JP5612521B2 - ZnO-based semiconductor film manufacturing method and ZnO-based semiconductor light emitting device manufacturing method - Google Patents

Description

  The present invention relates to a ZnO-based semiconductor film manufacturing method and a ZnO-based semiconductor light emitting device manufacturing method.

  Zinc oxide (ZnO) is a direct transition type semiconductor having a band gap energy of 3.37 eV at room temperature, and the exciton binding energy is 60 meV compared to other semiconductors (for example, GaN: 24 meV, ZnSe: 18 meV). It is expected to be used for a high-efficiency ultraviolet light-emitting device that uses an exciton emission process at room temperature. In addition, since the raw material is inexpensive and has a feature that there is little adverse effect on the environment and the human body, there is a possibility that a light-emitting element with high efficiency and low power consumption and excellent environmental performance may be realized.

  The structure of a light-emitting element using ZnO (see, for example, Patent Document 1) includes, for example, an n-type ZnO-based semiconductor layer, a ZnO-based semiconductor light-emitting layer, and a p-type ZnO-based semiconductor layer sequentially stacked above a substrate. In this structure, an n-side electrode is formed on the surface of the semiconductor layer or the back surface of the substrate (when a conductive substrate is used), and a p-side electrode is formed on the surface of the p-type semiconductor layer. The substrate is not particularly limited, but a ZnO-based semiconductor substrate with few lattice mismatches is most desirable.

  As a method for obtaining a p-type ZnO-based semiconductor layer, doping of a V group element into a ZnO-based semiconductor is mainly effective, and among them, doping with N atoms is effective and less toxic. Has been tried. However, it is difficult to obtain a p-type ZnO-based semiconductor layer by N doping.

International Publication No. 2007/018216 Pamphlet

  An object of the present invention is to provide a novel method for producing an N-doped ZnO-based semiconductor film having p-type conductivity, and a novel production of a ZnO-based semiconductor light-emitting device having an N-doped ZnO-based semiconductor layer having p-type conductivity. Is to provide a method.

According to one aspect of the present invention, a Zn source gun, an O radical gun, and an N radical gun are provided, and an Mg source gun is further provided as necessary. A radio frequency is applied to the electrode discharge tube to generate plasma containing N radicals, and a Zn beam is formed above the surface of the single crystal substrate using a crystal manufacturing apparatus using pBN or quartz as the material of the electrodeless discharge tube. ZnO-based semiconductor for growing an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film by simultaneous irradiation with an O radical beam, an N radical beam, and if necessary, an Mg beam In the film manufacturing method, the Zn source gun is formed above the growth surface side of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film as viewed from the normal direction of the substrate. , O la The Calgan, the N radical gun, and the Mg source gun are arranged side by side in the circumferential direction, and the azimuth angle in the beam irradiation direction of the N radical gun and the azimuth angle in the beam irradiation direction of the Zn source gun the angle theta _c of, as well as in the range of _{90 ° ≦ θ c ≦ 270 °} , the radio frequency applied to the electrodeless discharge tube power of the N radical gun sputtered from the electrodeless discharge tube In addition, a ZnO-based semiconductor film manufacturing method is provided in which B or Si is suppressed to a low power so that it is not taken into the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film.

By making the angle θ _c formed by the azimuth angle in the beam irradiation direction of the N radical gun and the azimuth angle in the beam irradiation direction of the Zn source gun into a range of 90 ° ≦ θ _c ≦ 270 °, The radio frequency power applied to the electrodeless discharge tube is such that B or Si sputtered from the electrodeless discharge tube is taken into the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film. Even if the power is suppressed to a low level, it is easy to increase the N concentration in the N _- doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film.

FIG. 1 is a schematic cross-sectional view showing an example of the overall structure of an MBE apparatus. FIG. 2A is a cross-sectional view illustrating a schematic structure example of a source gun, and FIG. 2B is a schematic cross-sectional view illustrating an example of a K-cell type Knudsen cell. FIG. 3 is a cross-sectional view showing a schematic structure example of a radical gun. 4A and 4B are a schematic plan view and a schematic cross-sectional view showing the positional relationship between a crystal growth substrate and a gun installed in each port in an MBE apparatus according to an example, respectively. FIG. 5A is a graph showing the relationship between the RF power applied to the N radical gun and the N concentration in the film in the preliminary experiment 1, and FIG. 5B shows the flux ratio and the N concentration in the film in the preliminary experiment 2. It is a graph which shows the relationship. FIG. 6A is a graph showing the relationship between the arrangement angle θ _c and the N concentration in the film in the first experiment, and FIGS. 6B and 6C respectively show a sample with an arrangement angle θ _c of 45 ° in the first experiment, and It is an AFM image of a 180 ° sample. FIG. 7A is a graph showing the relationship between the arrangement angle θ _c and the N concentration in the film in the second experiment, and FIGS. 7B and 7C show a sample in which the arrangement angle θ _c of the second experiment is 45 °, It is an AFM image of a 180 ° sample. FIG. 8 is a graph showing the relationship between the Zn beam flux intensity and the N concentration in the film in the third experiment. FIG. 9 is a schematic cross-sectional view showing a sample structure of the fourth experiment (and the fifth and sixth experiments). FIG. 10 collectively shows IV characteristics and light emission states (presence / absence of light emission and EL spectrum) in the fourth to sixth experiments. FIG. 11 collectively shows AFM images in the fourth to sixth experiments. FIG. 12 is an emission spectrum of plasma formed by an N radical gun. FIG. 13 is a graph showing a general IV characteristic of a diode.

  First, a molecular beam epitaxy (MBE) apparatus will be described as a crystal manufacturing apparatus for growing a ZnO-based compound semiconductor film. Note that the ZnO-based semiconductor contains at least Zn and O.

  FIG. 1 is a schematic cross-sectional view showing the overall structure of the MBE apparatus. In the vacuum vessel 1, a Zn source gun 2, an Mg source gun 3, an O radical gun 4, and an N radical gun 5 are provided. Hereinafter, the source gun or radical gun may be simply referred to as cancer. The guns 2 to 5 are each installed in a port of the vacuum vessel 1. The arrangement relationship of each gun will be described later with reference to FIGS. 4A and 4B. The arrangement relationship of the guns 2 to 5 shown in FIG. 1 is convenient for showing a schematic cross-sectional structure.

