LED is made of III-V compounds, such as GaAs (gallium arsenide), GaP (gallium phosphide), GaAsP (gallium arsenide phosphorous) and other semiconductors, and its core is a PN junction. Therefore, it has the I-V characteristics of a general PN junction, that is, forward conduction, reverse cutoff, and breakdown characteristics. In addition, under certain conditions, it also has light-emitting properties. Under the forward voltage, electrons are injected into the P zone from the N zone, and holes are injected into the N zone from the P zone. Part of the few carriers (minor carriers) that enters the opponent’s area recombines with the majority carriers (multiple carriers) to emit light, as shown in Figure 1.
Assuming that light emission occurs in the P region, the injected electrons and valence band holes directly recombine and emit light, or they are first captured by the luminescent center and then recombine with the holes to emit light. In addition to this recombination luminescence, some electrons are captured by the non-luminescent center (the center is near the middle of the conduction band and the medial band), and then recombine with the holes. The energy released each time is not large, and visible light cannot be formed. The greater the ratio of the luminous recombination to the non-luminous recombination, the higher the light quantum efficiency. Since the recombination emits light in the minority carrier diffusion region, the light is only generated within a few μm close to the PN junction surface. Theory and practice have proved that the wavelength or frequency of light emission depends on the energy gap Eg of the selected semiconductor material, and the unit of Eg is electron volt (eV)
Eg＝hv/q＝hc/(λq) (1-1)
λ＝hc/(qEg)＝1240/Eg (nm) (1-2)
Where: v is the velocity of electrons; h is Planck’s constant; q is the charge carried by the carriers; c is the speed of light; λ is the wavelength of light emission.
If it can produce visible light (wavelength of 380nm violet light ~ 780nm red light), the Eg of the semiconductor material should be between 3.26~1.63eV. The PN junction constitutes a certain potential according to its terminal voltage. When a forward bias is applied, the potential barrier drops, and the majority carriers in the P and N regions diffuse to each other. Since the electron mobility μ is much larger than the hole mobility, a large number of electrons diffuse to the P region, which constitutes the injection of minority carriers in the P region. These electrons recombine with the holes in the valence band, and the energy obtained during recombination is released in the form of light energy. This is the principle of PN junction luminescence.
Semiconductors can be divided into direct band gaps and indirect band gaps. Most LEDs use direct band gap materials, so that electrons can directly transition from the conduction band to the valence band and recombine with holes to emit light, with high efficiency. On the contrary, the use of indirect band gap materials, its efficiency is lower.
The output spectrum of LED determines its luminous color and light radiation purity, and also reflects the characteristics of semiconductor materials.
The PN junction has different height barriers for electrons and holes, both of which are very small, but the potential barrier of holes is much smaller than that of electrons, and the holes continue to diffuse from the P area to the N area, resulting in high injection efficiency. The rate at which electrons in the N region are injected into the P region is relatively small. In this way, the electrons in the N region transition to the valence band to recombine with the injected holes, and emit radiation determined by the energy gap of the N-type semiconductor. Due to the large energy gap of the P zone, light radiation cannot be sent to the conduction band, so no light absorption occurs, which can directly transmit outside the LED, reducing the loss of light energy.
The forward volt-ampere characteristic curve of the LED is relatively steep, and almost no current flows through the LED before the forward conduction. When the voltage exceeds the turn-on voltage, the current rises sharply. Therefore, LED is a current-controlled semiconductor device, and its luminous brightness L (unit cd/m2) is approximately proportional to the forward current IF
In the formula: K is the proportional coefficient.
In the small current range (IF=1~10mA), m=1.3~1.5, when IF＞10mA, m=1, the above formula can be simplified to
L＝KIF (1-4)
That is, the brightness of the LED is proportional to the forward current, and the forward voltage and forward current of the LED are related to the semiconductor material of the die. When using, you should select the appropriate IP value according to the required display brightness to ensure that the brightness is moderate, and the LED will not be damaged. If the current is too large, it will burn the PN junction of the LED.
