TWI549317B - Light emitting diode - Google Patents

Description

Light-emitting diode

The present invention relates to a light emitting diode, and more particularly to a light emitting diode (LED) which can improve luminous efficiency.

The light-emitting diode is a semiconductor element mainly composed of a III-V element compound semiconductor material. Because such a semiconductor material has the property of converting electrical energy into light, when a current is applied to the semiconductor material, the electrons inside it combine with the hole, and the excess energy is released in the form of light to achieve luminescence. effect.

In general, there is a problem of mismatch between the lattice constant of gallium nitride as the material of the epitaxial layer in the light-emitting diode and the lattice constant of the sapphire substrate, and the degree of lattice constant mismatch is about 16%. , causing a large number of defects to occur in the junction of the lattice growth, which in turn leads to a large attenuation of the luminous intensity. Although in the light-emitting diode, in the process of crystal growth of gallium nitride, it inevitably has certain defects. However, when the wavelength of the light emitted by the light-emitting diode is 450 nm, since the conventional lattice stress is released near the defect to form an indium self-polymerization region, the carrier easily enters the region of indium self-polymerization before moving to the defect. The so-called localized effect. Since the self-polymerization region of indium has quantum confinement effect, the recombination efficiency of the carrier can be improved. Therefore, even if the gallium nitride light-emitting diode has a high defect density in the active region due to the limitation of the crystal growth process, the wavelength is 450 nm. It is still possible to maintain a certain degree of luminous efficiency.

However, when the light-emitting band of the light-emitting diode is gradually shifted from the blue light to the ultraviolet light band, since the indium content in the active layer is gradually reduced, the formation region of the indium self-polymerization is relatively small, resulting in the light-emitting diode. The carrier is easily moved to the defect to produce non-radiative recombination, which causes the luminous efficiency of the LED to be greatly reduced in near-ultraviolet light. Furthermore, the polarization field effect of the nitride semiconductor itself causes the energy band of the active layer to bend, and the electron beam The hole pairs are not easily confined within the quantum well layer and thus cannot effectively radiate recombination. In addition, electrons are more likely to overflow to the P-type semiconductor layer, resulting in a decrease in luminous intensity. Furthermore, since the mobility of the hole is smaller than the mobility of the electron, when the hole is injected from the P-type semiconductor layer to the active layer, Most of the holes are confined to the quantum well layer closest to the P-type semiconductor layer, and it is difficult to uniformly distribute all of the quantum well layers, resulting in a decrease in luminous intensity. Therefore, the industry has been developing a light-emitting diode having high luminous intensity.

The invention provides a light-emitting diode which can improve the light-emitting diode in the 222 nm to 405 nm light-emitting band by making the number of layers of the quantum barrier layer doped with the N-type dopant in the quantum barrier layer conform to a specific ratio. Luminous efficiency.

The invention proposes another light-emitting diode which can improve the light-emitting diode at 222 nm by making the closest doping concentration of the quantum barrier layer doped with the N-type dopant to the P-type semiconductor. Luminous efficiency of the 405 nm luminescence band.

The invention further provides a light-emitting diode which can improve the light emission of the light-emitting diode in the 222 nm to 405 nm light-emitting band by satisfying a specific relationship of the doping concentration of the quantum barrier layer doped with the N-type dopant. effectiveness.

The invention provides a light emitting diode comprising a substrate, an N-type semiconductor layer, an active layer, a P-type semiconductor layer, a first electrode and a second electrode. The N-type semiconductor layer is on the substrate. The active layer has an active region having a defect density of DD, where DD ≧ 2 x 10 ⁷ /cm ³ . Located on a partial region of the N-type semiconductor layer, the active layer emits a light wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well, each quantum well Between any two layers of quantum barrier layers, and i is a natural number greater than or equal to 2, wherein the doped N-type dopant is in at least the k layer in the quantum barrier layer, and k is a natural number greater than or equal to 1, when i is When even, k≧i/2, when i is an odd number, k≧(i-1)/2. The P-type semiconductor layer is on the active layer. The first electrode is located on a partial region of the N-type semiconductor layer, and the second electrode is located on a partial region of the P semiconductor layer.

The present invention provides another light emitting diode comprising a substrate, an N-type semiconductor layer, an active layer, a P-type semiconductor layer, a first electrode, and a second electrode. The N-type semiconductor layer is on the substrate. The active layer has an active region having a defect density of DD, where DD ≧ 2 x 10 ⁷ /cm ³ . The active layer is located on a portion of the N-type semiconductor layer and emits light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm. The active layer includes an i-layer quantum barrier layer and (i-1) a layer of quantum wells. Between the two quantum barrier layers, and i is a natural number greater than or equal to 2, wherein the doped N-type dopant is in at least the k layer in the quantum barrier layer, k is a natural number greater than or equal to 1, when i is an even number When k≧i/2, when i is an odd number, k≧(i-1)/2. The P-type semiconductor layer is on the active layer, and the doping concentration of the quantum barrier layer closest to the P-type semiconductor in the k-layer quantum barrier layer is less than or equal to the doping concentration of the other quantum barrier layers in the k-layer quantum barrier layer. The first electrode is located on a partial region of the N-type semiconductor layer, and the second electrode is located on a partial region of the P semiconductor layer.

The invention further provides a light emitting diode comprising a substrate, an N-type semiconductor layer, an active layer, a P-type semiconductor layer, a first electrode and a second electrode. The active zone has a defect density DD of DD ≧ 2 x 10 ⁷ /cm ³ . The N-type semiconductor layer is on the substrate. An active layer is located on a portion of the N-type semiconductor layer, the active layer emits a light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well. Each quantum well is between any two layers of quantum barrier layers, and i is a natural number greater than or equal to 2, wherein the doped N-type dopant is at least k layers in the quantum barrier layer, and k is a natural number greater than or equal to When i is an even number, k≧i/2, when i is an odd number, the doping concentration of the k≧(i-1)/2,k layer quantum barrier layer is 5x10 ¹⁷ /cm ³ to 1x10 ¹⁹ /cm ³ . The P-type semiconductor layer is on the active layer. a first electrode and a second electrode, wherein the first electrode is located on a partial region of the N-type semiconductor layer, and the second electrode is located on a partial region of the P semiconductor layer.

Based on the above, in the light-emitting diode of the present invention, the number of layers of the quantum barrier layer doped with the N-type dopant in the active layer is made to conform to a specific relationship, or the active layer is doped with an N-type dopant. The closest to the P-type semiconductor in the quantum barrier layer has the smallest doping concentration, or the doping concentration of the quantum barrier layer doped with the N-type dopant satisfies a specific relationship, so that the N-type dopant can be tempered The effect of the flat defect on the carrier enhances the recombination efficiency of the carrier of the light-emitting diode. Therefore, the light-emitting diode of the present invention can greatly enhance the light-emitting diode at 222 nm to 405 nm by any of the above techniques. Luminous efficiency.

The invention provides a light-emitting diode by being closest to the P-type half In the three-layer quantum barrier layer of the conductor layer, the quantum barrier layer closest to the P-type semiconductor layer is larger than the thickness of the remaining two layers of the quantum barrier layer, and the electron hole pair is uniformly distributed in the quantum barrier layer of the active layer. In this way, the luminous intensity of the light-emitting diode in the 222 nm to 405 nm light-emitting band can be improved.

The present invention proposes another light-emitting diode which can uniformly distribute the electron hole pair in the quantum barrier layer of the active layer by making the thickness of the three-layer quantum barrier layer closest to the P-type semiconductor layer conform to a specific relationship. In this way, the luminous intensity of the light-emitting diode in the 222 nm to 405 nm light-emitting band can be improved.

The invention provides a light emitting diode comprising a substrate, an N-type semiconductor layer and a P-type semiconductor layer, an active layer, and a first electrode and a second electrode. The N-type semiconductor layer is between the substrate and the P-type semiconductor layer. The active layer is located between the N-type semiconductor layer and the P-type semiconductor layer, and the active layer emits light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and (i-1) layer quantum well Layer, each quantum well layer is between any two layers of quantum barrier layers, and i is a natural number greater than or equal to 2. The thickness of each quantum barrier layer in the i layer is sequentially T ₁ and T from the P-type semiconductor side. ₂ , T ₃ ... T _i , wherein T _{1 is} greater than T ₂ and T ₃ or T ₁ > T ₂ = T ₃ . The first electrode is located on a partial region of the N-type semiconductor layer, and the second electrode is located on a partial region of the P semiconductor layer.

