Recently, SMD aging, found a problem, the patch's light decay is very small (within 1000h), or can be said to be zero light decay, what is the reason for it? I found the answer through the spectrum, is the chip problem, due to SMD heat dissipation is better than LAMP, so the chip has no light decay (within 1000h), but the brightness will be added, so the brightness of the phosphor will increase, even if the powder has light decay (the conversion efficiency of the phosphor is reduced) And the light fades, and the chip's own light decay is very serious (due to the junction temperature rise), then the brightness of the phosphor will decrease, so the patch and the in-line light decay size are compared, and the glue is removed. The factor is mainly the reason of the chip, and the light decay of the phosphor can not be simply expressed by the LED light decay after the package, so people who praise the light fade of the phosphor should pay attention to it, look at your test. Is there a loophole unless the brightness of the chip you are using is constant, otherwise the phosphor will be affected.
The main reason for the rapid attenuation of white LED devices is the attenuation of blue LED chips or the attenuation of phosphors. At present, the opinions of LED industry insiders are not consistent. Some people use the same sapphire substrate GaN-based blue LED chip, bracket, epoxy resin packaging material to make φ5mm bracket type blue LED and white LED, and then compare the luminous flux attenuation of the two sets of devices after long-term operation. In the experiment, the lighting current is 20mA, the test current is also 20mA, and the ambient temperature and test equipment are the same. As a result, the white LEDs decay faster than the blue LEDs (as shown in Figure 1). Since the difference between the two devices is that the phosphor is added to the package of the white LED, it is natural to think whether the attenuation of the white LED is accelerated due to the attenuation of the phosphor.
Figure 1 Comparison of white LED and blue LED attenuation
We all know that if the phosphor is attenuated, the yellow-green light with a peak wavelength of 570 nm excited by the phosphor will inevitably decay, resulting in the change of the chromaticity coordinate of the white LED to the blue tone, that is, the chromaticity coordinate value is reduced. After testing, the actual results show that the chromaticity coordinates of the white LED device change in the direction of the yellow tone. Therefore, it can be considered that the phosphor is not a main factor causing the acceleration of white light attenuation. Those who hold this view believe that the difference between blue and white LEDs is that the wavelength of the light is shorter than that of ordinary LEDs. The epoxy resin oxidizes after absorbing short-wavelength light radiation, which in turn forms a chromophore. In addition, the epoxy resin is also changed by heat, which is called "yellowing", resulting in a decrease in light transmittance at a short wavelength. This phenomenon does not affect the red LED, but it has a great influence on the blue and white LEDs. This is an important reason for the fast decay of blue and white LEDs.
If a thin phosphor is applied to the chip of the white LED, and then the stroke of the light of the two devices is compared and analyzed, it is found that the light of the blue LED is directly emitted through the epoxy, and the white LED device exists due to the surface. In the phosphor layer, a part of the light is directly emitted, and a part of the light is directed to the phosphor particles. In addition to converting part of the blue light into yellow light, the phosphor particles have isotropic scattering due to the light, so that the short-wavelength light radiation and heat near the phosphor layer are highly concentrated, so that the epoxy resin near the phosphor layer is easier. A "yellow change" occurs. This will also have an effect on the attenuation of white LEDs, as shown in Figure 2.
Figure 2 Effect of isotropic scattering of light
Some people have done this experiment: use the same blue chip and different phosphors. The peak wavelength of the excitation spectrum of the selected phosphor is 460±5 nm. The emission wavelengths are 540 nm, 550 nm and 560 nm due to different rare earth doping. The luminous flux of three different emission wavelengths as a function of time is shown in Fig. 3.
Figure 3. Variation of luminous flux over three different emission wavelengths over time
It can be seen from Fig. 3 that the attenuation is relatively small when the emission wavelength is 560 nm. It can be concluded that the same type of blue chip excites phosphors of different impurities, which will cause different yellow wavelengths to be emitted, so the mixed white light also has different attenuation. This shows that phosphor attenuation is the main factor.
Others choose the same chip and the same phosphor (the glue and concentration are the same), and the thickness of the phosphor coated on the blue chip is different. In this way, it was found that the thicker phosphors were lighter than the thinner phosphors, as shown in Figure 4.
Figure 4: The thickness of the phosphor coating of the chip is different, and its luminous flux changes with time.
In summary, the most important reason for LED light decay is the chip. Only by fundamentally solving it can the LED light decay be reduced. Of course, some people will say that phosphors also account for a large part of LED light decay. In this way, different phosphors have different light decays, such as the yellow light and red powder. Many of the current market is very light decay or zero light decay, and the green light's light decay is slightly larger. However, in addition to the slower brightness improvement, it has been greatly improved in controlling the light decay. Therefore, it is expected that the chip technology should consider the light attenuation problem while improving the brightness. There are many reasons for the attenuation of white LEDs. It should be analyzed according to the actual situation, and then confirmed by experiments.
