JP2018200928A - Semiconductor laser element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a semiconductor laser device capable of oscillating a laser beam having a long wavelength, suppressing an increase in voltage, and further reducing absorption loss. SOLUTION: This is a semiconductor laser device having an n-side semiconductor layer, an active layer, and a p-side semiconductor layer, each of which is made of a nitride semiconductor, in order from upward, and the n-side semiconductor layer is an n-type. It has an n-side composition gradient layer containing impurities, and the p-side semiconductor layer has an undoped p-side composition gradient layer and an electron barrier layer, and is selected from the following (1) to (3). Has one or more structures. (1) The average bandgap energy of the p-side semiconductor layer is larger than the average bandgap energy of the n-side composition inclined layer, and (2) the bandgap energy of the lower end of the p-side composition inclined layer is the n-side. The p-side semiconductor layer, which is larger than the bandgap energy at the upper end of the composition inclined layer, (3) further has a p-side intermediate layer. [Selection diagram] FIG. 2A

Description

  The present invention relates to a semiconductor laser device.

  Nowadays, a semiconductor laser element having a nitride semiconductor (hereinafter also referred to as a “nitride semiconductor laser element”) can oscillate light ranging from the ultraviolet region to green. Has been used for a long time. As such a semiconductor laser element, a structure having an n-side cladding layer, an n-side light guide layer, an active layer, a p-side light guide layer, and a p-side cladding layer in this order on a substrate is known (for example, Patent Documents 1 and 2).

JP 2003-273473 A JP 2014-1331019 A

  Nitride semiconductor laser devices are required to have longer wavelengths and improved characteristics in the longer wavelength region. For example, a green laser having a long wavelength and good characteristics can be used as a light source for a projector, and is expected to expand the range of application.

  In a nitride semiconductor laser device, when a quantum well structure having a nitride semiconductor containing In as a well layer is used, the oscillation wavelength increases from the blue wavelength band to the green wavelength band. The light confinement to the light is reduced. As a result, the threshold current increases and the current density during laser oscillation increases. As the current density increases, the effective transition interval increases due to localized level shielding and band filling, and the oscillation wavelength shifts to a shorter wavelength.

The present application includes the following inventions.
A semiconductor laser element having an n-side semiconductor layer, an active layer, and a p-side semiconductor layer, each of which is made of a nitride semiconductor, in order upward.
The n-side semiconductor layer has a band gap energy that decreases upward, and includes n-type impurities at an n-type impurity concentration greater than 5 × 10 ¹⁷ / cm ^{3 and} less than or equal to 2 × 10 ¹⁸ / cm ^3. Having a side composition gradient layer,
The active layer is
A band that is disposed in contact with the n-side composition gradient layer, has a higher n-type impurity concentration and smaller film thickness than the n-side composition gradient layer, and is larger than the band gap energy at the upper end of the n-side composition gradient layer. An n-side barrier layer having gap energy;
A plurality of well layers disposed above the n-side barrier layer, and an intermediate barrier layer sandwiched between the plurality of well layers,
The p-side semiconductor layer is
The band gap energy increases upward, and a p-side composition gradient layer that is undoped;
An electron barrier layer disposed above the p-side composition gradient layer, having a band gap energy larger than that of the intermediate barrier layer and having a p-type impurity concentration larger than that of the p-side composition gradient layer;
Have
A semiconductor laser device having one or more structures selected from the following (1) to (3).
(1) When the film thickness of the n-side composition gradient layer is t, the average band gap energy of the p-side semiconductor layer in a distance range of t upward from the interface between the active layer and the p-side semiconductor layer Is larger than the average band gap energy of the n-side composition gradient layer,
(2) The band gap energy at the lower end of the p-side composition gradient layer is larger than the band gap energy at the upper end of the n-side composition gradient layer.
(3) The p-side semiconductor layer is further a p-side intermediate layer that connects the p-side composition gradient layer and the electron barrier layer, and is equal to or higher than a band gap energy at an upper end of the p-side composition gradient layer; The p-side intermediate layer has a band gap energy that is less than the band gap energy of the electron barrier layer and is undoped.

  According to such a semiconductor laser element, it is possible to oscillate long-wavelength laser light by reducing the threshold current, suppress an increase in voltage, and further reduce absorption loss. it can.

FIG. 1 is a schematic cross-sectional view of a semiconductor laser device according to an embodiment. FIG. 2A is a diagram schematically showing the active layer of the semiconductor laser device according to the embodiment and the band gap energy in the vicinity thereof. FIG. 2B is a diagram schematically illustrating another example of the semiconductor laser element according to the embodiment. FIG. 2C is a diagram schematically illustrating another example of the semiconductor laser element according to the embodiment. FIG. 3A is a diagram showing an active layer of the semiconductor laser device of FIG. 2A and a comparative semiconductor laser device and a refractive index distribution in the vicinity thereof. FIG. 3B is a diagram showing the active layer of the semiconductor laser device of FIG. 2B and the comparative semiconductor laser device and the refractive index distribution in the vicinity thereof. FIG. 4A is a diagram showing the light intensity distribution in the active layer of the semiconductor laser device of FIG. FIG. 4B is a diagram showing the light intensity distribution in the active layer and its vicinity of the semiconductor laser device of FIG. 2B and the semiconductor laser device for comparison. FIG. 5A is a partially enlarged view of the n-side composition gradient layer and the vicinity thereof. FIG. 5B is a partially enlarged view of the p-side composition gradient layer and the vicinity thereof. FIG. 6 is a graph showing an image of the relationship between the Si doping amount and the free carrier absorption loss in the composition gradient layer. FIG. 7 is a graph showing IL characteristics of the semiconductor laser elements of Example 1 and Comparative Example 1. FIG. 8 is a graph showing IV characteristics of the semiconductor laser elements of Example 1 and Comparative Example 1. FIG. 9 is a graph showing IL characteristics of the semiconductor laser elements of Example 2 and Comparative Example 1. FIG. 10 is a graph showing IV characteristics of the semiconductor laser elements of Example 2 and Comparative Example 1.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the embodiment described below exemplifies a method for embodying the technical idea of the present invention, and the present invention is not limited to the following embodiment. Furthermore, in the following description, the same name and symbol indicate the same or the same members, and detailed description thereof will be omitted as appropriate.

FIG. 1 is a schematic cross-sectional view of a semiconductor laser device 100 according to the present embodiment, and shows a cross section of the semiconductor laser device 100 in a direction perpendicular to the resonator direction. FIG. 2A is a diagram schematically showing the magnitude relationship between the active layer 3 of the semiconductor laser element 100 and the band gap energy in the vicinity thereof. The semiconductor laser element 100 includes an n-side semiconductor layer 2, an active layer 3, and a p-side semiconductor layer 4 each made of a nitride semiconductor in this order upward. The n-side semiconductor layer 2 has a band gap energy that decreases upward, and contains n-type impurities at an n-type impurity concentration greater than 5 × 10 ¹⁷ / cm ^{3 and} less than or equal to 2 × 10 ¹⁸ / cm ^3. A side composition gradient layer 26 is provided. The active layer 3 is disposed in contact with the n-side composition gradient layer 26, and has an n-type impurity concentration and a film thickness smaller than those of the n-side composition gradient layer 26, and the band gap at the upper end of the n-side composition gradient layer 26. The n-side barrier layer 31 has a band gap energy larger than the energy. The active layer 3 further includes a plurality of well layers 32A and 32B disposed above the n-side barrier layer 31, and an intermediate barrier layer 33 sandwiched between the plurality of well layers 32A and 32B. The p-side semiconductor layer 4 has a band gap energy that increases upward, and has a p-side composition gradient layer 41 that is undoped. The p-side semiconductor layer 4 is further disposed above the p-side composition gradient layer 41, and has an electron barrier layer 42 that has a larger band gap energy than the intermediate barrier layer 33 and a higher p-type impurity concentration than the p-side composition gradient layer 41. Have