The present invention provides another light emitting diode comprising a substrate, an N-type semiconductor layer and a P-type semiconductor layer, an active layer, and a first electrode and a second electrode. The N-type semiconductor layer is between the substrate and the P-type semiconductor layer. The active layer is located between the N-type semiconductor layer and the P-type semiconductor layer, and the active layer emits light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and (i-1) layer quantum well Layer, each quantum well layer is between any two layers of quantum barrier layers, and i is a natural number greater than or equal to 2. The thickness of each quantum barrier layer in the i layer is sequentially T ₁ and T from the P-type semiconductor side. ₂ , T ₃ ... T _i , wherein the thickness of the i-layer quantum barrier layer satisfies: T ₁ = T ₂ > T ₃ or T ₁ > T ₂ > T ₃ . . The first electrode and the second electrode are respectively located on a partial region of the N-type semiconductor layer and a partial region of the P semiconductor layer.

Based on the above, in the light-emitting diode of the present invention, the quantum barrier layer closest to the P-type semiconductor layer is larger than the remaining two quantum barrier layers by the third quantum barrier layer closest to the P-type semiconductor layer. Thickness, or by making the thickness of the quantum barrier layer in the active layer conform to a specific relationship, by which any of the above technical means can increase the uniform distribution of the electron hole in the active layer and increase the probability of the electron hole to the composite, thus The light-emitting diode of the invention can greatly improve the luminous intensity of the light-emitting diode at 222 nm to 405 nm by any of the above techniques.

The above described features and advantages of the present invention will be more apparent from the following description.

1 is a schematic cross-sectional view of a light emitting diode according to an embodiment of the present invention.

Referring to FIG. 1 , the light emitting diode 200 includes a substrate 210 , an N-type semiconductor layer 220 , an active layer 230 , a P-type semiconductor layer 240 , and a first electrode 250 and a second electrode 260 , and the substrate 210 is, for example, a sapphire substrate. Specifically, nitrides are sequentially formed on one surface of the sapphire substrate 210. A stack of a semiconductor cladding layer 212 (for example, undoped gallium nitride), an N-type semiconductor layer 220, an active layer 230, and a P-type semiconductor layer 240, the active layer 230 being located in the N-type semiconductor layer 220 and the P-type semiconductor layer Between 240, the N-type semiconductor layer 220 may include a stack of a first N-type doped gallium nitride layer 222 and a second N-type doped gallium nitride layer 224 sequentially on the nitride semiconductor cap layer 212. The P-type semiconductor layer 240 may include a stack of a first P-type doped gallium nitride layer 242 and a second P-type doped gallium nitride layer 244 sequentially disposed on the active layer 230, wherein the first N-type doped nitrogen The difference between the gallium GaN layer 222 and the second N-type doped gallium nitride layer 224, or between the first P-type doped gallium nitride layer 242 and the second P-type doped gallium nitride layer 244 may be a thickness Different or different doping concentrations. In addition, the material of the N-type semiconductor layer 220 and the P-type semiconductor layer 240 is, for example, aluminum gallium nitride. The person skilled in the art can select the grown nitride semiconductor cladding layer 212 and the first N/P according to actual needs. The thickness, doping concentration and aluminum content of the doped gallium nitride layers 222 and 242 and the second N/P type doped gallium nitride layers 224 and 244 are not limited thereto.

In detail, as shown in FIG. 1, a nitride semiconductor cladding layer 212 (for example, un-doped gallium nitride) is sequentially formed on the sapphire substrate 210, and the first N-type doping nitride is formed. a gallium layer 222 and a second N-type doped gallium nitride layer 224, an active layer 230, a first P-type doped aluminum gallium nitride layer 242, and a second P-type doped gallium nitride layer 244, respectively A first electrode 250 and a second electrode 260 are formed on a partial region of the surface of the two N-type doped gallium nitride layer 224 and the second P-type doped gallium nitride layer 244, so that the first electrode 250 is electrically connected to the N-type. The semiconductor layer 220 and electrically elects the second electrode 260 The P-type semiconductor layer 240 is connected. Of course, a layer of nitride buffer layer may be added between the sapphire substrate and the N-type semiconductor, and the invention is not limited thereto.

The configuration of the active layer 230 is, for example, as shown in FIGS. 2A and 2B, which may be a single quantum well active layer 230A or a multiple quantum well active layer 230B. 2A is a schematic cross-sectional view showing a single quantum well active layer in a light-emitting diode according to an embodiment of the present invention, and FIG. 2B is a cross-sectional view showing a multiple quantum well active layer in the light-emitting diode according to an embodiment of the invention. In general, the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well layer, and each quantum well layer is sandwiched between any two quantum barrier layers, so i is a natural ratio of 2 or more. number. For example, as shown in FIG. 2A, a single quantum well active layer 230A may be composed of two quantum barrier layers 232 and a quantum well layer 234 sandwiched therebetween to form a quantum barrier layer 232/quantum well 234/quantum barrier layer. 232. Taking the light-emitting diode 200 of the 222 nm to 405 nm light-emitting band as an example, the material of the quantum well 234 is, for example, Al _m In _n Ga _1-mn N, where 0 m<1,0 n 0.5, m+n 1, and x>m, n≧y, and the material of the quantum barrier layer 232 is, for example, Al _x In _y Ga _1-xy N, where 0 x 1,0 y 0.3, and x+y 1. Those skilled in the art can select the content of m, n, or x, y to be grown for actual needs such as different light-emitting bands, and the present invention is not limited thereto.

In addition, the configuration of the active layer may also be the type of the multiple quantum well active layer 230B as shown in FIG. 2B. As shown in FIG. 2B, the multiple quantum well active layer 230B may be composed of at least two pairs of quantum barrier layers 232 and quantum wells 234, such as three pairs of quantum barrier layers 232/quantum as depicted in FIG. 2B. Well 234 repeated laminate.

It is to be noted that the LED diode 200 of the present invention changes the doping of the quantum barrier layer 232 in the quantum barrier layer 232 by performing an N-type dopant doping process on the quantum barrier layer 232 in the active layer 230. The number of layers, the doping concentration, and the doping concentration distribution in the different doped quantum barrier layers 232 are used to improve the luminous efficiency of the light-emitting diode 200 in the 222 nm to 405 nm band. Specifically, although the gallium nitride growth techniques due to limitations in the manufacturing process and there is some defect density, but even if the light-emitting diode 200 in the active layer 230 of the present level of ^107, the sub-barrier layer by adjusting variable The number of layers, doping concentration, and the like of the doped quantum barrier layer 232 in 232 can reduce the influence of the defect density of the active region on the carrier by intentionally doping the N-type dopant, thereby effectively improving Luminous efficiency. In particular, in particular, the active layer 230 emits light having a wavelength range of 222 nm to 405 nm.

The efficacy of the light-emitting diode 200 of the present invention proposed by the inventors will be explained below with experimental results. In the following examples, the enthalpy is used as the N-type dopant as the implementation range, but those skilled in the art may also use other elements in the IVA family of the same family to replace the hydrazine in the embodiment, or may be optional. Instead of the ruthenium in the embodiment, such as oxygen, V or a group of VIA, as long as it can be substituted for the aluminum, indium or gallium elements of group IIIA, electrons can be provided as N-type doping, and the present invention can also be realized.

3 is an enlarged schematic cross-sectional view of an active layer of a light emitting diode. As shown in FIG. 3, the active layer 230 of the present embodiment includes six quantum barrier layers and five quantum well layers, and each quantum well layer is sandwiched between any two quantum barrier layers. The quantum barrier layer is sequentially 232a, 232b, 232c, 232d, 232e, and 232f from the N-type semiconductor side, and the quantum well layer is sequentially quantum wells 234a, 234b, 234c, and 234d from the N-type semiconductor side. 234e.

