RISE OF EXTERNAL QUANTUM EFFICIENCY OF LED BY ADDITIONAL TUNNEL TRANSPARENT LAYER.

Luskinovich P.N., Nanoattometria (www.nanoattometria.com)

Total efficiency of LED emission is determined by product of internal and external quantum efficiency. Internal quantum efficiency of current transformation into light is close to 100%. That’s why LED efficiency is first of all determined by efficiency of radiation coupling from LED - external quantum efficiency. That makes raising the external quantum efficiency the most important direction of development. The laws of physics don’t contradict further raise of external quantum efficiency.

The major problem of radiation coupling is significant reflection of light at the interface of light-emitting surface and lens.

The simplest way to solve this problem is to produce crystals and lenses from identical semiconductor materials, as you can see at fig.1.

Fig. 1. LED without matching layer.

Semiconductor and lens should have equal thermal coefficients of expansion.

Highly effective, but practically unrealizable because of the difference between thermal coefficients of expansion.

In this case the light comes out of crystal and lens with no internal reflection.

This type of infrared LEDs is produced from gallium arsenide, but it’s not economically feasible, because of the high price of the materials.

Such expensive materials could be replaced by cheaper ones, but in this case the material of lens and crystal should have equal or close optical index of refraction and coefficient of thermal expansion.

There are lots of materials with high optical index of refraction: chalcogenide materials, metal oxides, etc, but all this known materials have different from   LED crystal coefficient of thermal expansion. That leads to mechanical stresses at the interface of lens and crystal and destruction of LED crystal.

For the last decades, despite of multiple efforts, the desired material with similar to LED’s optical index of refraction and coefficient of thermal expansion wasn’t synthesized, so most of the developers quit trying.

That’s why by now organic materials are used to produce lenses. Such materials don’t make significant mechanical distortion to crystals but their optical index of refraction is low compare to semiconductor’s.

As a result, according to Snellius law, substantial part of luminous flux is totally reflected at the interface of crystal and lens, that reduces the external radiation power. According to the laws of geometrical optics, the radiation under total internal reflection doesn’t pass the interface that reflects it.

According to that categorical statement, further development on the basis of geometrical optics is almost futile!

Let’s take a look at LED design as a combination of 3 elements: light- extracting material (crystal) with high optical index of refraction, like on fig. 1, matching layer, made of organic materials with low index of optical refraction and a lens (fig.2).

Fig. 2. LED with matching layer thicker than the emission wavelength.

Equal thermal coefficients of expansion are not needed, but because of the low refractive index of matching layer most of the luminous flux reflects from the border of the layer. Low efficiency.

Let’s analyze this design from the points of view of ray and wave models of light propagation. By the law of geometrical optics part of the emission will be reflected at the interface of crystal. That will decrease the external quantum efficiency. Thickness of the organic material layer is measured in microns and ever millimeters. Such size is significantly more that the wavelength of optical  emission, so we can apply ray model to describe the process of light propagation.

Following that, when the light reaches the interface of the next layer with higher refractive index it will be partial reflected. It will lower the light-extracting efficiency. Upon that basis, further development of such constructions is futile.

To overcome that, we applied not ray, but wave model of light propagation. In 1855 Maxwell wrote equations that described propagation of electromagnetic waves and came to the conclusion that light is an electromagnetic wave. By the laws of electrodynamics, light propagation leads to new effects that sometimes are contrary to the old laws of geometrical optics.

For example, in geometrical optics the effect of total internal reflection is known, while electrodynamics describes the opposite effect of frustrated total internal reflection [1]. The core of the effect of frustrated total internal reflection is in penetration of light emission from one environment to another with a lower refractive index, in which the laws of geometrical optics it is not permitted to.

We demonstrate the results of the calculation of the propagation of light through a layer with a refractive index lower than the refractive index of the material layers in which it is located.

Fig. 3, 4 shows the propagation of radiation at different thicknesses of the matching layer.

Fig. 3. The propagation of radiation at layer thickness is much more than the wavelength of radiation.

When the layer is thicker that the wavelength, we can see the process of propagation of electromagnetic waves: the partial reflection, partial radiation entering in the middle layer and its recovery. The share of the transmitted wave over the reflected is negligible. This light-extracting layer with low refractive  index was another barrier in raising the effectiveness of LEDs emission.

We overcame this barrier, applying the theory of electromagnetic waves.

Fig.4. The propagation of radiation at layer thickness is much less than the wavelength of radiation.

When the layer thickness is much less than the wavelength of radiation, we can see that the transmitted wave proportion has significantly increased. It should be noted that this increases the transmission coefficients for all the incoming waves at the interface of the layer at various angles. Thus, the transmission coefficient through the tunnel-transparent layer increases when the thickness of the layer decreases. These calculations are made on the basis of Maxwell"s equations, which proved to be right describing the processes of propagation for more than 150 years. The results of the calculations showed that if we use the tunnel-transparent layer, despite its lower refractive index, the transmission coefficient through it can be quite large, up to 100%. For practical application, this means the possibility of using a transparent thin elastic layer with a low refractive index.