The semiconductor laser device 100 further has one or more structures selected from the following (1) to (3). An example of structures (1) and (2) is shown in FIG. 2A, and an example of structure (3) is shown in FIG. 2B.
(1) When the film thickness of the n-side composition gradient layer 26 is t, the average band gap energy of the p-side semiconductor layer 4 in a distance range t from the interface between the active layer 3 and the p-side semiconductor layer 4 upward. Is larger than the average band gap energy of the n-side composition gradient layer 26.
(2) The band gap energy at the lower end of the p-side composition gradient layer 41 is larger than the band gap energy at the upper end of the n-side composition gradient layer 26.
(3) The p-side semiconductor layer 4 is further a p-side intermediate layer 45 that connects the p-side composition gradient layer 41 and the electron barrier layer 42, and is equal to or higher than the band gap energy at the upper end of the p-side composition gradient layer 41. The p-side intermediate layer 45 has a band gap energy lower than that of the electron barrier layer 42 and is undoped.

  First, the n-side composition gradient layer 26 will be described. The n-side composition gradient layer 26 has a band gap energy that decreases upward, and the upper band gap energy is smaller than the band gap energy of the n-side barrier layer 31. In other words, the n-side composition gradient layer 26 has the first surface 26a on the n-side barrier layer 31 side and the second surface 26b opposite to the first surface 26a, and the band gap energy is the second surface. 26b decreases from 26b toward the first surface 26a. By having such a configuration, light confinement in the active layer 3 can be enhanced, and the laser oscillation threshold current density can be reduced. Thereby, shielding of the localized level can be suppressed, and a short wavelength shift accompanying an increase in current injection can be suppressed. Furthermore, since there is a band gap difference between the upper end of the n-side composition gradient layer 26 and the n-side barrier layer 31, the electron concentration locally increases in the vicinity of the first surface 26 a in the n-side composition gradient layer 26. . As a result, suppression of hole overflow is promoted, so that the film thickness of the n-side barrier layer 31 provided for suppressing hole overflow can be reduced as compared with the case where there is no such promotion. By reducing the thickness of the n-side barrier layer 31, which is an n-type impurity high concentration layer, it is possible to reduce shielding of localized levels due to electrons caused by the n-side barrier layer 31. Further, since the n-side composition gradient layer 26 can be disposed closer to the active layer 3 by reducing the thickness of the n-side barrier layer 31, light confinement in the active layer 3 can be enhanced.

  In this way, by suppressing the shielding of the localized level that causes the short wavelength shift, it is possible to realize the semiconductor laser element 100 that performs laser oscillation in a long wavelength region (for example, an oscillation wavelength of 530 nm or more). Furthermore, by doping the n-side composition gradient layer 26 with an n-type impurity at a relatively low concentration, it is possible to suppress an increase in voltage due to the provision of the n-side composition gradient layer 26.

Generally, Γg _th , which is the product of the confinement coefficient Γ and the threshold gain g _th , is called a threshold mode gain, and represents an essential gain required for the mode of the entire device. The threshold mode gain is generally expressed by the following model formula. Here, α _i and α _m are the average internal loss and the reflector loss, respectively. For convenience, the mode distribution is shown as an average without considering it.
Γg _th = α _i + α _m

As the current density increases, the mode gain Γg increases. From the above equation, it can be understood that when the gain g increases and reaches the threshold gain g _th , the internal loss and the reflector loss are overcome and laser oscillation is reached. During laser oscillation, a steady state of g = g _th is obtained inside the laser resonator. Since such modes in the steady state gain is monotonically dependent on the carrier density, the carrier density in the above lasing threshold current is clamped at the threshold carrier density N _th. As the injected carrier density is higher, the localized level is shielded, the substantial band gap is likely to be increased, and the laser oscillation wavelength is easily shifted to the short wavelength side. By increasing the confinement factor Γ and reaching the threshold gain g _th at a lower current, both the threshold current density j _th and the threshold carrier density N _th can be lowered. As a result, the injected carrier density is reduced, the blocking of localized levels is suppressed, and laser oscillation can be performed on the longer wavelength side.

  In addition, although suppression of the shielding of a localized level was demonstrated here, it is the same also about suppression of a band filling effect. That is, a short wavelength shift also occurs due to the band filling effect that the effective Ferrance level is separated from the band edge and the effective transition interval is widened by current injection, but by increasing the confinement factor Γ and reducing the threshold carrier density This can also be suppressed.

  On the other hand, when the n-side composition gradient layer 26 is provided, there is a concern about a voltage increase due to generation of fixed charges. In the n-side composition gradient layer 26, a layer having a small band gap energy is grown on a layer having a large band gap energy so that the band gap energy decreases upward. For this reason, a fixed charge is generated in the vicinity of the interface, and the barrier against electrons is increased. Then, since a high voltage is required to inject electrons into the active layer 3 across the barrier, there is a concern about a significant voltage increase. Even if the composition is finely switched in the n-side composition gradient layer 26, it is difficult to completely prevent the generation of fixed charges as long as the lattice constant difference occurs. Therefore, the n-side composition gradient layer 26 contains an n-type impurity. As a result, the fixed charges can be shielded, so that the voltage rise can be suppressed and laser oscillation can be performed at a low driving voltage. Note that by adding an impurity to the n-side composition gradient layer 26, light is absorbed by the impurity, and there is a concern that the laser characteristics may be deteriorated due to an increase in absorption loss. It is possible to prevent deterioration of the laser characteristics while suppressing abnormalities.

  As described above, by providing the n-side composition gradient layer 26, it is possible to suppress a short wavelength shift and cause laser oscillation in a long wavelength region, and it is possible to drive at a low voltage. The reliability of the semiconductor laser device 100 can be improved by keeping the driving voltage low.

  However, when the light confinement strengthening structure is employed in the n-side semiconductor layer 2 as described above, the intensity of light existing in the p-side semiconductor layer 4 is relatively increased. The p-side semiconductor layer 4 has a layer containing a p-type impurity like the electron barrier layer 42, and the loss α due to free carrier absorption increases as the light intensity in the p-type impurity-containing layer increases. An n-type impurity such as Si also absorbs free carriers, but Mg used as a p-type impurity for a nitride semiconductor has a lower activity than Si and therefore needs to be doped with a high concentration of p-type impurity. Since the absorption coefficient due to free carrier absorption is expressed by the product of the carrier concentration N and the absorption cross section σ, the p-side semiconductor layer 4 has a larger absorption loss than the n-side semiconductor layer 2. In particular, in the semiconductor laser device 100 that oscillates a long-wavelength laser beam having a wavelength of 530 nm or more, since the light emitting efficiency of the active layer 3 is not yet sufficient, the increase in the loss α tends to increase the laser oscillation threshold current. Therefore, by adopting at least one of the structures (1) to (3) described above, a p-type impurity-containing layer such as the electron barrier layer 42 while maintaining the effect of enhancing the optical confinement by the n-side composition gradient layer 26. To suppress an increase in light intensity.