4A is an optical simulation diagram of a comparative example of the light-emitting diode of the present invention, and FIG. 4B is an optical simulation diagram of the light-emitting diode of the present invention, wherein the defect density in FIGS. 4A and 4B is set to 1×10 ⁸ /cm. ³ . Please refer to FIG. 4A first. FIG. 4A is a diagram showing the relationship between the number of layers of the doped quantum barrier layer of the quantum barrier layer 232a-232f and the emission intensity of the wavelength band near the wavelength of 450 nm in the light-emitting diode. 3 and FIG. 4A, the horizontal axis represents the illuminating wavelength (unit: nanometer), and the vertical axis represents the illuminating intensity (unit: au), and the numbers before and after the slanting lines in the different line segments A, B, C, and D represent respectively as shown in FIG. The number of layers of the doped quantum barrier layer and the undoped quantum barrier layer is present in the quantum barrier layers 232a-232f, and the number of layers doped is from the side of the N-type semiconductor layer 220. For example, 6/0 in line A represents the total doping of the six-layer quantum barrier layers 232a-232f, and 4/2 in the line segment B represents the four-layer quantum barrier layer 232a-232d adjacent to the side of the N-type semiconductor layer 220. The quantum quantum barrier layer has two layers of undoped quantum barrier layers 232e-232f, and 2/4 of the line segments C represent two layers of quantum barrier layers 232a-232b adjacent to the N-type semiconductor layer 220 side. The quantum barrier layer is doped, and the number of layers of the undoped quantum barrier layer 232c-232f is four layers, and 0/6 of the line segment D represents that the six-layer quantum barrier layers 232a-232f are all undoped. As shown in FIG. 4A, the results show that increasing the number of layers of the doped quantum barrier layer reduces the luminous efficiency of the light-emitting diode in the vicinity of 450 nm.

In contrast, when the doped quantum barrier layer of the quantum barrier layer 232 is added When the number of layers is increased, the luminous intensity of the light-emitting diode in the 222 nm to 405 nm band can be effectively improved. In detail, FIG. 4B is a relationship diagram showing the relationship between the number of layers of the doped quantum barrier layer of the quantum barrier layer and the emission intensity of the wavelength band near the wavelength of 365 nm in the light-emitting diode, and the horizontal axis and the vertical axis in FIG. 4B. The definition of the line segment is similar to that of FIG. 4A. However, FIG. 4B shows the illuminating band in the range of 222 nm to 405 nm in which the main peak is around 365 nm. As shown in FIG. 4B, the results show that increasing the number of layers of the doped quantum barrier layer 232 helps to improve the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm band.

The inventors infer from the results of the foregoing FIG. 4A and FIG. 4B that when the light-emitting band emitted by the light-emitting diode is around 450 nm, the carrier is less susceptible to the defect density due to the strong localized effect of the quantum well. Therefore, the N-type dopant is doped in the quantum barrier layer, and the luminescence intensity near 450 nm cannot be enhanced. Excessive doping may cause the carrier overflow phenomenon to decrease the luminescence intensity, as shown in FIG. 4A. However, for a light-emitting diode having an emission band around 365 nm, the effect of doping the N-type dopant in the quantum barrier layer is completely opposite to that of the light-emitting diode having an emission band around 450 nm.

In detail, as shown in FIG. 4B, when the illuminating band emitted by the illuminating diode is in the 222 nm to 405 nm illuminating band near the main peak at 365 nm, the carrier is subjected to defects due to the weakening of the local effect of the quantum well. The effect of density is enhanced, and the doping of N-type dopants (such as germanium) in a given quantum barrier layer helps to compensate for the effect of defect density on the carrier. In other words, the N-type dopant can also provide electrons as a radiation composite. Therefore, it can effectively improve the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm light-emitting band. The so-called N-type dopant herein is a Group IV dopant which is provided as a substitute for a Group III element from the outside. As shown in FIG. 4B, the luminescence intensity of the 222 nm to 405 nm luminescence band increases as the number of layers of the doped quantum barrier layer increases, especially when the number of layers of the doped quantum barrier layer k and the total number of quantum barrier layers are increased. When i satisfies the following relation, the effect of improving luminous efficiency is remarkable: when i is an even number, k≧i/2; when i is an odd number, k≧(i-1)/2.

In order to further verify the above inference, for the light-emitting diodes of the 222 nm to 405 nm light-emitting band, the simulated pair illumination is further shown when changing the number of layers of the doped quantum barrier layer 232 of FIG. 3, respectively, in FIGS. 5A to 8D. A diagram of the relationship between the electron concentration of the diode, the concentration of the hole, the probability of electron hole cavities, and the probability of non-radiative compounding, wherein the horizontal axis of Figs. 5 to 8 represents the distance from the surface of the substrate (unit: nanometer), and Fig. 5 to The graphs A, B, C, and D in FIG. 8 respectively represent the number of layers of the doped quantum barrier layer and the undoped quantum barrier layer, and the definition thereof is the same as that of the line segment AD in FIGS. 4A and 4B, and will not be described again.

It can be seen from the electron concentration simulation diagrams of FIGS. 5A to 5D that the electron concentration is gradually increased as the number of layers of the doped quantum barrier layer is increased. From the hole concentration simulation diagrams of FIG. 6A to FIG. 6D, when the number of layers of the doped quantum barrier layer is increased, the hole concentration is gradually reduced, and the whole quantum barrier layer is not doped. The hole concentration is the highest. From the electron hole composite probability simulation diagrams of FIG. 7A to FIG. 7D, it can be seen that although the overall hole distribution of the quantum barrier layer is uniformly doped, all of the quantum barrier layers of the quantum barrier layer of FIG. 7D should be undoped. It has a high electron hole composite probability. However, as can be seen from the trend of FIG. 7A to FIG. 7D, all of the quantum barrier layers 232 of FIG. 7A have the highest electron hole composite probability when doped, but the entire amount of FIG. 7D. When the sub-barrier layer is not doped, the electron hole compounding probability is the lowest. Therefore, FIG. 7A to FIG. 7D can also verify that the N-type dopant can provide electrons as a radiation composite, and thus can effectively improve the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm emission band. Furthermore, from the non-radiative composite probability plots of the electron holes of FIG. 8A to FIG. 8D, it can be seen that all of the quantum barrier layers of FIG. 8A have the lowest non-radiative composite probability when doped, and all of the quantum barrier layers of FIG. 8D are The electron tunnel non-radiation composite rate is the highest when undoped. According to the results of FIG. 7A to FIG. 7D and FIG. 8A to FIG. 8D, it can be known that doping N-type dopant in the quantum barrier layer can provide electrons and improve electronic electricity. The hole radiates a composite probability, and effectively improves the luminous efficiency, and at the same time reduces the non-radiative composite probability of the electron hole in a non-emissive type such as heat. It can also be verified that the N-type dopant can enhance the light-emitting diode to emit light at 222 nm to 405 nm. The inference of the luminous intensity of the band.

Table 1 shows that when the structure of the active layer in the light-emitting diode is as shown in FIG. 3, the luminous intensity of the light-emitting diode at different currents, and the forward voltage performance along with the doped quantum barrier layer and The number of layers of the doped quantum barrier layer is changed, wherein in the experiment of Table 1, the doping concentrations C ₁ , C ₂ , . . . C _{k of} each doped quantum barrier layer are, for example, ₂ ×10 ¹⁸ /cm ^{3 .} In the embodiment in which the illuminating wavelength of the present invention is 365 nm, the material of the quantum well is In _c Ga _1-c N, where 0 ≦ c ≦ 0.05, and the material of the quantum barrier layer is Al _d Ga _1-d N, d is Between 0 and 0.25, in the present embodiment, the optimum value of the aluminum content is between 0.09 and 0.20, and the thickness of the quantum barrier layer is, for example, 5 nm to 15 nm. In the present embodiment, the thickness is preferably 6 nm to 11 nm. . The results of Table 1 are shown in FIG. 9A and FIG. 9B , wherein FIG. 9A is a diagram showing the relationship between the number of different doping layers and the current-light output power curve in the quantum barrier layer of the LED. 9B shows the relationship between the number of different doping layers in the quantum barrier layer of the light-emitting diode and the current-voltage curve.

As can be seen from the results of Table 1 and FIG. 9A, the light output power of the light-emitting diodes 200A-200E increases as the number of doped quantum barrier layers in the existing quantum barrier layer increases. In detail, firstly, when the N-type dopant is not doped, the doping concentration is 0, but the gallium nitride material has its own background doping concentration, and the concentration will be different according to different epitaxial techniques or different epitaxial qualities. There is a difference. In this embodiment, since the background doping concentration is not measured, the undoped concentration is represented by NA. At this time, the N-type dopant is not doped in the six-layer quantum barrier layer ( For example, 光) light output power is 9.5 mW (light emitting diode 200A). When there are two layers of doped N-type dopants in the six-layer quantum barrier layer (for example, there is a purpose in the quantum barrier layer 232a-232f shown in FIG. Ground-doped two layers of quantum barrier layers 232a-232b closest to the N-type semiconductor 220, the light output power of the light-emitting diode 200B can be increased from 9.5 mW to 10.6 mW, which is undoped, and more preferably, when When there are four layers of doped quantum barrier layer 232 in the layer quantum barrier layer 232 (such as purposely doping the four-layer quantum barrier layer 232a-232d closest to the N-type semiconductor 220 in FIG. 3), the light-emitting diode The optical output power of the 200C can be greatly increased from the undoped 9.5 mW to 17.0 mW, which is twice as large as the original, so when the number k of the doped quantum barrier layer 232 is greater than or equal to the quantum barrier layer 232 When half of the total number of layers i is obtained, the luminous efficiency of the light-emitting diode 200C can be effectively improved. In addition, when doped with a five-layer quantum barrier layer, the light output power of the light-emitting diode 200D is 24.2 mW, and when all the quantum barrier layers 232 are doped (such as the full six-layer quantum barrier in FIG. 3) The layers 232a-232f are purposely doped), and the light output power of the LED 200E can be increased to 31.1 mW, which is nearly three times as much as the original.