  Each of the Zn source gun 2 and the Mg source gun 3 includes a Knudsen cell that accommodates a solid source of Zn and Mg, and emits a Zn beam and an Mg beam, respectively.

  Each of the O radical gun 4 and the N radical gun 5 includes an electrodeless discharge tube to which, for example, a radio frequency (RF) of 13.56 MHz is applied, and oxygen gas and nitrogen gas are supplied in the electrodeless discharge tube to which RF power is applied. Radicalized to emit O radical beam and N radical beam.

  Oxygen gas is supplied from the oxygen cylinder to the O radical gun 4 via the mass flow controller 4a. Nitrogen gas is supplied to the N radical gun 5 from the nitrogen cylinder via the mass flow controller 5a.

  A stage 6 including a heater is disposed in the vacuum chamber 1, and the stage 6 holds the crystal growth substrate 7. A substrate 7 is disposed on the lower surface of the stage 6, and guns 2 to 5 are disposed below the substrate 7. The beams emitted from the guns 2 to 5 are incident obliquely on the lower surface of the substrate 7. By supplying a desired beam on the surface of the substrate 7, a desired ZnO-based semiconductor film can be grown.

  The MBE apparatus also includes a gun 8 for reflection high-energy electron diffraction (RHEED) and a screen 9 for projecting a RHEED image. From the RHEED image, the surface flatness of the grown crystal film can be evaluated. When the crystal grows two-dimensionally and the surface is flat, the RHEED image shows a streak pattern, and when the crystal grows three-dimensionally and the surface is not flat, the RHEED image shows a spot pattern.

By adding Mg to ZnO to form MgZnO, the band gap can be widened. However, since ZnO has a wurtzite structure and MgO has a rock salt structure, if the Mg composition is too high, phase separation occurs. In Mg _x Zn _1-x O in which the Mg composition of MgZnO is specified as _x , the Mg composition x is preferably 0.6 or less in order to maintain the wurtzite structure. In addition, by including Mg composition x = 0, ZnO to which Mg is not added is also included in the notation Mg _x Zn _1-x O.

  The n-type conductivity of the ZnO-based semiconductor can be obtained without doping an n-type impurity, but an n-type impurity such as Ga can also be doped in order to increase the n-type carrier concentration. If necessary, a source gun such as Ga can be installed in a vacant port of the vacuum vessel 1.

  In order to obtain the p-type conductivity of the ZnO-based semiconductor, for example, N can be doped. However, as will be described in the following preliminary experiment, it is difficult to produce an N-doped p-type ZnO-based semiconductor film.

  FIG. 2A is a cross-sectional view showing a schematic structure of the source gun. The structure of the source gun is the same between the Zn source gun 2 and the Mg radical gun 3.

  A Knudsen cell 22 and a filament heater 23 are disposed in the case 21. In the MBE apparatus exemplified here, as the Knudsen cell 22 of the Zn source gun 2 and the Mg source gun 3, a SUMO cell type Knudsen cell 22S as shown in FIG. 2A is used. As the Knudsen cell 22 of the Zn source gun 2 and the Mg source gun 3, a K cell type Knudsen cell 22K as shown in FIG. 2B may be used.

  A filament heater 23 is disposed outside the Knudsen cell 22 so as to surround the upper and lower portions of the Knudsen cell 22. A solid source 24 is contained in the Knudsen cell 22. FIG. 2A illustrates the Zn source gun 2 and shows zinc particles (purity 7N) as the solid source 24. As the solid source 24 of the Mg source gun 3, for example, magnesium grains having a purity of 6N can be used. When a K-cell type Knudsen cell 22K as shown in FIG. 2B is used, it is used by being filled with pellets of zinc or magnesium.

  By heating the filament heater 23 by energization, the temperature of the Knudsen cell 22 is raised and the solid source 24 is evaporated. The evaporated solid source 24 is emitted as a beam from the tip opening of the Knudsen cell 22. A spacer 25 is arranged between the edge of the case 21 on the beam emission side and the Knudsen cell 22. The MBE device fixing flange 26 attaches the source gun to the vacuum vessel port.

  FIG. 3 is a sectional view showing a schematic structure of the radical gun. The structure of the radical gun is the same for the O radical gun 4 and the N radical gun 5.

  An electrodeless discharge tube 32 and an RF coil 33 are arranged in the case 31. The electrodeless discharge tube 32 is made of, for example, pyrolytic boron nitride (pBN) or quartz. In the MBE apparatus exemplified here, quartz is used as the electrodeless discharge tube material of the O radical gun 4, and pBN is used as the electrodeless discharge tube material of the N radical gun 5. Since oxygen oxidizes pBN, quartz is preferable as the electrodeless discharge tube material of the O radical gun 4. Quartz may be used as the electrodeless discharge tube material of the N radical gun 5.

  A gas supply tube 35 is connected to the electrodeless discharge tube 32 via a connection portion 34. The gas path is indicated by arrows 36. By applying RF power of, for example, 13.56 MHz by the RF coil 33 wound outside the electrodeless discharge tube 32, plasma is formed in the electrodeless discharge tube 32, and the generated plasma is transferred to the electrodeless electrode. The light is emitted from the tip opening of the discharge tube 32 as a beam. In the example shown in FIG. 3, the opening of the electrodeless discharge tube 32 is one hole, but the number and size of the holes can be changed as appropriate. The radical gun is attached to the port of the vacuum vessel by the MBE device fixing flange 37.

  The electrodeless discharge tube 32 of the O radical gun 4 is supplied with, for example, oxygen gas having a purity of 6N, and the generated plasma mainly contains oxygen atom radicals as active species.

  The electrodeless discharge tube 32 of the N radical gun 5 is supplied with, for example, nitrogen gas having a purity of 7N. The generated plasma mainly includes nitrogen atom radicals, nitrogen molecule radicals, nitrogen atom ions, and nitrogen as active species. Molecular ions are included. FIG. 12 shows an emission spectrum of plasma formed by the N radical gun 5. By irradiation with a nitrogen atom radical, the O site of the ZnO-based semiconductor can be replaced with an N atom, whereby a p-type conductive ZnO-based semiconductor can be obtained.

  Of the active species emitted from the N radical gun 5, ions or the like damage the growing film and induce defects. By applying an electric field to the beam passing between the counter electrodes 38 by the counter electrode 38 attached to the tip of the beam output side of the case 31 and bending the ion emission direction, it is difficult to irradiate the growth film with ions. Can do.