  The structures (1) to (3) will be described below using specific examples. 3A and 3B are diagrams showing the refractive index distribution in the active layer 3 of the semiconductor laser device 100 and the vicinity thereof, and FIGS. 4A and 4B are diagrams showing the light intensity distribution. Both are graphs plotting values obtained by calculation. The solid lines in FIGS. 3A and 4A show the refractive index distribution and the light intensity distribution of the example shown in FIG. 2A (that is, examples having the structures (1) and (2)), and the solid lines in FIGS. The refractive index distribution and light intensity distribution of the example shown in 2B (that is, the example having the structure (3)) are shown. In both FIG. 3A and FIG. 3B, the broken line is for comparison in which the refractive index of the p-side composition gradient layer 41 is symmetric with the n-side composition gradient layer 26 across the active layer 3 and the p-side intermediate layer 45 is not provided. The semiconductor laser device is shown. 4A and 4B, the light intensity is normalized by the maximum intensity, and the left side in the figure is the n-side semiconductor layer 2 and the right side in the figure is the p-side semiconductor layer 4.

  As shown in FIGS. 3A and 4A, the leakage of light to the p-side semiconductor layer 4 can be reduced by making the refractive index of the p-side composition gradient layer 41 lower than the refractive index of the n-side composition gradient layer 26. it can. Since the p-side composition gradient layer 41 is undoped, it may be considered that there is substantially no free carrier absorption. The layers where free carrier absorption occurs are the electron barrier layer 42 and the first p-type semiconductor layer 43. By reducing the leakage of light to the p-side semiconductor layer 4, the light intensity in these layers can be reduced, so that the loss α due to free carrier absorption can be reduced.

  Further, as shown in FIG. 2A, the band gap energy of the p-side composition graded layer 41 is larger than that of the n-side composition graded layer 26, so that it is closer to the p-side semiconductor layer 4 than the semiconductor laser element for comparison. The overflow of electrons can be reduced. Thereby, it is possible to improve carrier injection efficiency into the active layer 3 particularly in a high current region (for example, 1.5 A or more).

  In order to make the refractive index of the p-side composition gradient layer 41 lower than the refractive index of the n-side composition gradient layer 26, the average refractive index of the p-side composition gradient layer 41 is lower than the average refractive index of the n-side composition gradient layer 26. And making the refractive index of the lower end of the p-side composition gradient layer 41 lower than the refractive index of the upper end of the n-side composition gradient layer 26. These correspond to structure (1) and structure (2), respectively. More preferably, both are provided as shown in FIG. 2A. In this specification, the average band gap energy refers to a value obtained by dividing the total value of multiplication of the band gap energy of each layer and the film thickness by the total film thickness. As described above, since the composition gradient layer can be regarded as an assembly of a plurality of layers having different band gap energies, the composition gradient layer multiplies the band gap energy and the film thickness of each layer, and the total value thereof. Is the average band gap energy of the composition gradient layer. The same applies to the average refractive index and the average composition ratio.

  As shown in FIG. 4A, when the structures (1) and / or (2) are used, light leakage to the n-side semiconductor layer 2 side slightly increases and light confinement to the active layer 3 slightly decreases. Since the threshold current increases as these levels increase, in order to maintain the low threshold current, the refractive index difference (band gap energy difference) between the n-side semiconductor layer 2 and the p-side semiconductor layer 4 in the vicinity of the active layer 3. Is preferably small. Specifically, in the structure (1), the difference between the average composition ratio in the same film thickness range as the n-side composition gradient layer 26 in the p-side semiconductor layer 4 and the average composition ratio of the n-side composition gradient layer 26 is calculated. 3% or less is preferable. On the other hand, in order to reduce light leakage to the p-side semiconductor layer 4 side, it is preferable that these differences be 0.05% or more.

In the structure (2), the difference between the In composition ratio at the upper end of the n-side composition gradient layer 26 and the In composition ratio at the lower end of the p-side composition gradient layer is preferably 3% or less. For example, when the composition at the upper end of the n-side composition gradient layer 26 is In _0.06 Ga _0.95 N, the composition at the lower end of the p-side composition gradient layer 41 is InGaN having an In composition ratio of 3% or more. preferable. More preferably, the difference between the In composition ratio at the upper end of the n-side composition gradient layer 26 and the In composition ratio at the lower end of the p-side composition gradient layer 41 is set to 2% or less. On the other hand, in order to reduce light leakage to the p-side semiconductor layer 4 side, the difference between the In composition ratio at the upper end of the n-side composition gradient layer 26 and the In composition ratio at the lower end of the p-side composition gradient layer 41 is It is preferably 0.1% or more.

  As shown in FIG. 2C, the n-side composition gradient layer 26 may have a first portion 26c and a second portion 26d in order from the active layer 3 side. The composition change rate of the first portion 26c is higher than the composition change rate of the second portion 26d. That is, in the n-side composition gradient layer 26, the side close to the active layer 3 has a relatively high refractive index. Thereby, since the light intensity of the active layer 3 and its vicinity tends to rise, the deviation of the light intensity peak from the active layer 3 can be reduced. The p-side composition gradient layer 41 may include a third portion 41c and a fourth portion 41d in order from the active layer 3 side. The composition change rate of the third portion 41c is higher than the composition change rate of the fourth portion 41d. That is, in the p-side composition gradient layer 41, the side close to the active layer 3 is set to a relatively high refractive index, and the side far from the active layer 3 is set to a low refractive index. Thereby, the light leakage to the p-side semiconductor layer 4 side can be reduced. These may be combined as shown in FIG. 2C. Further, by changing the composition change rate of the n-side composition gradient layer 26 and / or the p-side composition gradient layer 41 in this way, the In composition ratio at the upper end of the n-side composition gradient layer 26 and the p-side composition gradient layer The structure (1) can be obtained after making the In composition ratio at the lower end substantially the same.

  As shown in FIGS. 3B and 4B, by adopting the structure (3), that is, by providing the p-side intermediate layer 45, the peak of light intensity can be kept away from the p-type impurity-containing layer such as the electron barrier layer. . As a result, loss α due to free carrier absorption in the layer containing the p-type impurity can be reduced. Further, by providing the p-side intermediate layer 45 having a large band gap energy, it is possible to reduce the overflow of electrons to the p-side semiconductor layer 4 side as compared with the semiconductor laser device for comparison. Thereby, it is possible to improve the carrier injection efficiency into the active layer 3 particularly in a high current region. Since the structure (3) is a structure in which the p-type impurity-containing layer is kept away from the peak of light intensity as shown in FIGS. 3B and 4B, the free carriers in the p-side semiconductor layer 4 than the structures (1) and (2). Easy to reduce absorption loss. For this reason, it is easier to improve the light output than the structures (1) and (2), and it is easy to reduce the threshold current.