In addition, as can be seen from the results of Table 1 and FIG. 9B, the doping of the N-type dopant in the quantum barrier layer can effectively increase the luminous efficiency of the light-emitting diode 200A, and further reduce the resistance of the quantum barrier layer. Further, the forward voltage of the light-emitting diode is lowered. For example, the forward voltage drops from 4.36 V, which is undoped with all quantum barrier layers, to 4.14 V, which is doped with all quantum barrier layers. The above results represent that increasing the number of doped layers in the quantum barrier layer can compensate for the effect of defect density on the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm band (the main peak is around 365 nm). In other words, the N-type dopant incorporated in the quantum barrier layer can effectively provide electrons for radiation recombination, and reduce energy release such as non-radiation recombination, such as thermal form, thereby effectively improving luminous efficiency. The above experimental results again verified the simulation results of the aforementioned FIGS. 5 to 8.

Therefore, it can be seen from the above that the light-emitting diode of the present invention can make the number of layers of the quantum barrier layer doped with the N-type dopant in the quantum barrier layer of the active layer conform to a specific ratio, thereby effectively improving the light-emitting diode Luminescence efficiency of the body in the 222 nm to 405 nm band. Especially when the number k of layers of the doped quantum barrier layer is greater than or equal to half of the total number of layers i of the quantum barrier layer, the effect of improving the luminous efficiency is remarkable, specifically, when i is an even number, k≧i/2; When i is an odd number, k≧(i-1)/2.

The effect of the doping concentration of the N-type dopant in the doped quantum barrier layer on the luminescence efficiency of the luminescent diode in the 222 nm to 405 nm band is further discussed below.

Table 2 shows that when the structure of the active layer in the light-emitting diode is as shown in FIG. 3, the four-layer quantum barrier layers 232a-232d adjacent to the N-type semiconductor layer are fixedly doped, so that in each experimental example of Table 2 The doped quantum barrier layer 232 is four layers, and the quantum barrier layers 232e-232f that are otherwise adjacent to the P-type semiconductor layer are undoped. Table 2 shows the relationship between the different doping concentrations in the doped quantum barrier layer of the light-emitting diode and the luminous intensity performance and the forward voltage performance. The results of Table 2 are shown in FIG. 10A and FIG. 10B , wherein FIG. 10A is a diagram showing the relationship between different doping concentrations and current-light output power curves in the quantum barrier layer in the light-emitting diode. FIG. 10B is a graph showing the relationship between different doping concentrations and current-voltage curves in the quantum barrier layer of the light-emitting diode.

As can be seen from the results of Table 2 and FIG. 10A and with reference to FIG. 3, the light output power of the light-emitting diode increases as the doping concentration increases, for example, as described above, when the N-type dopant is not doped, due to measurement Not the background doping concentration, so the undoped concentration is expressed by NA, and its optical output power is 9.5 mW (light emitting diode 200A); when the doping concentration of the four-layer doped quantum barrier layer 232a-232d When the frequency is 8x10 ¹⁷ cm ^-3 , the light output power of the light-emitting diode 200F can be increased from 9.5 mW which is undoped to 11.8 mW, and more preferably, when the doping concentration is 2×10 ¹⁸ cm ^-3 , the light emitting diode The optical output power of the body 200G can be greatly improved from the undoped 9.5 mW to twice the 17.0 mW. When the doping concentration is 4× ^{10 18} cm ^-3 , the light output power of the LED 200H is 19.1 mW. When the doping concentration is 6x10 ¹⁸ cm ^-3 , the light output power of the light emitting diode 200E can be increased to 21.5 mW. Therefore, from Table 2 and FIG. 10A, it can be inferred that in the quantum barrier layer of the light-emitting diode, the number of doped layers exceeds half of the total number of layers, and the doping concentration is 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm. ^{At 3} o'clock, the luminous efficiency of the light-emitting diode 200F-200I can be effectively improved.

In addition, as can be seen from the results of Table 2 and FIG. 10B, when the doping concentration is 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm ³ in the four-layer doped quantum barrier layer, the N-type dopant can enhance the light-emitting diode. In addition to the luminous efficiency of the body, the resistance of the quantum barrier layer can be lowered, thereby reducing the forward voltage of the light-emitting diode. For example, the forward voltage of the light-emitting diode is reduced from 4.36 V with a doping concentration of 0 to 4.09 V with a doping concentration of 6× ^{10 18} . The above results represent that increasing the doping concentration of the N-type dopant (for example, germanium) in the quantum barrier layer 232 can effectively compensate the effect of the defect density on the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm band. In other words, the N-type dopant incorporated in the quantum barrier layer can effectively provide electrons as a radiation composite, and reduce the energy release of a non-radiative composite such as a thermal form, thereby effectively improving luminous efficiency, and the above experimental results The simulation results of the aforementioned FIGS. 5 to 8 are also verified again.

It is worth mentioning that, according to the embodiment of the above-mentioned light-emitting diode 200B-200I of the present invention, at least one of the IV, V and VIA groups can also be selected as the N-type dopant, which can also achieve the provision of electrons. Radiation recombination, which can effectively improve luminous efficiency. In addition, the doping concentration in the doped quantum barrier layer may be equal to that of Table 1 and Table 2, and the doping concentration may have a gradient change. For example, the total number of layers of the quantum barrier layer is 6 layers, and the doped quantum barrier layer is 4 layers of 6 layers. The doping concentration of the 4-layer doped quantum barrier layer is close to the N-type. The semiconductor side is sequentially calculated as C ₁ , C ₂ , . . . C _k , and C _k ≦C _k-1 , for example, the doping concentration of the 4-layer doped quantum barrier layer 232a-232d is 6× ^{10 18} cm in sequence ^{- 3} , 5x 10 ¹⁸ cm ^-3 , 4x 10 ¹⁸ cm ^-3 , 3x 10 ¹⁸ cm ^-3 , in other words, the doping concentration change of the doped quantum barrier layer is from the first quantum barrier near the N-type semiconductor side The layer 232a is gradually reduced to the fourth layer 232d closest to the P-type semiconductor side, and thus, the incorporated N-type dopant can also effectively provide electrons as a radiation composite, whereby the luminous efficiency can be effectively improved.

Furthermore, the gradient of the doping concentration C ₁ to C _k in the doped quantum barrier layer may be 6× ^{10 18} cm ⁻³ , 7× ^{10 18} cm ⁻³ , 8× ^{10 18} cm from the N-type semiconductor side ^{. 3} , 6x10 ¹⁸ cm ^-3 , in other words, the doping concentration change may be that the intermediate layer number doping concentration is greater than the closest N-type semiconductor and the closest P-type semiconductor. Further, the quantum barrier layer is doped in the doping concentration gradient may also be a self-starting near the N-type semiconductor side in order, it is ^{6x10 18 cm -3, 5x10 18 cm} -3, 8x10 18 cm -3, 6x10 18 cm - ³ . In summary, as long as the doping concentration of the doped quantum barrier layer closest to the P-type semiconductor layer is less than or equal to the doping concentration of other quantum barrier layers in the k-doped quantum barrier layer, the doping can be performed. The N-type dopant effectively provides electrons as a radiation composite, thereby effectively improving luminous efficiency.

In summary, in the light-emitting diode of the present invention, the number of layers of the quantum barrier layer doped with the N-type dopant in the active layer conforms to a specific relationship, or the active layer is doped with an N-type. The closest dopant to the P-type semiconductor in the doped quantum barrier layer has the smallest doping concentration, or the doping concentration of the quantum barrier layer doped with the N-type dopant satisfies a specific relationship, so that the N-type dopant The effect of the defect of the gallium nitride on the carrier can be smoothed, and the recombination efficiency of the carrier of the light-emitting diode can be improved, so that the light-emitting diode of the present invention has any of the above technologies The means can greatly improve the luminous efficiency of the light-emitting diode in the 222 nm to 405 nm band.