  FIG. 4A is a schematic plan view showing a positional relationship between the crystal growth substrate and guns installed at each port of the vacuum vessel. The top view seen from the normal line direction upper direction of the board | substrate 7 is shown. A port and a gun installed in the port are indicated by the same reference numeral.

  The vacuum vessel of the MBE apparatus illustrated here includes eight ports P1 to P8. The ports P <b> 1 to P <b> 8 are arranged side by side in the circumferential direction around the substrate 7. Guns P <b> 1 to P <b> 8 can be installed at each port so that the beam is irradiated onto the lower surface of the substrate 7.

The position of each gun is represented by the azimuth angle in the beam irradiation direction. Hereinafter, the azimuth angle theta _c of the beam irradiation direction, simply referred to as a azimuth or arrangement angles. The azimuth angle of the gun P1 is taken as the azimuth angle reference (θ _c = 0 °). Cancer P2~P8, respectively, the azimuth angle theta _c is 45 °, 90 °, 135 ° , 180 °, 225 °, are arranged in 270 °, and 315 ° and a position.

In the preliminary experiments and the first to sixth experiments described below, the N radical gun 5 was installed at the port P1 (θ _c = 0 °). The Zn source gun 2 was installed at the port P2 (θ _c = 45 °), the port P3 (θ _c = 90 °), the port P4 (θ _c = 135 °), or the port P5 (θ _c = 180 °). . The Mg source gun 3 was installed at the port P6 (θ _c = 225 °), and the O radical gun 4 was installed at the port P8 (θ _c = 315 °).

As will be described below, in a preliminary experiment, an angle (arrangement angle) θ _c formed by azimuth angles of the N radical beam and the Zn beam in the beam irradiation direction was set to 45 °. This is an arrangement angle setting conventionally used when the inventors of the present application form an N-doped ZnO-based semiconductor film.

In the first to sixth experiments, the relationship between the arrangement angle θ _c of the Zn beam with respect to the N radical beam and the N concentration in the N-doped ZnO-based semiconductor film was investigated.

FIG. 4B is a schematic cross-sectional view showing the positional relationship between the crystal growth substrate and guns installed at each port. Gun P1 / N radical gun 5 (θ _c = 0 °) and gun P5 / Zn source gun 2 (θ _c = 180 °) are shown. A polar angle θ _{i in the} beam irradiation direction is defined with respect to the normal direction of the substrate 7. In the MBE apparatus illustrated here, the guns P1 to P8 are installed so that the polar angle θ _{i in the} beam irradiation direction is 25 °.

Next, the VI / II ratio in the growth of Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) crystal will be described. The flux intensity of Zn beam expressed as _{J Zn,} the flux intensity of the Mg beam expressed as _{J Mg,} representing the flux intensity of O radical beam and _{J O.}

A coefficient (Zn adhesion coefficient, Mg adhesion coefficient) indicating the ease of adhesion of Zn and Mg to the O terminal surface of the Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) crystal is _expressed as K _Zn. , K _Mg . A coefficient indicating the ease of O adhesion to the Zn termination surface of Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) crystal (O adhesion coefficient) is defined as K 2 _O.

Zn adhesion coefficient K _Zn and flux strength J _Zn product K _Zn · J _Zn , Mg adhesion coefficient K _Mg and flux strength J _Mg product K _Mg · J _Mg , and O adhesion coefficient K _O and flux strength product _K O · _{J O} with J _O, respectively, Zn atoms adhering per unit time per unit area of the substrate, corresponding to the Mg atom, and the number of O atoms.

K _Zn · _{J Zn} and _K to the sum of Mg · _{J Mg} is the ratio of _{K O · J O (VI /} II ratio) _{K O ·} J O / a _{(K Zn · J Zn + K} Mg · J Mg), flux Define ratio. A case where the flux ratio is larger than 1 is called a VI group rich condition (or O rich condition), a case where the flux ratio is equal to 1 is called a stoichiometric condition, and a case where the flux ratio is smaller than 1 is called a group II rich condition. K _Zn · J _Zn is defined as the Zn beam flux amount, K _Mg · J _Mg is defined as the Mg beam flux amount, and K _O · J _O is defined as the O radical beam flux amount.

For _JZn and _JMg , values obtained using a film thickness monitor or the like can be applied. When the growth temperature (substrate surface temperature) is 850 ° C. or lower, the Zn adhesion coefficient K _Zn and the Mg adhesion coefficient K _Mg can be set to 1.

With respect to K 2 _O · J 2 _O , for example, a ZnO film is actually grown under the known condition of K 2 _Zn · J _Zn (condition of K _Zn = 1 at 850 ° C. or lower), and the growth rate G is obtained. Since the growth rate G of the ZnO growth film can be obtained by the following equation (1), by substituting the obtained growth rate G and K _Zn · J _Zn into the equation, the condition of the set O radical gun (O ₂ The amount of O radical beam flux K _O · J _O at a flow rate, RF power, etc. can be derived.
G = [(K _Zn · J _Zn ) ⁻¹ + (K _O · J _O ) ⁻¹ ] ⁻¹ (1)
Next, a preliminary experiment will be described. In a preliminary experiment, regarding the N concentration in an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film grown by MBE, the relationship between the RF power applied to the N radical gun and the N concentration is examined. The experiment (preliminary experiment 1) investigated and the experiment (preliminary experiment 2) which investigated the relationship between a flux ratio and N concentration were conducted. N concentration was measured by secondary ion mass spectrometry (SIMS).

  In Preliminary Experiment No. 1, a plurality of samples were prepared by changing the RF power applied to the N radical gun under the following conditions. The amount of N radical beam can be increased by increasing the RF power applied to the N radical gun and increasing the plasma density in the electrodeless discharge tube. In the following conditions, the flux ratio (VI / II ratio) is approximately 1, which is almost stoichiometric conditions.

-Substrate: Zn polar plane (+ c plane) ZnO single crystal substrate-Substrate temperature: 700 ° C
・ Oxygen gas supply flow rate to O radical gun: 2 sccm
・ Applied RF power to O radical gun: 300W
O radical beam flux amount (K _O · J _O ): 1.1 × 10 ¹⁵ atoms / (cm ² s)
Zn beam flux intensity (J _Zn ): 9 × 10 ¹⁴ atoms / (cm ² s)
Mg beam flux intensity (J _Mg ): 2 × 10 ¹⁴ atoms / (cm ² s)
・ Nitrogen gas supply flow rate to N radical gun: 2 sccm
Arrangement angle (azimuth angle) of the Zn source gun with respect to the N radical gun θ _c : 45 °
FIG. 5A is a graph showing the relationship between the RF power applied to the N radical gun and the N concentration in the film. The B concentration in the film is also shown. The left vertical axis shows the N concentration, and the right vertical axis shows the B concentration. N concentration is shown as a circle plot, and B concentration is shown as a triangle plot.