  In order to more efficiently move the light intensity peak away from the p-type impurity-containing layer such as the electron barrier layer 42, the total film thickness of the p-side composition gradient layer 41 and the p-side intermediate layer 45 is equal to that of the n-side composition gradient layer 26. It is preferably larger than the film thickness. Thereby, the loss α due to free carrier absorption can be reduced, and the light output can be improved. Since the light leakage toward the p-side semiconductor layer 4 can be reduced by increasing the thickness of the p-side intermediate layer 45, the thickness of the p-side intermediate layer 45 is the thickness of the p-side composition gradient layer 41. It is preferable that it is 1/10 or more of, or 20 nm or more. On the other hand, when the p-side intermediate layer 45 is thickened, it is considered that the overflow of electrons increases, the efficiency of carrier injection into the active layer 3 decreases, and the slope efficiency decreases. Therefore, in order to improve the slope efficiency, the thickness of the p-side intermediate layer 45 is preferably 400 nm or less, and more preferably 300 nm or less. In order to reduce light leakage to the p-side semiconductor layer 4 side, the thickness of the p-side intermediate layer 45 is preferably 10 nm or more, and more preferably 100 nm or more. Since the p-side intermediate layer 45 is made of, for example, GaN, there is an advantage that light leakage to the p-side semiconductor layer 4 side can be reduced while suppressing an increase in voltage.

  Although it is possible to combine the structure (1) and / or the structure (2) and the structure (3) in one semiconductor laser element 100, the internal light intensity distribution is widened and the light confinement in the active layer 3 is reduced. Therefore, the structure (3) is preferably not combined with the structures (1) and (2). That is, when the p-side intermediate layer 45 is provided, the p-side composition gradient layer 41 has substantially the same thickness as the n-side composition gradient layer 26 and is symmetrical to the n-side composition gradient layer 26 with the active layer 3 interposed therebetween. A composition ratio is preferred.

  Hereinafter, each member will be described in detail.

(Semiconductor laser element 100)
As shown in FIG. 2, the semiconductor laser device 100 includes a substrate 1, an n-side semiconductor layer 2, an active layer 3, and a p-side semiconductor layer 4 provided above the substrate 1. For example, a ridge 4 a is provided on the upper side of the p-side semiconductor layer 4. A portion of the active layer 3 immediately below the ridge 4a and its vicinity are an optical waveguide region. An insulating film 5 can be provided on the side surface of the ridge 4a and the surface of the p-side semiconductor layer 4 continuous from the side surface of the ridge 4a. The substrate 1 is made of, for example, an n-type semiconductor, and an n electrode 8 is provided on the lower surface thereof. A p-electrode 6 is provided in contact with the upper surface of the ridge 4a, and a p-side pad electrode 7 is provided thereon. The semiconductor laser element 100 can oscillate laser light in a long wavelength region, and can oscillate laser light having a wavelength of 530 nm or more, for example.

(Substrate 1)
For the substrate 1, for example, a nitride semiconductor substrate made of GaN or the like can be used. Examples of the n-side semiconductor layer 2, the active layer 3, and the p-side semiconductor layer 4 grown on the substrate 1 include semiconductors grown substantially in the c-axis direction. For example, each semiconductor layer can be grown on the c-plane using a GaN substrate having a c-plane ((0001) plane) as a main surface. Here, the c-plane as the main surface may include a surface having an off angle of about ± 1 degree. By using a substrate having a c-plane as a main surface, an advantage of excellent mass productivity can be obtained.

(N-side semiconductor layer 2)
The n-side semiconductor layer 2 can have a multilayer structure made of a nitride semiconductor such as GaN, InGaN, or AlGaN. Examples of the n-type semiconductor layer included in the n-side semiconductor layer 2 include a layer made of a nitride semiconductor containing an n-type impurity such as Si or Ge. For example, the n-side semiconductor layer 2 includes, in order from the substrate 1 side, a first n-type semiconductor layer 21, a second n-type semiconductor layer 22, a third n-type semiconductor layer 23, a fourth n-type semiconductor layer 24, a fifth n-type semiconductor layer 25, The n-side composition gradient layer 26 is included.

(First to fifth n-type semiconductor layers 21 to 25)
The first to fifth n-type semiconductor layers 21 to 25 contain an n-type impurity. The first n-type semiconductor layer 21 is made of, for example, AlGaN. For example, the second n-type semiconductor layer 22 is a layer having a larger band gap energy than the first n-type semiconductor layer 21. The second n-type semiconductor layer 22 is made of, for example, AlGaN. The third n-type semiconductor layer 23 is made of, for example, InGaN, and its In composition ratio is smaller than that of the well layers 32A and 32B. The fourth n-type semiconductor layer 24 is, for example, a layer having a larger band gap energy than the first n-type semiconductor layer 21 and may be the same as the second n-type semiconductor layer 22. The fourth n-type semiconductor layer 24 is made of, for example, AlGaN. Either one or both of the second n-type semiconductor layer 22 and the fourth n-type semiconductor layer 24 may have the maximum band gap energy in the n-side semiconductor layer 2, and typically function as an n-type cladding layer. The third n-type semiconductor layer 23 can function as a crack prevention layer. In this case, the film thickness is preferably smaller than both the second n-type semiconductor layer 22 and the fourth n-type semiconductor layer 24. The fifth n-type semiconductor layer 25 preferably has a band gap energy smaller than that of the fourth n-type semiconductor layer 24 and equal to or greater than the lower end of the n-side composition gradient layer 26. For example, it is made of GaN. The n-type impurity concentration of the fifth n-type semiconductor layer 25 is preferably larger than that of the n-side composition gradient layer 26.

(N-side composition gradient layer 26)
The n-side composition gradient layer 26 is a layer whose composition is changed stepwise so that the band gap energy decreases from the second surface 26b toward the first surface 26a. That is, in the n-side composition gradient layer 26, the band gap energy decreases stepwise from the second surface 26b toward the first surface 26a. Generally, a photon confinement structure can be formed by discontinuity of the refractive index at the semiconductor interface. By making the n-side composition graded layer 26 a layer whose composition is changed stepwise so that the refractive index n increases as it approaches the active layer 3, the barrier of the optical waveguide continues to the n-side composition graded layer 26. Formed. As a result, light confinement in the active layer 3 can be enhanced.

In the n-side composition gradient layer 26, the upper side (that is, the first surface 26a side) is In _a Ga _1-a N (0 <a <1), and the lower side (that is, the second surface 26b side) is In _b Ga _1. It is preferable that _−bN (0 ≦ b <a). Since the well layers 32A and 32B are typically InGaN, the termination on the first surface 26a side is preferably InGaN having a smaller In composition ratio than the well layers 32A and 32B. The termination on the second surface 26b side is, for example, GaN. The n-side composition gradient layer 26 is preferably made to function as an n-side light guide layer. The In composition ratio a closest to the first surface 26a of the n-side composition gradient layer 26 is preferably 0.01 or more, and more preferably 0.03 or more. Thus, by disposing InGaN near the well layers 32A and 32B, light confinement in the well layers 32A and 32B can be improved. The upper limit value of the In composition ratio a is, for example, 0.25. In consideration of suppression of deterioration of crystallinity, the In composition ratio a is preferably 0.1 or less.

Since the n-side composition gradient layer 26 is a layer whose composition is changed stepwise, as shown in FIG. 5A, when it is composed of a plurality of sub-layers 261 to 265 made of In _x Ga _1-x N having different compositions. It can also be said. FIG. 5A is a partially enlarged view of the n-side composition gradient layer 26 and the vicinity thereof, and shows that there are many sublayers other than those explicitly shown between the sublayer 263 and the sublayer 264. In such a laminated structure, it is preferable that the lattice constant difference between adjacent sub-layers is small. As a result, distortion can be reduced and the amount of fixed charges generated can be reduced. For this reason, it is preferable that the composition of the n-side composition gradient layer 26 is gradually changed with a small thickness. Specifically, in the n-side composition gradient layer 26, it is preferable that the In composition ratio decreases for each film thickness of 25 nm or less from the first surface 26a to the second surface 26b. That is, it is preferable that the film thickness of each of the sublayers 261 to 265 is 25 nm or less. Furthermore, the film thickness of each of the sublayers 261 to 265 is preferably 20 nm or less. The lower limit of the film thickness of each of the sublayers 261 to 265 is, for example, about one atomic layer (about 0.25 nm). Further, the difference in In composition ratio between adjacent sub-layers (for example, sub-layer 261 and sub-layer 262) is preferably 0.005 or less. More preferably, it is 0.001 or less. The lower limit is, for example, about 0.00007.