In addition, the embodiment of the light-emitting diode of the present invention is not limited to the above-described type, and may be a horizontal electrode configuration or a vertical electrode configuration, and the present invention may be implemented, and thus is not limited thereto.

The second type of the invention further proposes the following types of light-emitting diodes.

11A to FIG. 11C are respectively structural diagrams of several light-emitting diodes according to an embodiment of the present invention, wherein the horizontal axis represents the position of each quantum barrier layer in the stack positional relationship in the light-emitting diode, and the vertical axis represents the relative conductive band energy. In the diagram, the thickness of each quantum barrier layer is indicated above (thickness unit: nanometer).

The light-emitting diode 200A of FIG. 11A represents a structure in which the thicknesses of the quantum barrier layers 232a-232f in the light-emitting diode are the same, and the light-emitting diode 200B of FIG. 11B represents the quantum barrier layer 232a-232f in the light-emitting diode. The structure in which the thickness is increased from the P-type semiconductor layer 240 to the N-type semiconductor layer 220, and the light-emitting diode 200C of FIG. 11C is a structure in which the thickness of the quantum barrier layers 232a to 232f is decreased from the P-type semiconductor layer 240 to the N-type semiconductor layer 220. .

FIG. 12 further illustrates a simulation diagram of the luminous intensity of each of the light emitting diodes of FIGS. 11A to 11C. As shown in FIG. 12, the light-emitting diodes 200C whose thicknesses of the quantum barrier layers 232a-232f are decreased from the P-type semiconductor layer 240 to the N-type semiconductor layer 220 have higher luminous intensities than the quantum barrier layers 232a-232f. The diode 200A is also higher than the light-emitting diode 200B whose thickness of the quantum barrier layer 232a-232f is increased from the P-type semiconductor layer 240 to the N-type semiconductor layer 220. Wherein, the luminescence intensity is further increased by the thickness of the quantum barrier layers 232a-232f from the P-type semiconductor layer 240 to the N-type semiconductor layer 220. The diode 200B is the weakest.

Further explore the mechanism of the influence of the thickness variation of the quantum barrier layer on the luminous intensity in the light-emitting diode 200A-200C.

13A to 13C are diagrams showing electron and hole concentration simulations of the respective light-emitting diodes of FIGS. 11A to 11C, respectively, and the horizontal axis represents the position of the film layer stacking distance from the substrate (unit: nanometer), wherein the position is 2060 nm. It is close to the P-type semiconductor layer 240 side, and the position 2000 nm is close to the N-type semiconductor layer 220 side, and the thick line and the thin line are respectively indicating the electron concentration and the hole concentration (unit: cm ^-3 ).

The inventors inferred from the foregoing results of FIGS. 11A to 11C, FIG. 12, and FIG. 13A to FIG. 13C that the influence mechanism of the thickness variation of the quantum barrier layer on the luminous intensity of the light-emitting diode is as follows.

Referring to FIG. 11B and FIG. 13B simultaneously, for the electron transfer of the light-emitting diode 200B, when the thickness of the equivalent sub-barrier layer 232a-232f is increased from the P-type semiconductor layer 240 to the N-type semiconductor layer 220, The migration is implanted from the N-type semiconductor layer 220 and traverses the respective quantum barrier layers 232a-232f to migrate toward the P-type semiconductor layer 240, and thus the thickness of the equivalent-sub barrier layers 232a-232f is from the N-type semiconductor layer 220 to the P-type semiconductor layer 240. When it is gradually thinned, electrons are more easily moved toward the P-type semiconductor layer 240, so that the electron concentration of the quantum well layer 234a closest to the P-type semiconductor layer 240 is too high.

On the other hand, for the hole migration of the light-emitting diode 200B, as shown in FIG. 13B in conjunction with FIG. 11B, although the quantum barrier layer 232a close to the P-type semiconductor layer 240 is thin, the hole is more easily N-type half The conductor layer 220 moves. However, as described above, since there is excessive electrons in the quantum well layer 234a closest to the P-type semiconductor layer 240, electrons overflow, rather than recombining in the quantum well layer, resulting in ineffective electrons and holes. The radiation recombination is generated, so that the concentration of the overall hole injection is lowered to cause the luminescence intensity to decrease.

On the other hand, as shown in FIG. 11C and FIG. 13C simultaneously, for the electron transfer of the light-emitting diode 200C, when the thickness of the equivalent sub-barrier layer 232a-232f is increased from the N-type semiconductor layer 220 to the P-type semiconductor layer 240, Since the migration of electrons from the N-type semiconductor layer 220 is moved to the P-type semiconductor layer 240 after passing through the quantum barrier layers 232a-232f, the thickness of the quantum barrier layer 232 is gradually increased to slightly slow down electrons to the P-type semiconductor layer 240. The tendency to move, as such, allows electron concentration to be evenly distributed among the quantum well layers 234a-234e of the active layer 230. In addition, the light-emitting diode 200C is gradually thickened from the N-type semiconductor layer 220 to the P-type semiconductor layer 240 by the thickness of the quantum barrier layer 232a-232f, whereby the electrons of the light-emitting diode 200C can be avoided as the light-emitting diode 2 The phenomenon that the polar body 200B is concentrated in the last quantum well layer 234a, so the overall electron injection concentration is not affected by the overflow effect caused by the excess electrons in the last quantum well layer 234a.

On the other hand, for the hole migration of the light-emitting diode 200C, as shown in FIG. 13C in conjunction with FIG. 11C, when a hole is injected from the P-type semiconductor layer 240 to the quantum well layer 234 closest to the P-type semiconductor layer 240 ( When the quantum well layer 234a) is as shown in FIG. 11C, since the thickness of the quantum barrier layers 232a-232f is thinned from the P-type semiconductor layer 240 to the N-type semiconductor layer 220, It is advantageous to inject holes into the next quantum well layer 234a-234e. As a result, compared with the light-emitting diode 200A and the light-emitting diode 200B, the hole concentration of the light-emitting diode 200C is relatively uniform in the quantum well layer 234, so that the structure of the light-emitting diode 200C has the most Good luminous intensity.

14A and 14B are energy band simulation diagrams of the light-emitting diodes of Figs. 11B and 11C, respectively, in which the definition of the horizontal axis is the same as that of Figs. 13A to 13C. As shown in FIG. 14A in conjunction with FIG. 11B, when the thickness of the quantum barrier layer 232a closest to the P-type semiconductor layer 240 is thin, the conductive band is lower than the Fermi level indicated by a broken line, which causes the closest The quantum well layer 234a of the P-type semiconductor layer 240 has no limiting effect and electrons will overflow into the P-type semiconductor layer 240.

On the other hand, as shown in FIG. 14B in conjunction with FIG. 11C, the quantum barrier layer 232a of the light-emitting diode 200C closest to the P-type semiconductor layer 240 is thicker, so that the conductive strip is higher than the Fermi level indicated by the broken line. The quantum well layer 234a closest to the P-type semiconductor layer 240 is caused to exert an appropriate limiting effect to prevent electrons from overflowing into the P-type semiconductor layer 240, resulting in a phenomenon that the electron hole non-radiative recombination reduces the luminous intensity.

Fig. 15 is a simulation diagram of electron current density of each of the light-emitting diodes of Figs. 11A to 11C, in which the definition of the horizontal axis is the same as that of Figs. 13A to 13C, and the vertical axis is the electron current density (unit: A/cm ² ). Referring to FIG. 8, the electron current density of the light-emitting diode 200B in the P-type semiconductor layer 240 is higher than that of the light-emitting diode 200A and the light-emitting diode 200C. This shows a phenomenon in which the light-emitting diode 200 B is in a current overflow.

Table 3 is a simulation diagram of the probability of overlapping wave functions of each quantum well layer 234.

Referring to Table 3 and FIG. 15, since the conductive strip of the LED 200B is lower than the Fermi level, the quantum well layer 234a closest to the P-type semiconductor layer 240 cannot confine the carrier to cause excessive electron overflow to the P. In the type semiconductor layer 240, a phenomenon in which the electron concentration of the light-emitting diode 200B in the quantum well layer 234a closest to the P-type semiconductor layer 240 exists in the quantum well layer 234a may be collectively verified in conjunction with FIG. 11B. In addition, as can be seen from FIG. 15, the electron overflow phenomenon of the light-emitting diode 200B is severe, and the wave function of the pair of electron holes cannot overlap and the composite light is emitted.