The RF power applied to the N radical gun was varied in the range of 50W to 300W. It can be seen that the N concentration in the film increases from the order of 10 ¹⁸ cm ⁻³ to the latter half of the order of 10 ²⁰ cm ^{−3 as the} RF power increases.

However, the electrodeless discharge tube made of pBN used for the N radical gun is sputtered when the applied RF power is high. As a result, B is mixed in the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film, and the crystallinity is deteriorated.

In the sample of this experiment, the RF power applied to the N radical gun was 130 W or more, and B was detected in the film. For example, at an RF power of 200 W, the B concentration reaches the order of 10 ¹⁷ cm ⁻³ . At RF power lower than 130 W (for example, 80 W), the B concentration was below the SIMS detection limit (6 × 10 ¹⁵ cm ⁻³ ), and B was not incorporated into the film. In the O radical gun, for example, 300 W of RF power is applied, but sputtering of the quartz electrodeless discharge tube material (mixing of Si into the film) is not particularly problematic.

  In Preliminary Experiment 2, a plurality of samples were prepared by changing the flux ratio (VI / II ratio) by changing the Zn beam flux intensity under the following conditions.

-Substrate: Zn polar plane (+ c plane) ZnO single crystal substrate-Substrate temperature: 700 ° C
O radical beam flux amount (K _O · J _O ): 1.1 × 10 ¹⁵ atoms / (cm ² s)
Mg beam flux intensity (J _Mg ): 2 × 10 ¹⁴ atoms / (cm ² s)
・ Nitrogen gas supply flow rate to N radical gun: 2 sccm
・ Applied RF power to N radical gun: 80W
Arrangement angle (azimuth angle) of the Zn source gun with respect to the N radical gun θ _c : 45 °
FIG. 5B is a graph showing the relationship between the flux ratio and the N concentration in the film. The flux ratio is changed from 1.24 to 0.44 by changing the Zn beam flux intensity in the range of 6.9 × 10 ¹⁴ atoms / (cm ² s) to 2.3 × 10 ¹⁵ atoms / (cm ² s). The range was changed. It can be seen that the N concentration in the film increases from the order of 10 ¹⁸ cm ^{−3 to the} order of 10 ²⁰ cm ^{−3 as the} Zn beam flux intensity increases (decrease in the flux ratio). In order to obtain an N concentration of 10 ²⁰ cm ⁻³ or more, a II group rich condition with a Zn beam flux of 1.4 × 10 ¹⁵ atoms / (cm ² s) or more and a flux ratio of 0.69 or less is required. ing. The RF power applied to the N radical gun is as low as 80 W, which is a condition that suppresses the incorporation of B by sputtering of the electrodeless discharge tube.

  As can be seen from the result of the preliminary experiment 1, the N concentration in the film can be increased by increasing the RF power of the N radical gun and increasing the plasma density. However, in this method, the electrodeless discharge tube material is sputtered and taken into the growth film due to the increase in RF power.

  Impurity B resulting from the electrodeless discharge tube material acts as a donor in the ZnO-based semiconductor and compensates for p-type carriers, which is not preferable from the viewpoint of obtaining a p-type ZnO-based semiconductor.

  Also. Sputtered particles inhibit the surface migration of constituent elements such as Zn on the crystal growth surface and cause pit formation. When the obtained film is used as a p-type layer of a light-emitting element, such pits cause a leakage current and reduce the light emission intensity.

On the other hand, as can be seen from the result of the preliminary experiment No. 2, the N concentration in the film can be increased by sufficiently setting the flux ratio to the group II rich condition. In this method, the N concentration can be increased even when the RF power applied to the N radical gun is low to some extent, and sputtering of the electrodeless discharge tube material can be suppressed. However, under the II group-rich condition, oxygen radicals to be supplied are insufficient in the first place, so that many oxygen vacancies are likely to be generated in the grown N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6). This oxygen deficiency acts as a donor in the crystal and again compensates for p-type carriers.

In order to obtain good p-type conductivity, the RF power applied to the N radical gun can be suppressed, sputtering of the electrodeless discharge tube can be suppressed, and oxygen vacancies can be suppressed by a flux ratio that does not become excessively Group II rich. Therefore, a technique for producing an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film is desired.

Next, first to sixth experiments conducted for examining such a N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film manufacturing technique will be described. In the description of the first to sixth experiments, the N concentration in the N _- doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film is simply referred to as the N concentration, and RF applied to the N radical gun. The power is sometimes simply referred to as RF power.

In the first to third experiments, an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single film sample was produced. For the single film samples prepared in the first to third experiments, the impurity concentration in the film was confirmed by SIMS, and the surface property was confirmed by an atomic force microscope (AFM).

In the fourth to sixth experiments, a light emitting device including an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film as a p-type semiconductor layer was manufactured. For the device samples prepared in the fourth to sixth experiments, the surface property was confirmed by AFM, the IV characteristics were measured, and the light emission state was confirmed.

  The presence or absence of pits was confirmed from the observation results by AFM. A general IV characteristic of the diode will be described with reference to FIG. The characteristic indicated by the solid line in FIG. 13 is an ideal diode characteristic, and there is no leakage current in both the forward bias characteristic and the reverse bias characteristic. The characteristic indicated by the broken line has a low withstand voltage in the reverse bias characteristic, and in the forward bias characteristic, a minute current starts to flow from a voltage lower than the threshold voltage, and can be said to have a leak current. The characteristic indicated by the dotted line has no rectification and shows a short circuit. Regarding the light emission state, an electroluminescence (EL) spectrum was measured.

  The first experiment will be described. In the first experiment, the dependence of the N concentration on the arrangement angle (azimuth angle) between the N radical gun and the Zn source gun under high RF power conditions was examined.

First, in the MBE apparatus, the + c plane ZnO single crystal substrate was subjected to thermal annealing to clean the substrate surface. Thermal annealing was performed at 900 ° C. for 30 minutes under a high vacuum of 1 × 10 ⁻⁹ Torr.