Such a range is preferably satisfied over the entire n-side composition gradient layer 26. That is, it is preferable that all the sublayers are within such a range. For example, in the 230 nm-thick n-side composition graded layer 26, when the lowermost side is GaN and the uppermost side is In _0.065 Ga _0.935 N, the composition is grown under manufacturing conditions that gradually change in 120 steps. Let The number of times the composition changes in the n-side composition gradient layer 26 is preferably about 90 times or more. More preferably, it is about 120 times or more.

  By making the composition change rate of the n-side composition gradient layer 26 (that is, the difference in the composition ratio of adjacent sublayers) substantially constant in the vertical direction, the distribution of fixed charges can be made substantially uniform. . As will be described later, the n-side composition is sufficient as long as the effect of suppressing the voltage rise is obtained by setting the n-type impurity concentration to a substantially constant n-type impurity concentration from the second surface 26b to the first surface 26a of the n-side composition gradient layer 26. The composition change rate of the inclined layer 26 may vary somewhat. For example, in Example 1 described later, the change rate of the In composition ratio fluctuates between 0.0002 and 0.0009 and falls within the range of 0.001 or less. Note that the flow rate of the raw material gas described here indicates a set value of the manufacturing apparatus. Each numerical value may be adjusted according to the specifications of the manufacturing apparatus.

An n-type impurity is added to the n-side composition gradient layer 26. In the n-side composition gradient layer 26 having a structure in which the composition is changed, it is difficult to avoid the generation of fixed charges even if the composition change rate is reduced. Therefore, the fixed charge is shielded by adding an n-type impurity. The lower limit of the n-type impurity concentration of the n-side composition gradient layer 26 is preferably such that the voltage increase can be sufficiently suppressed, and specifically, is preferably greater than 5 × 10 ¹⁷ / cm ³ . When the n-type impurity concentration of the n-side composition gradient layer 26 is 5 × 10 ¹⁷ / cm ³ or less, there is a concern that the fixed charge generated in the n-side composition gradient layer 26 cannot be sufficiently shielded and the voltage increases. Further, according to the band calculation, the higher the n-type impurity concentration, the lower the barrier height. By setting the concentration to 5 × 10 ¹⁷ / cm ³ , the above comparative semiconductor laser element (instead of the composition gradient layer) It is almost the same height as that provided with one composition layer. Therefore, it is preferable that the n-type impurity concentration exceeds 5 × 10 ¹⁷ / cm ³ . Thereby, a drive voltage can be reduced rather than the case where a single composition layer is used. The n-type impurity concentration is more preferably 7 × 10 ¹⁷ / cm ³ or more, and may be 8 × 10 ¹⁷ / cm ³ or more in consideration of a deviation at the time of manufacture, and further may be 1 × 10 ¹⁸ / cm ³ or more. Good.

On the other hand, the n-type impurity concentration of the n-side composition gradient layer 26 is preferably 2 × 10 ¹⁸ / cm ³ or less. When the n-type impurity concentration increases, there is a concern that the absorption loss due to free carriers increases and the light output decreases, but this can be suppressed by setting it to the upper limit value or less. When the n-type impurity concentration of the n-side composition gradient layer 26 is about 2 × 10 ¹⁸ / cm ³ , the optical output is considered to decrease to the same extent as when the composition gradient layer is undoped. The n-type impurity concentration of the n-side composition gradient layer 26 may be less than 2 × 10 ¹⁸ / cm ³ . Further, if the n-type impurity concentration becomes excessive (for example, 1 × 10 ¹⁹ / cm ³ or more), the flatness of the surface of the semiconductor layer tends to deteriorate. Therefore, from this viewpoint, the n-type impurity concentration may not be too high. preferable.

An image of the relationship between the Si concentration (n-type impurity concentration) and the free carrier absorption loss in the n-side composition gradient layer 26 is shown in FIG. The numerical values shown in FIG. 6 are obtained by calculation. The loss α due to free carrier absorption in the impurity-containing layer depends on the carrier density. That is, as the carrier density increases, the loss due to free carrier absorption increases. Since the carrier density increases as the Si concentration increases, it is considered that as the carrier density increases as the Si concentration increases, the loss increases and the light output tends to decrease. As shown in FIG. 6, when the Si concentration in the n-side composition gradient layer 26 exceeds 2 × 10 ¹⁸ / cm ³ , it is considered that the absorption loss further increases and the light output decreases more remarkably. Note that it is considered that the Si concentration for voltage reduction is about 1 × 10 ¹⁸ / cm ³ .

  The n-type impurity concentration of the n-side composition gradient layer 26 is preferably substantially constant from the first surface 26a to the second surface 26b. In order to suppress the voltage rise, it is sufficient to add an n-type impurity to a portion where a fixed charge is generated. However, in the n-side composition gradient layer 26 in which the composition is finely changed as described above, fixed charge generation is performed. It is difficult to add an n-type impurity by accurately aiming at a place, that is, a composition change place. Accordingly, it is preferable that the n-side composition gradient layer 26 is substantially constant from one end to the other end in the thickness direction so that the n-type impurity can be reliably added to the fixed charge generation site. In addition, it is preferable that the composition change rate of the n-side composition gradient layer 26 is also substantially constant. That is, it is preferable to change the composition so that the band gap energy substantially monotonously decreases from the second surface 26b to the first surface 26a. In the case of the n-side composition gradient layer 26 having a substantially constant composition change rate, the distribution of the generated fixed charges becomes substantially uniform. For this reason, the effect of suppressing the voltage rise can be obtained substantially uniformly over the entire thickness direction by containing the n-type impurity at a substantially constant concentration. In order to obtain a substantially constant n-type impurity concentration, for example, when the n-side composition gradient layer 26 is grown, the set value of the flow rate of the n-type impurity supply source (for example, silane gas) is set from the beginning of the growth. Constant until the end. Deviation during manufacturing is allowed. In addition, even if the flow rate of the n-type impurity source gas is constant, the concentration of the n-type impurity actually taken in may vary somewhat due to a change in the flow rate of the In source gas. Also in this case, it can be said that the n-type impurity concentration of the n-side composition gradient layer 26 is substantially constant.

  The n-side composition gradient layer 26 is made thicker than the n-side barrier layer 31 in order to obtain the effect of confining light. Specifically, the film thickness of the n-side composition gradient layer 26 is preferably 200 nm or more. Further, the film thickness of the n-side composition gradient layer 26 is preferably 500 nm or less, and more preferably 300 nm or less. In order to obtain a light confinement effect, the n-side composition gradient layer 26 is preferably disposed near the active layer 3. Specifically, the distance from the well layer 32A of the n-side composition gradient layer 26 is preferably 20 nm or less.