It can be seen from the above-mentioned inference that the structure in which the thickness of the quantum barrier layer 232 in the light-emitting diode 200 is gradually thinned from the N-type semiconductor layer 220 to the P-type semiconductor layer 240 like the light-emitting diode 200B cannot effectively enhance the light emission. strength. In contrast, when the thickness of the quantum barrier layer 232 in the light-emitting diode 200 is gradually changed from the N-type semiconductor layer 220 to the P-type semiconductor layer 240 as in the light-emitting diode 200 C, the electron and hole concentration can be effectively effective. The ground is evenly distributed in all the quantum well layers 234, and at this time, the wave function overlapping probability of the electron hole pair is higher than the wavelength function overlapping probability of the uniformity of the quantum barrier layer 232, so the light emitting two The polar body 200 C has the best luminous intensity in this embodiment.

It is worth mentioning that in the quantum barrier layer 232 of the active layer 230, the illuminating intensity of the illuminating diode 200 is more affected by the thickness variation of the first few quantum barrier layers 232 closest to the P-type semiconductor layer 240. The effect of the thickness variation in the quantum barrier layer 232 on the luminescence intensity of the luminescent diode 200 in the 222 nm to 405 nm band is further discussed below.

Table 4 shows that when the structure of the active layer 230 in the light-emitting diode 200 is as shown in FIG. 3, when the thickness of the quantum barrier layers 232a-232f at different positions (unit: nanometer) is changed, the current is 350 mA and 700 mA. The performance of the underlying luminescence intensity is injected, wherein each quantum well layer 234a-234e has a thickness of 3 nm. Moreover, in the present embodiment, the material of the quantum well layers 234a-232f is, for example, In _c Ga _1-c N, where 0 c 0.05, and the material of the quantum barrier layers 232a-232f is, for example, Al _d Ga _1-d N, where 0 d 0.25, and preferably 0.09 d 0.20.

In other words, in the present embodiment, the active layer 230 has a total of six quantum barrier layers 232a-232f, the structure of which can be referred to FIG. 3, the quantum barrier layers 232a-232f of the 6 layers from the P-type semiconductor side 240. The thickness is sequentially T ₁ , T ₂ , T ₃ ... T ₆ , that is, T ₁ is the thickness of the quantum barrier layer 232a closest to the P-type semiconductor side 240, and T ₆ (T _i , this embodiment is i=6 is an example) the thickness of the quantum barrier layer 232f closest to the N-type semiconductor side 220.

As can be seen from Table 4, the light-emitting diode I had a luminous intensity of 17.0 mW under a current of 350 mA. From the results of Table 4 and referring to FIG. 3, in the first three layers of quantum barrier layers 232a-232c of the light-emitting diode 200 closest to the P-type semiconductor layer 240, when the quantum barrier layer 232a is adjacent to the P-type semiconductor layer 240. When the thickness T _{1 is} greater than the thicknesses T ₂ and T ₃ of the quantum barrier layers 232b-232c closer to the N-type semiconductor layer 220, that is, when T _{1 is} greater than T ₂ and T ₃ , the light-emitting diode can be effectively improved. The luminous intensity of the body.

Specifically, the light-emitting intensity of the light-emitting diode II is greatly reduced to 5.9 mW compared to the light-emitting diode I, since the light-emitting diode II is close to the thickness T ₁ of the quantum barrier layer 232a of the P-type semiconductor layer 240. Thinner, but not effective in confining electrons into quantum wells, limits the luminescence intensity to a significant extent, which is consistent with the mechanism inferred above.

On the other hand, after the light-emitting diode III thins the thicknesses T ₃ and T ₄ of the quantum barrier layers 232c and 232d of the intermediate layer of the light-emitting diode I, the luminous intensity can be increased to 24.0 mW, which represents a hole. It is easier to inject more of the quantum wells 234a-234e into the N-type semiconductor layer 220 under this thickness design. Further, as the light-emitting diode IV further thins the thickness of the quantum barrier layers 232e and 232f, the light output power thereof is further increased to 30.3 mW.

Further, as the light-emitting diode V further reduces the thickness T _{1 of the} quantum barrier layers 232a-232f from the P-type semiconductor layer 240 to the N-type semiconductor layer 220, as shown in Table 4, T _{1 is} gradually reduced to T ₆ The luminous intensity is increased to about 33.1 mW. In other words, in the light-emitting diode, when the thickness of the three-layer quantum barrier layer 232 closest to the P-type semiconductor layer 240 satisfies the relationship that T _{1 is} greater than T ₂ and T ₃ , the hole can be effectively made uniform. It is distributed in the quantum well of the active layer 230, and suppresses the overflow phenomenon of electrons, thereby effectively improving the luminous intensity of the light-emitting diode.

Figure 16 is a graph of light output curve versus injection current for each of the light-emitting diodes of Table 4. As shown in Table 4 and FIG. 16, the effect of improving the light output efficiency of the light-emitting diode can be achieved by modulating the thickness of the quantum barrier layers 232a-232f in the active layer 230, especially for the mobility of the hole. For the three-layer quantum barrier layers 232a-232c closest to the P-type semiconductor layer 240, the effect of effectively improving the light-emission intensity can be achieved by appropriately modulating the thickness of the three-layer quantum barrier layers 232a-232c.

Specifically, when the thickness of the quantum barrier layer 232 in the i layer of the active layer 230 satisfies the thickest T ₁ , the effect of improving the luminous intensity of the light emitting diode can be achieved.

According to the results of the foregoing Table 4, the thickness of the quantum barrier layer of the intermediate layer, for example, T ₃ , T ₄ , may be thinner than the quantum barrier of the N-type semiconductor layer 220 and the P-type semiconductor layer 240, like a light-emitting diode. As shown in III, the light output efficiency can be effectively improved. On the other hand, the thickness of the quantum barrier layer 232 close to the N-type semiconductor layer 220 can be made smaller than the thickness of the quantum barrier layers 232a, 232b of the P-type semiconductor layer 240, and the thickness of the quantum barrier layers 232c-232f can be made uniform. As shown by the light-emitting diode IV, the light output efficiency can be effectively improved. Further, the thickness of the equivalent sub-barrier layer 232 is gradually thinned from the P-type semiconductor layer 240 to the N-type semiconductor layer 220, and as shown by the light-emitting diode V, the light-emission intensity is optimum.

According to the foregoing experimental results and the inference mechanism of the inventors, it can be known that the electron hole pairs are uniformly distributed in the quantum well of the active layer 230, and the limitation effect of the quantum barrier layer carriers close to the P-type semiconductor layer 240 is improved. Effectively improve the luminous efficiency of the light-emitting diode.

Taking the six-layer quantum barrier layer 232 in the foregoing experiment as an example, the thickness T ₁ of the first quantum barrier layer 232a closest to the P-type semiconductor layer 240 is the largest, and the thickness T _{2 of the} second quantum barrier layer 232b. It is less than or equal to the thickness T ₁ of the first quantum barrier layer 232a, thereby making the first quantum well closest to the P-type semiconductor layer 240 have a better limiting effect, avoiding the overflow effect of the electron, and elevating the electron Radiation recombination efficiency of the hole.

It can be seen from the above experiments and inference that by maximizing the thickness T ₁ of the first quantum barrier layer 232a closest to the P-type semiconductor layer 240, the overflow effect of electrons can be effectively avoided, thereby improving the electron hole. Radiation recombination efficiency. Thus in this area skilled in the art can infer, when the layer thickness of the quantum barrier layer 232b second T ₂ is equal to ₁ when a first layer of a quantum barrier, will be such that the P-type semiconductor layer 240 closest to the thickness T of layer 232a The first layer of quantum wells simultaneously exerts better confinement effects, thereby also achieving the effect of avoiding the electron overflow effect and improving the radiation recombination efficiency of the electron holes.

Furthermore, the thickness T _{3 of the} third quantum barrier layer 232c is the thinnest between T1 and T3, as shown in the light-emitting diodes III to V of Table 4, thereby facilitating the injection of the holes. The holes are more efficiently injected into the quantum well layer 234 on the side of the N-type semiconductor layer 220, so that the distribution of the holes in the active layer 230 is more uniform. Further, as shown by the light-emitting diode I of Table 4, when T ₁ > T ₂ = T ₃ , the light output efficiency can be effectively improved as compared with the light-emitting diode II. Further, as shown by the light-emitting diodes IV and V of Table 4, the thickness T _i of the quantum barrier layer closest to the N-type semiconductor side of T ₆ (this embodiment i=6, thus T ₆ ) is the most When thin, the illuminating intensity of the LEDs IV and V at 350 mA and 700 mA current injection is excellent, that is, by the thickness T _i closest to the N-type semiconductor layer in the i-layer quantum barrier layer The thinnest, can effectively improve the light output efficiency.