  Subsequently, the substrate temperature was set to 350 ° C., and a ZnO buffer layer was grown by simultaneously irradiating a Zn beam and an O radical beam on the ZnO substrate. Subsequently, in order to improve the crystallinity of the ZnO buffer layer, the substrate temperature was raised to 800 ° C. and annealing was performed for 20 minutes. The thickness of the ZnO buffer layer was about 30 nm.

Subsequently, the substrate temperature is set to 700 ° C., and the ZnO buffer layer is simultaneously irradiated with a Zn beam, an Mg beam, an O radical beam, and an N radical beam, so that N-doped Mg _x Zn _1-x O (x = 0. 25) Growing layer. The beam irradiation conditions in N-doped Mg _x Zn _1-x O (x = 0.25) layer growth are as follows.

In the irradiation of the Zn beam, 7N purity Zn was used as a solid source, and the flux intensity was set to 9 × 10 ¹⁴ atoms / (cm ² s).

The Mg beam was irradiated with Mg having a purity of 6N as a solid source and a flux intensity of 2 × 10 ¹⁴ atoms / (cm ² s).

The O radical beam was irradiated by introducing pure oxygen gas having a purity of 6N into an electrodeless discharge tube (made of quartz) at 2 sccm, and converting it to plasma with an RF power of 300 W. This condition corresponds to 1.1 × 10 ¹⁵ atoms / (cm ² s) as the amount of O radical beam flux.

  The irradiation of the N radical beam was performed by introducing pure nitrogen gas with a purity of 7N into an electrodeless discharge tube (manufactured by pBN) at 0.5 sccm and converting it into plasma under a high RF power condition of RF power 200W.

The above conditions have a flux ratio (VI / II ratio) of approximately 1, and are almost stoichiometric conditions. The thickness of the N-doped Mg _x Zn _1-x O (x = 0.25) layer was 100 nm.

Under such conditions, the Zn source gun arrangement angle θ _c with respect to the N radical gun was changed to 45 °, 90 °, 135 °, and 180 °, and samples with different arrangement angles θ _c were produced. Hereinafter, the arrangement angle θ _c of the Zn source gun with respect to the N radical gun may be simply referred to as the arrangement angle θ _c .

  As described with reference to FIGS. 1, 4 </ b> A, and 4 </ b> B, in the illustrated crystal manufacturing apparatus, each beam is irradiated from below, so that the apparatus is below the substrate (with respect to the ground). Each layer will be laminated. However, for convenience of description, in the description of the film forming process, the growth surface side is referred to as “upward”. Therefore, in this view, each gun is disposed above the growth surface side with respect to the substrate.

FIG. 6A is a graph showing the relationship between the arrangement angle θ _c and the N concentration in the film in the first experiment. The B concentration in the film is also shown. The left vertical axis shows the N concentration, and the right vertical axis shows the B concentration. N concentration is shown as a circle plot, and B concentration is shown as a triangle plot.

Even when the flux ratio (VI / II ratio) is almost stoichiometric, the RF power of the N radical gun is as high as 200 W, so that a high N concentration of 10 ²⁰ cm ⁻³ or more is obtained. N concentration is not significantly changed even by changing the arrangement angle theta _c. However, since the RF power is high, it is considered that the electrodeless discharge tube made of pBN is sputtered, and B is also doped with the order of 10 ¹⁷ cm ⁻³ regardless of the arrangement angle θ _c .

FIG. 6B and FIG. 6C are AFM images of a sample with an arrangement angle θ _c of 45 ° and a sample with a 180 ° of the first experiment, respectively. An AFM image with an observation range of 15 μm square is shown on the upper side, and an AFM image with an observation range of 1 μm square is shown on the lower side.

Sample arrangement angle theta _c is 45 °, nor samples of 180 °, countless pit occurs on the membrane surface. These pits are considered to be caused by sputtered particles generated by sputtering the electrodeless discharge tube.

  The second experiment will be described. In the second experiment, the dependence of the N concentration on the arrangement angle under low power conditions was examined.

The sample of the second experiment is the same as that of the first experiment, except that the RF power of the N radical gun when the N-doped Mg _x Zn _1-x O (x = 0.25) layer is set to a low power of 80 W. It produced on the same conditions. Samples with different arrangement angles θ _c were prepared by changing the arrangement angle θ _c of the Zn source gun with respect to the N radical gun to 45 °, 90 °, 135 °, and 180 °.

Figure 7A is a graph showing the relationship between the arrangement angle theta _c and film N concentration in the second experiment. The B concentration in the film is also shown. The left vertical axis shows the N concentration, and the right vertical axis shows the B concentration. In addition, it is thought that the graph which shows N density | concentration in a film | membrane when arrangement | positioning angle (theta) _c is 180 degrees-360 degrees becomes the shape which reversed the graph of FIG. 7A at 180 degrees from the symmetry of arrangement angle (theta) _c . .

By reducing the RF power of the N radical gun to 80 W, the doping of B into the film is suppressed to below the SIMS detection lower limit (6 × 10 ¹⁵ cm ⁻³ ) at any arrangement angle θ _c . It is considered that the sputtering in the electrodeless discharge tube was suppressed by lowering the RF power. In addition, when the B concentration is suppressed to the SIMS detection lower limit (6 × 10 ¹⁵ cm ⁻³ ) or less, it can be determined that B is not taken into the film.

On the other hand, the N concentration in the film is reduced as compared with the first experiment, for example, by lowering the RF power. However, the degree of reduction differs by the arrangement angle theta _c of Zn source gun for N radicals cancer, arrangement angle theta _c is 0 ° (360 °) to about N concentration decrease in the film is large near, arrangement angle theta _c is The closer to 180 °, the less. If the arrangement angle θ _c is in the range of 90 ° ≦ θ _c ≦ 270 °, it can be said that an N concentration of 10 ²⁰ cm ⁻³ or more can be obtained.

As described above, by setting the arrangement angle θ _c within the range of 90 ° ≦ θ _c ≦ 270 °, the flux ratio (VI / II ratio) is caused by sputtering of the electrodeless discharge tube even in the vicinity of the stoichiometry condition. It was found that an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film having an N concentration of 10 ²⁰ cm ⁻³ or more can be obtained with such a low RF power that B doping is suppressed.

7B and 7C, respectively, the arrangement angle theta _c is 45 ° samples of the second experiment, an AFM image of a sample of 180 °. An AFM image with an observation range of 15 μm square is shown on the upper side, and an AFM image with an observation range of 1 μm square is shown on the lower side.

