material that has a performance comparable to AlN but the mechanical properties of a MCPCB would be ideal. That alternative is nanoceramics.
The nanoceramic approach
Using a patented electro-chemical oxidation (ECO) process, Cambridge Nanotherm transforms the surface of an aluminium board into a thin layer of alumina (Al2O3) ceramic measuring just tens of microns thick. This alumina layer acts as the dielectric between the circuit above and the aluminium below. The thinness of the layer more than makes up for alumina’s relatively low thermal efficiency compared to aluminium nitride, and allows heat to be conducted extremely effectively.
Thin-film processing follows the ECO process to sputter the copper circuit layer directly to the nanoceramic dielectric to further improve the thermal efficiency of the stack. Nanotherm DMS (a direct replacement for AlN for UV modules) has a composite thermal performance of 152W/mK, slightly lower then high-grade AlN but more than enough to cope with all but the most demanding UVA applications.
Because nanoceramics are essentially aluminium PCBs, mechanically they can be treated in the same fashion as a standard MCPCB, there is no loss of yield due to brittleness and they can be screw mounted to a heatsink without any risk of breakage. Ultimately nanoceramics offer the best of both worlds, the thermal performance of AlN and the robust characteristics of an aluminium MCPCB.
More than just UVA
UVA isn’t the only area where the unique properties of nanoceramics can be used. It can also be used in UVC, where the unique germicidal attributes of UVC radiation give it the ability to kill bacteria by disrupting its DNA preventing reproduction — making UVC ideal to sterilise and disinfect water, air and surfaces from a rage of bacteria and pathogens.
As with UVA, HID lamps, albeit without any phosphor coating, have traditionally been used for UVC. Mercury lamps are extremely effective for large-scale applications such as municipal water disinfection where their slow warm-up times and relatively high cost are irrelevant. In these applications, the potentially toxic bulbs can be tightly controlled and replaced quickly if damaged before the mercury can cause harm. HID lamps have, again much like with the UVA market, limited the range of possible applications.
Now, LED technology is creating a whole range of new markets in small, portable disinfection devices, thanks to the compact size, robustness and low power of LEDs over HIDs. Products like portable disinfection ‘wands’ are coming to market, which sterilise everyday items like smartphones, tablets and keyboards. Water disinfection devices that connect to the pipes in your water supply to sterilise groundwater and the like are also on the market.
Moreover, consumer goods manufacturers can now embed LED UVC technology into their products to turn them into self-disinfecting items. For example, a toothbrush could disinfect itself after you put it back into its holder, a baby’s milk bottle could self-sterilise, and a bottle could sterilise water at the push of a button. The possibilities are virtually endless and could have profound implications for public health.
The growth potential for the UVC industry is huge with industry analysts Yole Développment predicting meteoric growth from 2018. But the thermal challenges of UVC LEDs are substantially higher than UVA and is a challenge the industry will have to deal with these to ensure the predictions of the analysts come to fruition.
Around 95% of the power of put into a UVC LED is converted to heat meaning ony 5% of the power is converted to useful light because of the short wavelength at which UVC operates. While UVA operates at 380nm, UVC disinfection devices operate at a wavelength of 260nm — the ‘germicidal range’. The challenge is that as the wavelength gets shorter, the external quantum efficiency (EQE) of the LED — which is made up of three factors: light extraction efficiency (LEE), electron injection efficiency (EIE), and internal quantum efficiency (IQE) — plummets. At a wavelength of 380nm (UVA) the EQE is, at best, 40%. Continue up to 260nm (UVC) and the EQE drops to just a few percent, which means that for UVC LEDs, around 95% of the power put into an LED is converted to waste heat, which needs to be removed quickly.
There are significant developments underway in all these areas to reduce EQE. These range from using bulk-AlN wafers as the base of the epitaxy process because it has a lower defect density, which increases IQE. Bulk AlN is also able to transmit light below 280nm, potentially opening up new applications. Projects around quantum tunnelling are exploring ways to cut the resistance and improve the electrical efficiency of die aim to improve EIE, whilst different chip geometries are being used to improve LEE.
While these projects may offer significant breakthroughs at the time of writing, they are very much R&D. The fact is that, at the moment, shorter-wavelength UV applications are incredibly inefficient and generate a huge percentage of heat. To get a suitable output requires putting a lot of power in. This produces excess heat that needs to be removed from the LED to keep the junction under its maximum operating temperature.
With UVC LEDs, AlN has traditionally been the only option. MCPCBs simply do not have the required thermal performance, and they are degraded by UVC light because of the organic epoxy used to create the dielectric layer. Nanoceramics present the perfect option — thermally effective, robust, inorganic and cost effective. And if the thermal management issues can be addressed by a material that is mechanically robust enough to withstand the day-to-day use of a portable consumer product, the possible applications are huge.
Thermally managed LEDs: the key to UV success
industry.
LED technology is having a profound impact on industries enabled by UV. The printing industry has seen costs drop and efficiencies increase as manufacturers have shifted to LEDs. UVC LED applications promise something rather more profound, nothing short of a revolution in how we disinfect and sterilise our environment. Yet until the heat issue is dealt with, thermal management will remain a central issue. The unique qualities of nanoceramic can have a big impact in realising the ambition of UVC LED manufacturers to transform what is one of the most fundamental aspects of daily life.
About the author:
John Cafferkey is Marketing Manager at Cambridge Nanotherm – www.camnano.com
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