The following further changes the number of layers of quantum wells and quantum barrier layers in the active layer. Table 5 shows changes in the number of layers of quantum wells and quantum barrier layers in the active layer (six, nine, and eleven layers of quantum barriers) and changes. The thickness of the quantum barrier layer at different positions (unit: nanometer), the luminescence intensity at 350 mA and 700 mA current injection, wherein the thickness of each quantum well layer is 3 nm. In other words, in the structure column of Table 5, the numbers represent the thicknesses T ₁ -T _{i of the} respective layers of the quantum barrier layers 232a-232i, and in the column, the numbers from right to left represent the values from the P-type semiconductor side, respectively. The thickness of the quantum barrier layers 232a-232i is sequentially T ₁ , T ₂ , T ₃ ... T _i

As can be seen from the results of Tables 4 and 5, whether the active layer 230 is a structure using eight or ten quantum well layers 234 (ie, nine or eleven layers of quantum barrier layers 232), in the active layer 230. When the thickness of each of the i-level quantum barrier layers 232 satisfies: T ₁ >T ₂ ≧T ₃ , the quantum barrier layer 232 is designed to have a gradual thickness to effectively improve the luminous efficiency of the light-emitting diode. For example, comparing the structure of the eight-layer quantum well layer 234 of the light-emitting diode VI and the light-emitting diode VII, it is found that in the structure of the eight-layer quantum well layer 234 of the light-emitting diode VI, the quantum barrier is The layer 232 thickness T ₈ to T _{2 was} set to 9 nm, and T _{1 was} set to 11 nm. When the thickness T ₉ to T ₁ of the modulating sub-block layer 232 of the light-emitting diode VII is sequentially 2/3/3/5/5/7/7/9/11 nm, the luminescent light can be The light output efficiency (luminous intensity) of the polar body is effectively increased from the original 16.4 mW to 24.7 mW.

On the other hand, comparing the structures of the ten quantum well layers 234 of the light-emitting diode VIII and the light-emitting diode IX, it can be found that the thicknesses T ₁₁ to T _{2 of} the quantum barrier layer 232 are both set to 9 nm, and The thickness T _{1 of the} quantum barrier layer 232 is set to 11 nm. When the thickness of the variable sub-barrier layer 232 is T ₁₁ to T ₁ sequentially, the thickness is 2/2/3/3/3/5/5/7/7/9/11 nm, thereby The light output efficiency (luminous intensity) of the light-emitting diode is effectively increased from the original 11.3 mW to 20.7 mW.

It is worth mentioning that, in order to further increase the luminous intensity of the light-emitting diode, the inventors propose that the number of layers of the quantum barrier layer 232, the doping concentration, etc., can be borrowed by the variable sub-block layer 232. The N-type dopant is intentionally doped to reduce the influence of the defect density of the active region on the carrier, and the luminous efficiency is effectively improved, especially for the active layer 230 to emit a wavelength range of 222 nm to 405 nm. The light has a significant boost.

Especially when the number of layers k of the doped quantum barrier layer 232 and the total number i of the quantum barrier layers 232 satisfy the following relationship, the effect of improving the luminous efficiency is remarkable: when i is an even number, k≧i/2; When it is odd, k≧(i-1)/2. In other words, when the number of doped layers exceeds half of the total number of layers in the quantum barrier layer 232 of the light-emitting diode, and the doping concentration is 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm ³ , it can be effectively further improved. Luminous efficiency of the light-emitting diode.

In summary, in the light-emitting diode of the present invention, by designing the thickness of the quantum barrier layer in the active layer to conform to a specific relationship, the holes can be uniformly distributed in the quantum well layer, thereby improving the light-emitting diode. The composite efficiency of the carrier of the body, so that the light-emitting diode of the present invention can greatly improve the luminous intensity of the light-emitting diode in the wavelength range of 222 nm to 405 nm by any of the above techniques.

Further, the embodiment of the light-emitting diode of the present invention is not limited to the foregoing The present invention can be implemented in a horizontal electrode configuration or a vertical electrode configuration, and thus is not limited thereto.

Figure 17 is an embodiment of the light-emitting diode of the present invention. As shown in FIG. 17, the film layer of the light-emitting diode 300 includes a contact layer 310 from top to bottom, the N-type semiconductor layer 220, the active layer 230, the electron barrier layer 270 and the intermediate layer 280, and the P-type. a semiconductor layer 240; a reflective layer 320; a bonding layer 330; and a carrier substrate 340.

Fig. 18 is another embodiment of the light-emitting diode of the present invention. As shown in FIG. 18, the film layer of the light-emitting diode 400 includes the substrate 210, the nitride semiconductor cladding layer 212, the N-type semiconductor layer 220, and the carrier substrate 340, and the N-type semiconductor layer, in order from top to bottom. 220 is stacked on the carrier substrate 340. The active layer 230, the electron barrier layer 270 and the intermediate layer 280, the P-type semiconductor layer 240, the contact layer 310, and the bonding layer 330 are mainly disposed on the left side of FIG. a first laminate formed, and a second laminate located to the right of the first laminate and spaced apart from the first laminate, wherein the second laminate is mainly composed of the contact layer 310 and the bonding layer 330 Composition. Moreover, the light-emitting diode 400 can be disposed between the contact layer 310 of the first first laminate on the right side and the bonding layer 330, or the carrier substrate 340 is disposed adjacent to the surface of the second second laminate. The invention is not limited thereto.

Fig. 19 is still another embodiment of the light-emitting diode of the present invention. As shown in FIG. 19, the film structure of the light-emitting diode 500 is similar to that of FIG. 20 except that the light-emitting diode 500 of FIG. 19 further omits the N-type semiconductor compared to the light-emitting diode 400 of FIG. Substrate above layer 220 210 and the nitride semiconductor cladding layer 212 are the same as those in the foregoing FIG. 20 and will not be described again. Moreover, the light-emitting diode 500 can be disposed between the contact layer 310 of the first first laminate on the right side and the bonding layer 330, or the carrier substrate 340 is adjacent to the second second stack. On the surface, the invention is not limited thereto.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

200, 200A-200I, I-IX‧‧‧Light Emitting Diodes

210‧‧‧Substrate

212‧‧‧ nitride semiconductor coating

220‧‧‧N type semiconductor layer

222‧‧‧First N-type doped gallium nitride layer

224‧‧‧Second N-type doped gallium nitride layer

230‧‧‧ active layer

230A‧‧‧Single quantum well active layer

230B‧‧‧Multiple quantum well active layer

232, 232a, 232b, 232c, 232d, 232e, 232f‧‧‧ quantum barrier layer

234, 234a, 234b, 234c, 234d, 234e‧‧‧ quantum wells

240‧‧‧P type semiconductor layer

242‧‧‧First P-type doped GaN layer

244‧‧‧Second P-type doped GaN layer

250‧‧‧first electrode

260‧‧‧second electrode

T ₁ , T ₂ , T ₃ ... T _i ‧ ‧ thickness of the quantum barrier layer

310‧‧‧Contact layer

320‧‧‧reflective layer

330‧‧‧ joint layer

340‧‧‧bearing substrate

1 is a schematic cross-sectional view of a light emitting diode according to an embodiment of the present invention.

2A is a schematic cross-sectional view showing a single quantum well active layer in a light-emitting diode according to an embodiment of the invention.

2B is a schematic cross-sectional view showing a multiple quantum well active layer in a light-emitting diode according to an embodiment of the invention.

3 is an enlarged schematic cross-sectional view of an active layer of a light emitting diode.

Fig. 4A shows a comparative example of the light-emitting diode of the present invention.

Fig. 4B shows an embodiment of the light-emitting diode of the present invention.

5A to 5D are graphs showing the relationship between the simulated electron concentration of the light-emitting diodes when the number of layers of the doped quantum barrier layer of Fig. 3 is changed, respectively.

6A to 6D respectively show changes in the doped quantum barrier of FIG. A graph showing the relationship between the hole concentration of the light-emitting diodes in the number of layers of the layer.

7A to 7D are diagrams showing the relationship between the simulated electron-electrode composite probability of the light-emitting diode when the number of layers of the doped quantum barrier layer of FIG. 3 is changed, respectively.

8A to 8D are graphs showing the relationship between the simulated non-radiative composite probability of the light-emitting diodes when the number of layers of the doped quantum barrier layer of FIG. 3 is changed, respectively.

FIG. 9A is a graph showing the relationship between the number of different doping layers and the current-light output power curve in the quantum barrier layer of the light-emitting diode.