Sample arrangement angle theta _c is 45 °, nor samples of 180 °, the pit is small. It is considered that the sputtering of the N radical gun electrodeless discharge tube was suppressed by the low RF power, and the sputtered particles were reduced.

When a light emitting element such as a light emitting diode (LED) is manufactured, the hole carrier density is required to be 1 × 10 ¹⁶ cm ⁻³ or more. N element doped as an acceptor in Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) has a low activation rate, and if it is not doped at least 1 × 10 ¹⁹ cm ⁻³ or more, effective hole carriers are not present. I can't get it. However, if the concentration of N is more than 5 × 10 ²⁰ cm ⁻³ , many defects are generated in the p-type Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film, and the device May cause a leak when a current is passed through. Therefore, the concentration of N is preferably in the range of 1 × 10 ¹⁹ cm ^{−3 to} 5 × 10 ²⁰ cm ⁻³ . The range of 1 × 10 ²⁰ cm ^{−3 to} 3 × 10 ²⁰ cm ⁻³ is more preferable.

  The third experiment will be described. In the third experiment, the flux ratio dependence of N concentration under low power conditions was examined.

Samples of the third experiment, the arrangement angle theta _c of Zn source gun for N radicals cancer and 45 ° and 180 °, produced in the second experiment similar to the low power condition. However, the Zn beam flux intensity (which is equal to the beam flux amount from K _Zn = 1) when forming the N-doped Mg _x Zn _1-x O (x = 0.25) layer is 5 × 10 ¹⁴ atoms. / (Cm ² s) to 2 × 10 ¹⁵ atoms / (cm ² s). Thereby, the flux ratio (VI / II ratio) was changed in the range of 1.6 to 0.5.

FIG. 8 is a graph showing the relationship between the Zn beam flux intensity and the N concentration in the film in the third experiment. Arrangement angle theta _c indicates the result of 45 ° in the plot of the triangular arrangement angle theta _c indicates the result of the 180 ° in the plot of the circle.

Even when the arrangement angle theta _c is even 180 ° In the case of 45 °, with increasing Zn beam flux intensity increased film N concentration (decrease with the flux ratio), in the fully Group II rich conditions side, in both cases Also, the N concentration is 10 ²⁰ cm ⁻³ or more.

However, if the arrangement angle theta _c is 45 °, N concentration as the Group II rich conditions side closer to stoichiometric conditions side is rapidly decreased, the stoichiometric conditions, the order of 10 ¹⁹ cm ^-3 N concentration.

On the other hand, when the arrangement angle θ _c is 180 °, an N concentration of 10 ²⁰ cm ⁻³ or more is obtained in a region where the flux ratio is 1.1 or less, and the VI group rich (O Even under a rich condition, there is a region where the N concentration is 10 ²⁰ cm ⁻³ or more. As a result, high-concentration N-doping is possible under stoichiometric conditions with a flux ratio of 1 or more or group VI-rich (O-rich) conditions, and N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) can be obtained.

In the second experiment, an N concentration of 10 ²⁰ cm ⁻³ or more is obtained under almost stoichiometric conditions when the arrangement angle θ _c is in the range of 90 ° ≦ θ _c ≦ 270 °. Based on this, the relationship between the flux ratio and the N concentration obtained when the arrangement angle θ _c is 180 ° in the third experiment does not hold in a wide range of 90 ° ≦ θ _c ≦ 270 °. It is thought.

The fourth experiment will be described. In the fourth experiment, a single heterojunction type ZnO-based semiconductor light-emitting device was fabricated, and an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film as a p-type layer was applied to an N radical gun. applying RF power as high as 200 W, the arrangement angle theta _c was formed under the conditions of 45 °.

FIG. 9 is a schematic cross-sectional view showing a sample structure of the fourth experiment (and the fifth and sixth experiments). First, thermal annealing was performed on the + c-plane ZnO single crystal substrate 41 having n-type conductivity, and the substrate surface was cleaned. Thermal annealing was performed at 900 ° C. for 30 minutes under a high vacuum of 1 × 10 ⁻⁹ Torr.

  Subsequently, the substrate temperature was set to 350 ° C., and a ZnO buffer layer 42 was grown on the ZnO substrate 41 by simultaneously irradiating a Zn beam and an O radical beam. Subsequently, in order to improve the crystallinity of the ZnO buffer layer 42, the substrate temperature was raised to 800 ° C. and annealing was performed for 20 minutes. The thickness of the ZnO buffer layer 42 was about 30 nm.

Next, the substrate temperature was set to 900 ° C., and a Zn beam and an O radical beam were simultaneously irradiated on the ZnO buffer layer 42 to grow an n-type ZnO layer 43. In the irradiation of the Zn beam, 7N purity Zn was used as a solid source, and the flux intensity was 8 × 10 ¹⁴ atoms / (cm ² s). The O radical beam was irradiated by introducing pure oxygen gas with a purity of 6N into an electrodeless discharge tube at 1 sccm and converting it to plasma with an RF power of 300 W. This condition corresponds to 7.5 × 10 ¹⁴ atoms / (cm ² s) as the amount of O radical beam (K _O · J _O ). The thickness of the n-type ZnO layer 43 was 200 nm.

Next, the substrate temperature was set to 700 ° C., and a ZnO active layer 44 was grown on the n-type ZnO layer 43 by simultaneously irradiating a Zn beam and an O radical beam. In the irradiation of the Zn beam, the flux intensity was set to 1.6 × 10 ¹⁴ atoms / (cm ² s). Irradiation with the O radical beam was performed by introducing pure oxygen gas into an electrodeless discharge tube at 3 sccm and turning it into plasma with an RF power of 300 W. This condition corresponds to 1.2 × 10 ¹⁵ atoms / (cm ² s) as the amount of O radical beam (K _O · J _O ). The thickness of the ZnO active layer 44 was 10 nm.

Subsequently, the ZnO active layer 44 is simultaneously irradiated with a Zn beam, an Mg beam, an O radical beam, and an N radical beam while keeping the substrate temperature at 700 ° C., and an N-doped Mg _x Zn _1-x O is formed as a p-type layer. The (x = 0.25) layer 45 was grown to a thickness of 40 nm. The irradiation conditions of the Zn beam, Mg beam, O radical beam, and N radical beam were the same as in the first experiment. Applying RF power to the N radicals cancer is 200 W, the arrangement angle theta _c of Zn source gun for N radicals cancer was 45 °.