(Active layer 3)
The active layer 3 can have a multilayer structure composed of nitride semiconductor layers such as GaN and InGaN. The active layer 3 has a single quantum well structure or a multiple quantum well structure. In order to obtain a sufficient gain, a multiple quantum well structure is preferable. The active layer 3 having a multiple quantum well structure includes a plurality of well layers 32A and 32B and an intermediate barrier layer 33 sandwiched between the well layers 32A and 32B. For example, the active layer 3 includes an n-side barrier layer 31, a well layer 32A, an intermediate barrier layer 33, a well layer 32B, and a p-side barrier layer 34 in this order from the n-side semiconductor layer 2 side.

  For the n-side barrier layer 31, the intermediate barrier layer 33, and the p-side barrier layer 34, a semiconductor having a larger band gap energy than the well layers 32A and 32B is used. The n-side barrier layer 31, the intermediate barrier layer 33, and the p-side barrier layer 34 are preferably made of InGaN or GaN. This is because the In composition ratio of the well layers 32A and 32B increases as the oscillation wavelength becomes longer, so that the difference in lattice constant between them is not increased too much. The In composition ratio of the well layers 32A and 32B in the case of a semiconductor laser element having an oscillation wavelength of 530 nm or more slightly increases or decreases depending on the layer structure other than the active layer 3, but is, for example, 0.25 or more (25% or more). The upper limit of the In composition ratio of the well layers 32A and 32B is, for example, 0.50 or less (50% or less). At this time, it is considered that the oscillation wavelength of the semiconductor laser element is about 600 nm or less. The well layers 32A and 32B are preferably undoped from the viewpoint of improving crystallinity and reducing light absorption. In the present specification, undoped means not intentionally doped. In an analysis result such as SIMS analysis, it may be said that it is undoped if the concentration is below the detection limit.

  The film thickness of the well layers 32A and 32B is, for example, 4 nm or less, and preferably 3 nm or less. A layer having a thickness (for example, 1 nm or less) smaller than the thickness of each barrier layer 31, 33, 34 may be disposed between each barrier layer 31, 33, 34 and the well layers 32A, 32B. . In other words, the barrier layers 31, 33, and 34 and the well layers 32A and 32B approach each other so that the shortest distance is less than the film thickness of each barrier layer 31, 33, and 34 (for example, 1 nm or less, including 0). It is preferable that they are arranged.

(N-side barrier layer 31)
The n-side barrier layer 31 is a layer having a higher n-type impurity concentration than the n-side composition gradient layer 26 and a larger band gap energy than the first surface 26 a side of the n-side composition gradient layer 26. With the band gap energy having such a relationship, electrons can be stored in the vicinity of the first surface 26 a of the n-side composition gradient layer 26. Thereby, even if the n-side barrier layer 31 is thinned, hole overflow can be suppressed. Since the n-side barrier layer 31 has a high n-type impurity concentration, there is a concern that the localized level of the well layer closest to the n-side barrier layer 31 may be shielded. Wavelength can be achieved. Although free carrier absorption is caused by n-type impurities and absorption loss is increased, this can also be suppressed by thinning, so that the light output can be improved. In order to collect electrons in the vicinity of the first surface 26a of the n-side composition gradient layer 26, the band gap energy closest to the first surface 26a of the n-side composition gradient layer 26 (that is, the minimum band gap in the n-side composition gradient layer 26). Energy) is preferably smaller than that of the intermediate barrier layer 33.

Specifically, the thickness of the n-side barrier layer 31 is preferably 20 nm or less. Furthermore, the film thickness of the n-side barrier layer 31 is preferably 15 nm or less, more preferably 6 nm or less. The n-type impurity concentration of the n-side barrier layer 31 is preferably 1 × 10 ¹⁹ / cm ³ or more. More preferably, it is set to 3 × 10 ¹⁹ / cm ³ or more. Since the n-side barrier layer 31 is the first barrier layer for electrons in the active layer 3, it is preferable to contain an n-type impurity at such a high concentration to reduce the barrier against electrons. In addition, by increasing the n-type impurity concentration of the n-side barrier layer 31, hole overflow can be further suppressed. Since the n-side barrier layer 31 is provided near the well layer 32A, the upper limit of the n-type impurity concentration is preferably such that the crystallinity does not deteriorate. In order to suppress deterioration of characteristics, the n-type impurity concentration of the n-side barrier layer 31 is preferably 5 × 10 ¹⁹ / cm ³ or less. In addition, the n-side barrier layer 31 is preferably GaN so that the deterioration of crystallinity can be suppressed.

  An intervening layer may be provided between the n-side barrier layer 31 and the nearest well layer 32A. When the intervening layer is provided, the shortest distance between the n-side composition gradient layer 26 and the nearest well layer 32A is 200 nm or less. Further, since the n-side barrier layer 31 is likely to be a barrier against electron injection when the n-side barrier layer 31 is thick, the thickness of the n-side barrier layer 31 is preferably 10 nm or less. In addition, the film thickness of each layer shall be 1 atomic layer or more.

(P-side barrier layer 34)
When the p-side composition gradient layer 41 is disposed, p having a band gap energy larger than the minimum band gap energy of the p-side composition gradient layer 41 between the p-side composition gradient layer 41 and the well layers 32A and 32B. The side barrier layer 34 is preferably disposed. If an n-type impurity is included in the p-side barrier layer 34, there is a risk of light absorption or hole trapping, and Mg, which is a p-type impurity, creates a deep level and causes light absorption. 34 is preferably undoped. For example, the p-side barrier layer 34 is made of undoped GaN.

(P-side semiconductor layer 4)
The p-side semiconductor layer 4 can have a multilayer structure made of a nitride semiconductor layer such as GaN, InGaN, or AlGaN. Examples of the p-type nitride semiconductor layer included in the p-side semiconductor layer 4 include a layer made of a nitride semiconductor containing a p-type impurity such as Mg.

(P-side composition gradient layer 41)
The p-side composition gradient layer 41 has a band gap energy that increases upward. In other words, the p-side composition gradient layer 41 has a third surface 41a on the active layer 3 side and a fourth surface 41b on the electron barrier layer 42 side, and the band gap energy is from the third surface 41a to the fourth. It increases toward the surface 41b. The band gap energy on the third surface 41a side is smaller than that on the fourth surface 41b side. That is, in the p-side composition gradient layer 41, the band gap energy increases stepwise from the third surface 41a toward the fourth surface 41b. By providing not only the n-side composition gradient layer 26 but also the p-side composition gradient layer 41, light can be confined in a balanced manner from both sides with respect to the active layer 3. Thereby, the electric field strength in the active layer 3 can be increased, and the threshold current can be reduced.

The p-side composition gradient layer 41 is undoped, or the p-type impurity concentration of the p-side composition gradient layer 41 is 5 × 10 ¹⁷ / cm ³ or less. For the same reason as the p-side barrier layer 34, the p-side composition gradient layer 41 is preferably undoped. The p-side barrier layer 34 functions as, for example, a p-side light guide layer. A preferable range of the composition, composition change rate, and film thickness of the p-side composition gradient layer 41 is the same as that of the n-side composition gradient layer 26 on the assumption that any one or more of the above structures (1) to (3) is satisfied. Can be adopted. The film thickness of the p-side composition gradient layer 41 is preferably 350 nm or less, and more preferably 300 nm or less. The third surface 41a side of the p-side composition gradient layer 41 preferably has a smaller band gap energy than the p-side barrier layer 34, and is made of, for example, InGaN. The fourth surface 41b side of the p-side composition gradient layer 41 may have a band gap energy equal to or higher than that of the p-side barrier layer 34, and is made of, for example, GaN. Since the p-side composition gradient layer 41 suppresses electron overflow while bringing light to the active layer 3, it is effective to decrease the In composition substantially monotonically from the active layer 3 side.