FIG. 9B is a graph showing the relationship between the number of different doping layers and the current-voltage curve in the quantum barrier layer of the light-emitting diode.

FIG. 10A is a graph showing the relationship between different doping concentrations and current-light output power curves in a quantum barrier layer in a light-emitting diode. FIG.

FIG. 10B is a graph showing the relationship between different doping concentrations and current-voltage curves in the quantum barrier layer of the light-emitting diode.

11A to 11C are respectively structural diagrams of the structure of several light emitting diodes according to an embodiment of the present invention.

Fig. 12 is a simulation diagram of luminous intensity of each of the light-emitting diodes of Figs. 11A to 11C.

13A to 13C are diagrams showing electron and hole concentration simulations of the respective light-emitting diodes of Figs. 11A to 11C, respectively.

14A and 14B are energy band simulation diagrams of the light-emitting diodes of Figs. 11B and 11C, respectively.

Fig. 15 is a simulation diagram of electron current density of each of the light-emitting diodes of Figs. 11A to 11C.

Figure 16 is a graph of light output curve versus injection current for each of the light-emitting diodes of Table 4.

Figure 17 is an embodiment of the light-emitting diode of the present invention.

Fig. 18 is another embodiment of the light-emitting diode of the present invention.

Fig. 19 is still another embodiment of the light-emitting diode of the present invention.

Lines A, B, C, D‧‧

Claims (23)

A light emitting diode comprising: a substrate; an N-type semiconductor layer on the substrate; an active layer having a defect density DD, wherein DD ≧ 2×10 ⁷ /cm ³ , the active layer is located in the N-type semiconductor layer The wavelength λ of the emitted light is 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and (i-1) layer quantum wells, and each quantum well is in any two layers of quantum barrier layers. And i is a natural number greater than or equal to 2, wherein the doped N-type dopant is in at least k layers of the quantum barrier layers, k is a natural number greater than or equal to 1, and when i is an even number, k≧i /2, when i is an odd number, k ≧ (i-1) /2; a P-type semiconductor layer on the active layer, and the quantum barrier closest to the P-type semiconductor in the k-layer quantum barrier layer The doping concentration of the layer is less than or equal to the doping concentration of the other quantum barrier layers in the k-layer quantum barrier layer; and a first electrode and a second electrode, wherein the first electrode is located in a portion of the N-type semiconductor layer And the second electrode is located on a partial region of the P-type semiconductor layer. The light-emitting diode according to claim 1, wherein the n-type doped n-layer quantum barrier layer is located in the k-layer of the quantum barrier layer closest to the N-type semiconductor layer. The light-emitting diode according to claim 2, wherein the k-layer quantum barrier layer has a doping concentration of at least 5× ^{10 17} /cm ³ . The light-emitting diode according to claim 2, wherein the doping concentration of each quantum barrier layer in the k layer is sequentially C ₁ , C ₂ , ... C _k from the N-type semiconductor side. And C _k ≦C _k-1 . The light-emitting diode according to claim 1, wherein the material of the quantum barrier layer comprises Al _x In _y Ga _1-xy N, wherein 0 x 1,0 y 0.3, and x+y 1. The light-emitting diode according to claim 1, wherein each of the quantum barrier layers has a thickness of between 5 nm and 15 nm. The light-emitting diode according to claim 5, wherein the materials of the quantum wells include Al _m In _n Ga _1-mn N, wherein 0 m<1,0 n 0.5, m+n 1, and x>m, n≧y. A light emitting diode comprising: a substrate; an N-type semiconductor layer on the substrate; an active layer having a defect density DD, wherein DD ≧ 2×10 ⁷ /cm ³ , the active layer is located in the N-type semiconductor In a partial region of the layer, the active layer emits a light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well, each quantum well in any two-layer quantum Between the barrier layers, and i is a natural number greater than or equal to 2, wherein the doped N-type dopant is in at least the k-layer of the quantum barrier layers, k is a natural number greater than or equal to 1, when i is an even number , k≧i/2, when i is an odd number, k≧(i-1)/2, the doping concentration of the k-layer quantum barrier layer is 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm ³ ; a semiconductor layer on the active layer; and a first electrode and a second electrode, wherein the first electrode is located on a partial region of the N-type semiconductor layer, and the second electrode is located in a portion of the P-type semiconductor layer The doping concentration of the quantum barrier layer closest to the P-type semiconductor in each of the quantum barrier layers in the k layer is less than or equal to the k-layer quantum barrier layer. Other doping concentration of the quantum barrier layer. A light emitting diode comprising: a substrate; an N-type semiconductor layer and a P-type semiconductor layer, wherein the N-type semiconductor layer is between the substrate and the P-type semiconductor layer; and an active layer is located in the N-type semiconductor Between the layer and the P-type semiconductor layer, the active layer emits a light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well layer, each quantum well The layer is between any two layers of quantum barrier layers, each quantum barrier layer is composed of the same material, and i is a natural number greater than or equal to 2, and the thickness of each quantum barrier layer in the i layer is from the P-type semiconductor side The first order is T ₁ , T ₂ , T ₃ ... T _i , wherein T _{1 is} greater than T ₂ and T ₃ ; and a first electrode and a second electrode, wherein the first electrode is located in the N-type semiconductor layer And a portion of the second electrode is located on a portion of the P-type semiconductor layer. The light-emitting diode according to claim 9, wherein T ₂ ≧T ₃ . The light-emitting diode according to claim 9, wherein in the i-layer quantum barrier layer, the thickness T _i closest to the N-type semiconductor layer is the thinnest. The light-emitting diode according to claim 9, wherein the doped N-type dopant is in at least k layers of the quantum barrier layers, and k is a natural number greater than or equal to 1, when i is an even number, K≧i/2, when i is an odd number, k≧(i-1)/2. The light-emitting diode according to claim 12, wherein the k-layer quantum barrier layer has a doping concentration of 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm ³ . The light-emitting diode according to claim 9, wherein in the i-layer quantum barrier layer, the thickness T ₁ of the quantum barrier layer closest to the P-type semiconductor layer is between 6 nm and 15 nm. The light-emitting diode according to claim 9, wherein the material of the quantum barrier layer comprises Al _x In _y Ga _1-xy N, wherein 0 x 1,0 y 0.3, and x+y 1. The light-emitting diode according to claim 15, wherein the material of the quantum well layer comprises Al _m In _n Ga _1-mn N, wherein 0 m<1,0 n 0.5, m+n 1, and x>m, n≧y. A light emitting diode comprising: a substrate; an N-type semiconductor layer and a P-type semiconductor layer, wherein the N-type semiconductor layer is between the substrate and the P-type semiconductor layer; and an active layer is located in the N-type semiconductor Between the layer and the P-type semiconductor layer, the active layer emits a light having a wavelength λ of 222 nm ≦ λ ≦ 405 nm, and the active layer includes an i-layer quantum barrier layer and an (i-1) layer quantum well layer, each quantum well The layer is between any two layers of quantum barrier layers, each quantum barrier layer is composed of the same material, and i is a natural number greater than or equal to 2, and the thickness of each quantum barrier layer in the i layer is from the P-type semiconductor side The first order is T ₁ , T ₂ , T ₃ ... T _i , wherein the thickness of the i-layer quantum barrier layer satisfies: T ₁ = T ₂ > T ₃ ; and a first electrode and a second electrode, The first electrode and the second electrode are respectively located on a partial region of the N-type semiconductor layer and a partial region of the P-type semiconductor layer. The light-emitting diode according to claim 17, wherein in the i-layer quantum barrier layer, the thickness T _i closest to the N-type semiconductor layer is the thinnest. The light-emitting diode according to claim 17, wherein the doped N-type dopant is at least k layers of the quantum barrier layers, and k is a natural number greater than or equal to 1, when i is an even number, K≧i/2, when i is an odd number, k≧(i-1)/2. The light-emitting diode according to claim 19, wherein the k-layer quantum barrier layer has a doping concentration of 5× ^{10 17} /cm ³ to 1×10 ¹⁹ /cm ³ . The light-emitting diode according to claim 17, wherein in the i-layer quantum barrier layer, the thickness T ₁ of the quantum barrier layer closest to the P-type semiconductor layer is between 6 nm and 15 nm. The light-emitting diode according to claim 17, wherein the material of the quantum barrier layer comprises Al _x In _y Ga _1-xy N, wherein 0 x 1,0 y 0.3, and x+y 1. The light-emitting diode according to claim 22, wherein the materials of the quantum well layers comprise Al _m In _n Ga _1-mn N, wherein 0 m<1,0 n 0.5, m+n 1, and x>m, n≧y.