For LEDs, lattice matching is a major issue, because for most III-V semiconductors, there is no suitable substrate that can support the upper heteroepitaxy. The crystal lattice size of the grown barrier crystal layer must match the crystal lattice of the substrate so as not to cause lattice defects due to stress. Otherwise, the photons emitted by the module will be absorbed by the lattice defects, which will greatly reduce the luminous efficiency of the module. The earliest heterogeneous barrier crystals of group III-V semiconductors used GaAs as a substrate, and a GaAlAs barrier crystal layer was grown on it. Because the crystal lattices of these two materials are very similar, the stress between the barrier layer and the substrate is extremely small, so the development process can be carried out smoothly. However, the barrier crystals that have been developed one after another are grown on GaAs substrates or GaAsxP1-x are grown on GaP substrates, both of which have problems due to stress. Therefore, in optoelectronic materials, the method of adjusting the ratio of binary, ternary or even quaternary materials is often used to adjust the ratio of multi-element atoms of different sizes to match the crystal lattice structure of the substrate. This method can also adjust the energy gap of the semiconductor, thereby adjusting the wavelength of light emitted by the light-emitting module. Using such a method is complicated in adjusting the parameters of the barrier crystal. Therefore, it can be seen that the barrier crystal technology can be called the core of the semiconductor light-emitting module technology.
As the growth method of the barrier crystal is improved, the structure of the barrier crystal is also continuously improved. The earliest structure is the traditional surface structure PN junction, but its luminous efficiency cannot be significantly improved. Therefore, the method of using a single heterojunction surface structure has begun to be used in the full-crystal manufacturing process, which can improve the minority carrier injection efficiency in the PN junction, thereby significantly improving the luminous efficiency. Later, it developed into a double heterojunction structure. The energy gap of the material on both sides of this structure is higher than that in the middle, so the carriers on both sides can be injected into the intermediate layer very effectively and these carriers are completely trapped in this range, resulting in very high photoelectric conversion efficiency. The latest method is to use a quantum structure in the barrier layer. When the thickness of the intermediate layer of the double heterojunction structure is gradually reduced to a few 10 angstroms (
), the electrons or holes produce quantum effects, which can greatly improve the efficiency of photoelectric conversion.
The barrier crystal technology is mainly aimed at the GaAs series LEDs of III-V materials whose emission wavelengths are concentrated in the red and yellow wavelength bands. If you want to obtain a full-color semiconductor light source, you must develop LEDs in the blue and green wavelength bands, and the GaN series of LEDs have also made significant progress in recent years under such demand. The materials used in blue and green LEDs were mainly ZnSe and GaN materials in the early days. Because ZnSe materials have reliability problems, GaN materials have more room for development. In the early research on GaN, because it has been unable to find a substrate that matches the lattice constant of GaN, the defect density in the barrier crystal is too high, and therefore the luminous efficiency cannot be improved. Another reason for the failure of GaN to achieve breakthroughs lies in the growth of the P-GaN part of the module. Not only is the doping of P-GaN too low, but its hole mobility is also low.
Until 1983, S. Yoshida and others in Japan used high-temperature growth aluminum nitride (AIN) as a buffer layer on a sapphire substrate, and then grew GaN to obtain better crystals. Later, Professor I. Akasaki of Nagoya University and others used MOCVD to grow the AN buffer layer at a low temperature (600°C) to obtain a mirror-like GaN after high temperature growth. In 1991, Nichia Co. researchers used low-temperature growth of an amorphous buffer layer of GaN, and then grown at high temperature to obtain the same mirror-like GaN. In 1989, Professor Yu Akasaki used electron beams to irradiate magnesium (Mg)-doped P-GaN to obtain P-type GaN. In the same year, Nichia’s Shuji Nakamura directly used 700°C thermal annealing to complete the production of P-type GaN, which made breakthroughs in two major issues that plagued the development of GaN.