Similarly to the n-side composition gradient layer 26, the p-side composition gradient layer 41 can be said to be composed of a plurality of sub-layers 411 to 415 made of In _y Ga _1-y N having different compositions as shown in FIG. 5B. FIG. 5B is a partially enlarged view of the p-side composition gradient layer 41 and the vicinity thereof, and shows that there are many sublayers other than those explicitly shown between the sublayer 413 and the sublayer 414. For the same reason as the n-side composition gradient layer 26, it is preferable that the p-side composition gradient layer 41 has an In composition ratio that decreases every 25 nm or less from the third surface 41a to the fourth surface 41b. That is, it is preferable that the film thickness of each sublayer 411-415 is 25 nm or less. Furthermore, it is preferable that the film thickness of each sublayer 411-415 is 20 nm or less. The lower limit value of the film thickness of each of the sublayers 411 to 415 is, for example, about one atomic layer (about 0.25 nm). Further, the difference in In composition ratio between adjacent sublayers (for example, the sublayer 411 and the sublayer 412) is preferably 0.005 or less. More preferably, it is 0.001 or less. The lower limit is, for example, about 0.00007.

(Electron barrier layer 42)
The electron barrier layer 42 contains a p-type impurity such as Mg. The electron barrier layer 42 is made of, for example, AlGaN. The electron barrier layer 42 may be provided as a layer having the highest band gap energy in the p-side semiconductor layer 4 and having a smaller film thickness than the p-side composition gradient layer 41.

(First p-type semiconductor layer 43, second p-type semiconductor layer 44)
The first p-type semiconductor layer 43 and the second p-type semiconductor layer 44 contain a p-type impurity such as Mg. The first p-type semiconductor layer 43 is made of, for example, AlGaN. The first p-type semiconductor layer 43 functions as, for example, a p-type cladding layer, and may have the highest band gap energy next to the electron barrier layer 42 in the p-side semiconductor layer 4. The film thickness of the first p-type semiconductor layer 43 is larger than that of the electron barrier layer 42. The second p-type semiconductor layer 44 is made of, for example, GaN and functions as a p-type contact layer.

(Insulating film 5, n-electrode 8, p-electrode 6, p-side pad electrode 7)
The insulating film 5 can be formed of, for example, a single layer or a stacked layer of oxides or nitrides such as Si, Al, Zr, Ti, Nb, and Ta. The n electrode 8 is provided, for example, in almost the entire lower surface of the n-type substrate 1. The p electrode 6 is provided, for example, on at least the upper surface of the ridge 4a. When the width of the p electrode 6 is narrow, a p-side pad electrode 7 wider than the p electrode 6 may be provided on the p electrode 6, and a wire or the like may be connected to the p-side pad electrode 7. The material of each electrode is, for example, a metal or alloy such as Ni, Rh, Cr, Au, W, Pt, Ti, and Al, or a simple oxide such as a conductive oxide containing at least one selected from Zn, In, and Sn. A layer film or a multilayer film is mentioned. Examples of the conductive oxide include ITO (Indium Tin Oxide), IZO (Indium Zinc Oxide), GZO (Gallium-doped Zinc Oxide), and the like. The thickness of an electrode should just be the thickness which can function as an electrode of a semiconductor element normally. For example, about 0.1 μm to 2 μm can be mentioned.

Example 1
As Example 1, the following semiconductor laser device was fabricated. An MOCVD apparatus was used for producing an epitaxial wafer to be a semiconductor laser element. The raw materials include trimethylgallium (TMG), triethylgallium (TEG), trimethylaluminum (TMA), trimethylindium (TMI), ammonia (NH ₃ ), silane gas, bis (cyclopentadienyl) magnesium (Cp ₂ Mg). ) Was used as appropriate.

An Al _0.02 Ga _0.98 N layer (first n-type semiconductor layer 21) containing Si was grown to a thickness of 1.0 μm on the c-plane GaN substrate (substrate 1).
Next, an Al _0.08 Ga _0.92 N layer (second n-type semiconductor layer 22) containing Si was grown to a thickness of 250 nm.
Next, an In _0.04 Ga _0.96 N layer (third n-type semiconductor layer 23) containing Si was grown to a thickness of 150 nm.
Next, an Al _0.08 Ga _0.92 N layer (fourth n-type semiconductor layer 24) containing Si was grown to a thickness of 650 nm.
Next, a GaN layer containing Si (the fifth n-type semiconductor layer 25) was grown to a thickness of 300 nm.
Next, a composition-graded layer (n-side composition gradient layer 26) doped with Si at a concentration of about 1 × 10 ¹⁸ / cm ³ was grown to a thickness of 260 nm. In the composition gradient layer, the growth start is GaN, the growth termination is In _0.05 Ga _0.95 N, and the In composition is substantially monotonically increased in 120 steps so that the composition gradient is almost linear. Grown up. That is, the composition gradient layer is set so that the In composition ratio increases substantially 0.02% to 0.09% (average 0.04%) every time it grows with a thickness of 2.1 nm. Grew.
Next, a GaN layer (n-side barrier layer 31) doped with Si at a concentration of about 3 × 10 ¹⁹ / cm ³ was grown to a thickness of 3 nm.
Next, an undoped In _0.25 Ga _0.75 N layer (well layer 32A) was grown to a thickness of 2.7 nm.
Next, an undoped GaN layer (intermediate barrier layer 33) was grown to a thickness of 3.4 nm.
Next, an undoped In _0.25 Ga _0.75 N layer (well layer 32B) was grown to a thickness of 2.7 nm.
Next, an undoped GaN layer (p-side barrier layer 34) was grown to a thickness of 2.2 nm.
Next, an undoped composition gradient layer (p-side composition gradient layer 41) was grown to a thickness of 260 nm. In the composition gradient layer, the growth start is In _0.045 Ga _0.955 N, the growth termination is GaN, and the In composition is substantially monotonically decreased in 120 steps so that the composition gradient is almost linear. Grown up. That is, the composition gradient layer is set so that the In composition ratio is substantially monotonously reduced, that is, the In composition ratio decreases by 0.02 to 0.09% (average 0.04%) every time the film grows with a thickness of 2.1 nm. Grew.
Next, an Al _0.16 Ga _0.84 N layer (electron barrier layer 42) containing Mg was grown to a thickness of 11 nm.
Next, an Al _0.04 Ga _0.96 N layer (first p-type semiconductor layer 43) containing Mg was grown to a thickness of 300 nm.
Next, a GaN layer containing Mg (second p-type semiconductor layer 44) was grown to a thickness of 15 nm.

  Then, the epitaxial wafer on which the above layers are formed is taken out from the MOCVD apparatus, and the ridge 4a, the p-electrode 6, the p-side pad electrode 7, the n-electrode 8 and the like are formed using photolithography, RIE, and sputtering. As a result, a semiconductor laser device 100 was obtained. The semiconductor laser device 100 had a ridge width of 15 μm, a resonator length of 1200 μm, and an element width of 150 μm. The semiconductor laser device 100 according to Example 1 oscillated at about 531 nm.