  Next, a titanium layer having a thickness of 10 nm was deposited on the back surface of the ZnO substrate 41, and an aluminum layer having a thickness of 500 nm was deposited on the titanium layer to form an n-side electrode 46. Further, a nickel layer having a thickness of 1 nm is deposited on the p-type layer 45, a gold layer having a thickness of 10 nm is deposited on the nickel layer, and a p-side transparent electrode 47 is formed. A p-side bonding electrode 48 was fabricated by depositing a gold layer having a thickness of 500 nm on a part of the p-side bonding electrode 48. Thereafter, for example, an electrode alloying treatment was performed in an oxygen atmosphere at 400 ° C. for a treatment time of 2 minutes. In this manner, a ZnO-based light emitting diode according to the fourth experiment was fabricated (however, as described later, the light emitting element according to the fourth experiment did not actually emit light).

The fifth experiment will be described. In the fifth experiment, a single heterojunction type ZnO-based semiconductor light-emitting device was fabricated, and an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film as a p-type layer was applied to an N radical gun. The applied RF power was as low as 80 W, and the arrangement angle θ _c was 45 °.

The light emitting device of the fifth experiment is the same as the light emitting device of the fourth experiment except that the N-doped Mg _x Zn _1-x O (x = 0.25) layer 45 and the RF power applied to the N radical gun are set to 80 W. It was produced under the same conditions.

The sixth experiment will be described. In the sixth experiment, a single heterojunction type ZnO-based semiconductor light-emitting device was fabricated, and an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) film serving as a p-type layer was applied to an N radical gun. The applied RF power was as low as 80 W, and the arrangement angle θ _c was 180 °.

In the light emitting device of the sixth experiment, the N-doped Mg _x Zn _1-x O (x = 0.25) layer 45, the RF power applied to the N radical gun is 80 W, and the arrangement angle of the Zn source gun with respect to the N radical gun It was fabricated under the same conditions as the light-emitting element of the fourth experiment except that _θc was 180 °.

  10 and 11 summarize the characteristics of the light-emitting elements obtained in the fourth to sixth experiments. FIG. 10 shows the IV characteristics and the light emission state (the presence or absence of light emission and the EL spectrum), and FIG. 11 shows an AFM image.

  The IV characteristics and light emission state are as follows. The sample of the fourth experiment showed an IV characteristic with a severe current leakage, and did not emit light. Samples of the fifth and sixth experiments showed IV characteristics as diodes and emitted light.

  Although the sample of the fifth experiment was confirmed to emit light, no ultraviolet light emission peak was observed, and white light emission of a broad spectrum having a peak near the wavelength of 580 nm was shown. The sample of the sixth experiment showed ultraviolet light emission having a peak at a wavelength of 380 nm, and also showed white light emission with a broad spectrum having a peak at around a wavelength of 580 nm.

The AFM observation results are as follows. The sample of the fourth experiment has many small pits. This pit is considered to be caused by sputtered particles generated from the N radical gun during the growth of the N-doped Mg _x Zn _1-x O (x = 0.25) layer, and a leak current flows through this pit. It seems that there is. On the other hand, the samples of the fifth and sixth experiments have very few pits. This seems to be the effect of lowering the RF power of the N radical gun.

The samples of the fifth experiment and the sixth experiment showed light emission, and it was confirmed that the N-doped Mg _x Zn _1-x O (x = 0.25) layer was a p-type layer. However, the IV characteristic of the sample of the fifth experiment shows that the current hardly flows and the resistance of the p-type layer is high. This is presumably because the N concentration in the film of the p-type layer is low. P-type layer of the sample in the fifth experiment, the arrangement angle theta _c of Zn source gun is formed under conditions of 45 ° with respect to N radicals cancer. As a result, as described in the second experiment, it is considered that the N concentration in the p-type layer was low.

On the other hand, in the sample of the sixth experiment, the current flows through the p-type layer more easily than the sample of the fifth experiment. The p-type layer of the sample of the sixth experiment is formed under the condition that the arrangement angle θ _c is 180 °, and as described in the second experiment, the N concentration in the film of the p-type layer is on the order of 10 ²⁰ cm ^−3. It is thought that it is getting higher. Furthermore, the RF power applied to the N radical gun is sufficiently low, so that B doping and pit formation are suppressed, and the oxygen vacancies are reduced by making the flux ratio almost stoichiometric. It is done.

From the first to sixth experiments described above, the sputtering of the N radical gun electrodeless discharge tube is suppressed during the growth of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film. Even at a low RF power, the angle (arrangement angle) θ _c formed by the azimuth angle in the beam irradiation direction of the N radical gun and the azimuth angle in the beam irradiation direction of the Zn source gun is 90 ° ≦ θ _c ≦ 270 °. It was found that the N concentration in the film can be increased to, for example, 1 × 10 ²⁰ cm ⁻³ or more by setting the value within the range.

Further, by setting the arrangement angle θ _c within the range of 90 ° ≦ θ _c ≦ 270 °, even if the flux ratio is 1 or more (stoichiometry condition or O-rich condition), the N concentration in the film is, for example, 1 × 10 ^20. It turned out that it can raise to cm ^<-3 > or more.

In the above experiment, the growth temperature of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film was set to 700 ° C., but if the growth temperature is around 700 ° C., the same tendency Can be obtained. The growth temperature of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film could be in the range of 600 ° C. to 800 ° C., for example.

  In the above experiment, an N-doped MgZnO single crystal film to which Mg was added was grown, but it is considered that the same tendency can be obtained for the growth of an N-doped ZnO single crystal film to which no Mg is added. Mg can be added as needed.

In the above experiment, nitrogen gas was introduced into an N radical gun to generate plasma, and N radicals were obtained. However, it seems that the N radical source is not limited to nitrogen gas. As the compound containing a nitrogen atom, for example, nitrogen oxides such as N ₂ O, NO, NO ₂ , N ₂ O ₃ , and N ₂ O ₄ may be used alone or in combination.

  In the above experiment, an electrode made of pBN was used as the electrodeless discharge tube of the N radical gun, but a product made of quartz can also be used. Even when a quartz electrodeless discharge tube is used, if the RF power is too high, the inside of the electrodeless discharge tube is sputtered, resulting in pit formation by sputtered particles and compensation of acceptors by Si doping in the film.