The main object of the elastic intermediate layer is to reduce mechanical stresses on the interface.

This requirement can be successfully carried out with a tunnel-transparent layer thickness smaller than the wavelength of radiation. That way the problem of mechanical stresses on the interface of crystal and light-extracting material was solved. We designed LEDs based on electrodynamics. Raising the effectiveness of light emission’s transmission between the materials with high refractive index through a thin layer with low refractive index is called the tunneling effect.

LED design with elastic material tunnel-transparent layer and increased refractive index material lens is shown on fig 5. LED with the applied effect of  light tunneling we called Tunneling Light Emitting Diode - TLED.

Fig. 5. Patented TLED with matching layer ensuring tunneling of light. 

Semiconductor and lens shouldn’t have equal thermal coefficients of expansion. High efficiency.

As a result we achieved to follow the basic rules of LED design: use the light- extracting materials with high refractive index and reduce mechanical stresses at the interface of dissimilar materials used.

Electromagnetic waves theory proves that the characteristics of wave propagation are affected not by the parameters of the material at the point, but by the effective parameters at the area which is commensurable with the length of wave emission. Using that knowledge, we created LED design with averaged by space in nanometric scale, effective characteristics, matching required. This means that it is possible to combine areas with different properties and achieve set averaged parameters.

In this design, light emission of crystal passes through the tunnel-transparent layer and then through the high refractive index material lens. The effectiveness of the output radiation coming from the LED can reach values close to 100%.

Type of the light-extracting device doesn’t matter for TLED, so the Weierstrass hemispheres, Frenel lenses and modern micro-and nano-optical elements can be used.

It should be pointed that several process operations in LED manufacturing are not only unnecessary, but also impair the achievement of maximum possible parameters of TLED. Roughness of light-emitting surface for raising the radiation power, for example. In fact, such roughness lower the effective refractive index in area commensurate with the length of wave emission, which reduces the efficiency of transmission of radiation while tunneling.

Placing a tunnel-transparent matching layer and it’s further connection with  lens with higher refractive index is easily implemented in the existing LEDs manufacturing processes.

TLED technology can be applied to LEDs of various designs and spectral ranges.

We created TLEDs experimental samples of infrared, red and blue spectral ranges. We used usual epitaxial structures for them.

For infrared TLEDs we took crystals with planarized surface with no roughness.

Measurement results of LED’s with usual polymer coating and high refractive index lens connected through the tunnel-transparent elastic layer with infrared crystal’s surface radiation pattern are on fig. 6.

TLED gives 40% raise in integral emission output compared to traditional LED.

Identical flip-chip LEDs with flat emitting surface before (blue) and after (red) TLED applied.

LED TLED

Integral optical power (mW) 7.366 10.863

Fig. 6. Rise of integral emission output of infrared TLED compared to traditional technology LED without increase in energy consumption.

The bottom curve in the graph shows the directional pattern of infrared LED with usual silicone coating. The upper curve is for TLED directional pattern with tunnel-transparent elastic layer and high refractive index lens. The measurements were made by opto- goniometer. Processing of measurement results by integrating the radiated power in all areas of the hemisphere shown TLED’s over 40% raise compared with LED with usual silicone coating in radiation power.

For blue TLEDs we took well-known crystals with rough surface.

Measurement results of radiation pattern of blue LED and similar LED with applied TLED are shown on fig. 7.

TLED gives 18,9% raise in integral emission output compared to traditional rough emitting surface LED.

Identical flip-chip LEDs with rough emitting surface before (blue) and after (purple) TLED applied.

LED TLED

Integral optical power (mW) 52.010 66.18

Fig.7. Rise of radiated power of blue TLED made of LED with rough emitting surface without increase in energy consumption.

The bottom curve in the graph shows well-known blue LED.

Upper curve is for TLED. Similar LED with tunnel-transparent elastic layer and high refractive index lens.

Measurement results of TLED’s radiation power shown almost 27% increase compared to LED with no silicone layer. Measurement results of emission power  of TLED compared with LED with silicone layer shows increase by 18%.

We made red and infrared LEDs and TLEDs of various design and constantly TLEDs shown raise of efficiency without increase in energy consumption.

Fig. 8. Tunneling Light Emitting Diode Design’s major advantages.

TLED’s design advantages that rise light-extracting efficiency are:

  • Use of materials with high refractive index
  • Rise of optical radiation transmission from semiconductor to lens with  higher refractive index
  • Compensation of denary difference between semiconductor’s and lens’s thermal expansion coefficient
  • Optical tunneling effect application.

Fig. 9. Rise of efficiency technology’s major advantages.

TLED’s technology’s advantages are:

Adaptivity. The technology is easily implemented in any production string.

Universality. The technology is applied for raising diodes efficiency in various spectral ranges.

Economy. No major costs on modernization of production needed. We hope that our technology will find use in LED manufacturing. We are ready to demonstrate the efficiency of TLED on any crystals.


1. Zhu S., A. W. Yu, D. Hawley, and R. Roy, “Frustrated total internal reflection: A demonstration and review,” American Journal of Physics. 54 (7), 601-606 (1986).