(Example 2)
As Example 2, after the growth end of the p-side composition gradient layer 41 is In _0.05 Ga _0.95 N and the growth end is grown as GaN, an undoped GaN layer (p-side intermediate layer 45) The semiconductor laser device 100 was fabricated in the same manner as in Example 1 except that the film was grown with a film thickness of 200 nm. The semiconductor laser device 100 according to Example 2 oscillated at about 532 nm.

(Comparative Example 1)
As Comparative Example 1, a semiconductor laser element similar to that of Example 2 was manufactured except that the p-side intermediate layer 45 was not provided.

(Experimental result 1)
The IL characteristics of the semiconductor laser elements of Example 1 and Comparative Example 1 are shown in FIG. 7, and the IV characteristics are shown in FIG. 7 and 8, the solid line indicates the semiconductor laser element 100 of Example 2, and the broken line indicates the semiconductor laser element of Comparative Example 1. As shown in FIGS. 7 and 8, it was confirmed that the semiconductor laser device 100 of Example 1 has improved optical output and the voltage is comparable to that of the semiconductor laser device of Comparative Example 1. . The reason why the light output is improved is that the refractive index of the p-side composition gradient layer is lowered as compared with Comparative Example 1, so that the light leakage to the region doped with the p-type impurity is reduced, whereby the p-side semiconductor layer 4 This is thought to be due to the reduction of free carrier absorption loss. The threshold current slightly increased, but this is thought to be due to a decrease in light confinement in the active layer because the internal light intensity peak slightly shifted to the n-side semiconductor layer side.

(Experimental result 2)
The IL characteristics of the semiconductor laser elements of Example 2 and Comparative Example 1 are shown in FIG. 9, and the IV characteristics are shown in FIG. 9 and 10, the solid line indicates the semiconductor laser element 100 of Example 2, and the broken line indicates the semiconductor laser element of Comparative Example 1. As shown in FIGS. 9 and 10, it was confirmed that the semiconductor laser device 100 of Example 2 had improved optical output and the voltage was comparable to that of the semiconductor laser device of Comparative Example 1. . The reason why the light output is improved is that the distance from the active layer 3 to the electron barrier layer 42 is increased as compared with the comparative example 1, thereby reducing the light leakage to the region doped with the p-type impurity. It is considered that the free carrier absorption loss in the side semiconductor layer 4 can be reduced. In addition, it is considered that the threshold current also decreased because the free carrier absorption loss could be greatly reduced.

DESCRIPTION OF SYMBOLS 100 Semiconductor laser element 1 Substrate 2 N side semiconductor layer 21 1st n type semiconductor layer 22 2nd n type semiconductor layer 23 3rd n type semiconductor layer 24 4th n type semiconductor layer 25 5th n type semiconductor layer 26 n side composition inclination layer 3 Active layer 31 n-side barrier layer 32A, 32B well layer 33 intermediate barrier layer 34 p-side barrier layer 4 p-side semiconductor layer 41 p-side composition gradient layer 42 electron barrier layer 43 first p-type semiconductor layer 44 second p-type semiconductor layer 45 p-side intermediate Layer 4a Ridge 5 Insulating film 6 P electrode 7 P side pad electrode 8 N electrode 26a First surface, 26b Second surface 41a Third surface, 41b Fourth surface 26c First portion, 26d Second portion 41c Third portion, 41d 4th part 261-265, 411-415 sublayer

Claims (13)

A semiconductor laser element having an n-side semiconductor layer, an active layer, and a p-side semiconductor layer, each of which is made of a nitride semiconductor, in order upward.
The n-side semiconductor layer has a band gap energy that decreases upward, and includes n-type impurities at an n-type impurity concentration greater than 5 × 10 ¹⁷ / cm ^{3 and} less than or equal to 2 × 10 ¹⁸ / cm ^3. Having a side composition gradient layer,
The active layer is
A band that is disposed in contact with the n-side composition gradient layer, has a higher n-type impurity concentration and smaller film thickness than the n-side composition gradient layer, and is larger than the band gap energy at the upper end of the n-side composition gradient layer. An n-side barrier layer having gap energy;
A plurality of well layers disposed above the n-side barrier layer, and an intermediate barrier layer sandwiched between the plurality of well layers,
The p-side semiconductor layer is
The band gap energy increases upward, and a p-side composition gradient layer that is undoped;
An electron barrier layer disposed above the p-side composition gradient layer, having a band gap energy larger than that of the intermediate barrier layer and having a p-type impurity concentration larger than that of the p-side composition gradient layer;
Have
A semiconductor laser device having one or more structures selected from the following (1) to (3).
(1) When the film thickness of the n-side composition gradient layer is t, the average band gap energy of the p-side semiconductor layer in the distance range t upward from the interface between the active layer and the p-side semiconductor layer Is larger than the average band gap energy of the n-side composition gradient layer,
(2) The band gap energy at the lower end of the p-side composition gradient layer is larger than the band gap energy at the upper end of the n-side composition gradient layer.
(3) The p-side semiconductor layer is further a p-side intermediate layer that connects the p-side composition gradient layer and the electron barrier layer, and is equal to or higher than a band gap energy at an upper end of the p-side composition gradient layer; The p-side intermediate layer has a band gap energy that is less than the band gap energy of the electron barrier layer and is undoped. Having the structure of (1) and / or the structure of (2),
The n-side composition gradient layer has a first portion and a second portion in order from the active layer side, and a composition change rate of the first portion is lower than a composition change rate of the second portion. 2. The semiconductor laser device according to 1. Having the structure of (1) and / or the structure of (2),
The p-side composition gradient layer has a third portion and a fourth portion in order from the active layer side, and the composition change rate of the third portion is higher than the composition change rate of the fourth portion. 3. The semiconductor laser device according to 1 or 2. Having the structure of (3),
4. The semiconductor laser device according to claim 1, wherein a total film thickness of the p-side composition gradient layer and the p-side intermediate layer is larger than a film thickness of the n-side composition gradient layer. 5.   5. The semiconductor laser device according to claim 1, wherein an n-type impurity concentration of the n-side composition gradient layer is substantially constant in a vertical direction.   The semiconductor laser device according to claim 1, wherein the n-side composition gradient layer has a thickness of 200 nm or more. The n-side composition gradient layer is composed of a plurality of sub-layers composed of In _x Ga _1-x N having different compositions from each other,
The uppermost sublayer of the n-side composition gradient layer is made of In _a Ga _1-a N (0 <a <1),
7. The semiconductor laser device according to claim 1, wherein a lowermost sub-layer of the n-side composition gradient layer is made of In _b Ga _1-b N (0 ≦ b <a).   The semiconductor laser device according to claim 7, wherein the sub layer has a thickness of 25 nm or less.   9. The semiconductor laser device according to claim 7, wherein a difference in In composition ratio between adjacent sub-layers is 0.005 or less.   The semiconductor laser device according to claim 1, wherein a distance of the n-side composition gradient layer from the plurality of well layers is 20 nm or less. The p-side composition gradient layer includes a plurality of sub-layers composed of In _y Ga _1-y N having different compositions from each other,
The lowermost sublayer of the p-side composition gradient layer is made of In _c Ga _1-c N (0 <c <1),
The semiconductor laser device according to claim 1, wherein the uppermost sublayer of the p-side composition gradient layer is made of In _d Ga _1-d N (0 ≦ d <c).   The semiconductor laser device according to claim 11, wherein the sub-layer has a thickness of 25 nm or less.   The semiconductor laser device according to claim 1, wherein the semiconductor laser device can oscillate laser light having a wavelength of 530 nm or more.

Images