When a quartz electrodeless discharge tube is used for the N radical gun, it is preferable to suppress the RF power applied to the N radical gun to such an extent that Si resulting from sputtering of the electrodeless discharge tube is not taken into the film. When the Si concentration is suppressed to the SIMS detection lower limit (5 × 10 ¹⁶ cm ⁻³ ) or less, it can be determined that Si is not taken into the film.

  Note that the lower limit of the applied RF power at which the electrodeless discharge tube of the N radical gun is sputtered can vary depending on the crystal manufacturing apparatus. The RF power to the extent that the electrodeless discharge tube of the N radical gun is not sputtered (to the extent that impurities due to sputtering of the electrodeless discharge tube of the N radical gun are not taken into the film) can be obtained experimentally.

Note that a substrate used for manufacturing a ZnO-based semiconductor light-emitting device includes a zinc oxide (ZnO) substrate, a sapphire (Al ₂ O ₃ ) substrate, a silicon carbide (SiC) substrate, a gallium nitride (GaN) substrate, and Mg _x Zn _{1. -x} O board (0 <x ≦ 0.6), and the like. In order to obtain a ZnO-based semiconductor layer with good crystallinity, a substrate having a smaller lattice mismatch is better, and a ZnO substrate is particularly preferable. Further, in the case of manufacturing a light emitting element, an Mg _x Zn _1-x O substrate (0 <x ≦ 0.6) having a larger band gap than ZnO is used so that the substrate does not absorb the emitted light from the active layer. It is also preferable to use it.

A ZnO-based semiconductor substrate such as a ZnO substrate or a Mg _x Zn _1-x O substrate (0 <x ≦ 0.6) uses various surfaces such as a + c plane, a −c plane, an a plane, and an m plane. A ZnO-based compound semiconductor layer can be grown. Further, various substrates having off angles in the m direction, the a direction, and the like can be used.

  The n-type ZnO-based semiconductor layer of the ZnO-based semiconductor light-emitting element is an undoped layer formed in the above experiment, but may be a Ga-doped or Al-doped ZnO layer or MgZnO layer, or a laminated structure of a ZnO layer and an MgZnO layer. But it ’s okay. In the above experiment, an N-doped MgZnO layer was formed as the p-type ZnO-based semiconductor layer, but it may be an N-doped ZnO layer or a stacked structure of both.

  Although the present invention has been described with reference to the embodiments, the present invention is not limited thereto. It will be apparent to those skilled in the art that various modifications, improvements, combinations, and the like can be made.

DESCRIPTION OF SYMBOLS 1 Vacuum vessel 2 Zn source gun 3 Mg source gun 4 O radical gun 5 N radical gun 6 Stage 7 Substrate 8 RHEED gun 9 Screen 21 Case 22 Knudsen cell 22S SUMO cell 22K K cell 23 Filament heater 24 Solid source 25 Spacer 26 MBE Device fixing flange 31 Case 32 Electrodeless discharge tube 33 RF coil 34 Connection portion 35 Gas supply tube 36 MBE device fixing flange 37 Counter electrode P1 to P8 Port or source gun / radical gun 41 installed at the port ZnO substrate 42 ZnO buffer layer 43 n-type ZnO layer 44 ZnO active layer 45 N-doped Mg _x Zn _1-x O layer 46 n-side electrode 47 p-side transparent electrode 48 p-side bonding electrode

Claims (4)

A Zn source gun, an O radical gun, and an N radical gun are provided, and an Mg source gun is provided as necessary. The N radical gun applies a radio frequency to an electrodeless discharge tube into which a gas containing nitrogen is introduced. A plasma containing N radicals is generated, and a crystal manufacturing apparatus using pBN or quartz as a material for the electrodeless discharge tube is used. Above the surface of the single crystal substrate, a Zn beam, an O radical beam, an N radical beam, And a ZnO-based semiconductor film manufacturing method of growing an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film by simultaneously irradiating with a Mg beam as necessary,
When viewed from the normal direction of the substrate, the Zn source gun, the O radical gun, the upper side of the growth surface side of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film, An N radical gun and the Mg source gun are arranged side by side in the circumferential direction, and an angle formed by an azimuth angle in the beam irradiation direction of the N radical gun and an azimuth angle in the beam irradiation direction of the Zn source gun θ _c is set in the range of 90 ° ≦ θ _c ≦ 270 °, and the radio frequency power applied to the electrodeless discharge tube of the N radical gun is set to B or B sputtered from the electrodeless discharge tube. A method for producing a ZnO-based semiconductor film, wherein Si is suppressed to a low power so that Si is not taken into the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film.   2. The method for producing a ZnO-based semiconductor film according to claim 1, wherein the Zn beam, the O radical beam, and, if necessary, the Mg beam are irradiated so that a flux ratio becomes a stoichiometric condition or an O-rich condition. The N radical beam is irradiated so that an N concentration in the N _- doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film is 1 × 10 ²⁰ cm ⁻³ or more. 3. The method for producing a ZnO-based semiconductor film according to 1 or 2. Forming an n-type ZnO-based semiconductor layer above the surface of the single crystal substrate;
Forming a p-type ZnO-based semiconductor layer above the surface of the n-type ZnO-based semiconductor layer,
The step of forming the p-type ZnO-based semiconductor layer includes:
A Zn source gun, an O radical gun, and an N radical gun are provided, and an Mg source gun is provided as necessary. The N radical gun applies a radio frequency to an electrodeless discharge tube into which a gas containing nitrogen is introduced. Using a crystal manufacturing apparatus in which pBN or quartz is used as a material for the electrodeless discharge tube, a Zn beam, an O radical beam, By simultaneously irradiating an N radical beam and, if necessary, an Mg beam, an N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film is grown,
When viewed from the normal direction of the substrate, the Zn source gun, the O radical gun, the upper side of the growth surface side of the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film, An N radical gun and the Mg source gun are arranged side by side in the circumferential direction, and an angle formed by an azimuth angle in the beam irradiation direction of the N radical gun and an azimuth angle in the beam irradiation direction of the Zn source gun θ _c is set in the range of 90 ° ≦ θ _c ≦ 270 °, and the radio frequency power applied to the electrodeless discharge tube of the N radical gun is set to B or B sputtered from the electrodeless discharge tube. A method for producing a ZnO-based semiconductor light-emitting device, wherein Si is suppressed to a low power so that Si is not taken into the N-doped Mg _x Zn _1-x O (0 ≦ x ≦ 0.6) single crystal film.