Methods for adjusting the electrical properties of oxide semiconductors and development of highly conductive p-type and n-type Ga2O3

Abstract

Methods for bipolar doping of oxide semiconductor materials, methods for n-type doping of oxide semiconductor materials, methods for p-type doping of oxide semiconductor materials, and products thereof are described. Also described are p-type Ga2O3 having hydrogen atoms as dopants. 16 cm 2 and / or a sheet carrier concentration of at least about 100 cm at room temperature 2 Also described is n-type Ga2O3 with both mobilities of 1000 volts and 1000 volts.

Description

[Technical field]

[0001] Related Applications This application claims priority under 35 U.S.C. §111(b) to U.S. Provisional Application No. 62 / 858,092, filed June 6, 2019, the entire disclosure of which is incorporated herein by reference.

[0002] STATEMENT REGARDING FEDERALLY FUNDED RESEARCH This invention was made without government support. The Government has no rights in this invention. [Background technology]

[0003] Ga2O3 is a desirable material in many high power devices and optoelectronic applications. However, p-type Ga2O3 has not been reported and is considered a major challenge in the art. Methods for producing p-type Ga2O3 or highly conductive n-type Ga2O3 are currently unknown, even though such materials are highly desirable for certain types of devices. It would be advantageous to develop p-type Ga2O3 and highly conductive n-type Ga2O3, as well as new and improved methods for producing them. Summary of the Invention

[0004] A composition is provided that includes an oxide semiconductor material having cation vacancies filled with hydrogen. In certain embodiments, the oxide semiconductor material has a hydrogen-containing oxide semiconductor material having a hydrogen-containing cation vacancy. Ca ) 1- In one particular embodiment, the oxide semiconductor material has a p-type conductivity. Ca ) 1+ In one particular embodiment, the oxide semiconductor material comprises Ga2O3.

[0005] Further provided is a composition comprising p-type Ga2O3. In certain embodiments, the p-type Ga2O3 comprises hydrogen atoms as dopants. In certain embodiments, the hydrogen atoms are present in Ga vacancies in the p-type Ga2O3. Also provided are power devices and optoelectronic devices comprising the composition.

[0006] Further provided is a composition comprising n-type Ga2O3, wherein the n-type Ga2O3 comprises hydrogen atoms as dopants. In certain embodiments, the n-type Ga2O3 has a crystal structure with four hydrogen atoms in Ga vacancies. In certain embodiments, the n-type Ga2O3 has a crystal structure with at least about 10 16 cm -2 In one particular embodiment, the n-type GaO has a sheet carrier concentration of at least about 100 cm at room temperature. 2 / VS. Also provided are optoelectronic and power devices comprising the composition.

[0007] Furthermore, n-type Ga2O3 is at least about 10 16 cm -2 and at least about 100 cm at room temperature. 2 Also provided is a composition comprising n-type Ga2O3 having both a mobilities of about 10 / VS. -4 The composition has a resistivity of Ω·cm. In certain embodiments, the n-type Ga2O3 includes hydrogen atoms as a dopant. Also provided is an optoelectronic device comprising the composition. In certain embodiments, the composition is a thin film having a thickness in the range of about 100 nm to about 900 nm. In certain embodiments, the composition is a thin film having a thickness in the range of about 200 nm to about 700 nm. In certain embodiments, the composition is a thin film having a thickness of about 500 nm. Also provided is an optoelectronic device and a power device comprising the composition.

[0008] Also provided is an optoelectronic or power device comprising a pn junction formed from p-type Ga2O3 and n-type Ga2O3. In certain embodiments, the p-type Ga2O3 includes hydrogen atoms as a dopant. In certain embodiments, the n-type Ga2O3 includes hydrogen atoms as a dopant. In certain embodiments, both the p-type Ga2O3 and the n-type Ga2O3 include hydrogen atoms as a dopant. In certain embodiments, the n-type Ga2O3 has a dopant concentration of at least about 10 16 cm -2 or at least about 100 cm at room temperature 2 In a particular embodiment, the n-type Ga2O3 has at least one of a mobility of about 10 -4 In certain embodiments, the power device is a high power diode, switch, or transistor, or the optoelectronic device is a photodiode, a phototransistor, a photomultiplier tube, an optical isolator, an integrated optical circuit element, a photoresistor, a charge-coupled imaging device, a light emitting diode (LED), an organic light emitting diode (OLED), a solar-blind UV detector, a gas sensor, a photodetector, a power transistor, or a solar cell.

[0009] Additionally, a method of bipolar doping is provided that includes either partially filling cation vacancies in the oxide semiconductor material with hydrogen to reduce acceptor states to act as shallow acceptors, thereby doping the oxide semiconductor material p-type, or filling the cation vacancies with additional H ions in addition to hydrogen, thereby doping the oxide semiconductor material n-type.

[0010] Further provided is a method for doping an oxide semiconductor material to be p-type, the method including placing the oxide semiconductor material in a sealed system; evacuating air from the sealed system; introducing hydrogen gas into the sealed system; and annealing the oxide semiconductor material in the sealed system at an elevated temperature for a duration of time to diffuse the hydrogen gas into the oxide semiconductor material, thereby doping the oxide semiconductor material to be p-type.

[0011] In certain embodiments, the duration is at least about 1 hour. In certain embodiments, the duration is about 1 hour to about 5 hours. In certain embodiments, the duration is about 1 hour to about 2 hours. In certain embodiments, the elevated temperature is in the range of about 310° C. to about 1000° C. In certain embodiments, the elevated temperature is in the range of about 350° C. to about 1000° C. In certain embodiments, the elevated temperature is in the range of about 700° C. to about 950° C. In certain embodiments, the elevated temperature is about 950° C. In certain embodiments, the sealed system is at a pressure of less than 1 atm during annealing. In certain embodiments, the sealed system is at a pressure of about 300 Torr to about 700 Torr during annealing. In certain embodiments, the sealed system is at a pressure of about 500 Torr to about 650 Torr during annealing. In certain embodiments, the sealed system is at a pressure of about 580 Torr during annealing. In one particular embodiment, the elevated temperature is about 950° C. and the duration is about 2 hours. In one particular embodiment, the oxide semiconductor material includes Ga2O3. In one particular embodiment, the elevated temperature is about 950° C. and the duration is about 2 hours, the oxide semiconductor material includes Ga2O3, and the sealed system is at a pressure of about 580 Torr during annealing.

[0012] Also provided is a method for doping an oxide semiconductor material n-type, the method including annealing the oxide semiconductor material at a first temperature in air for a first duration; placing the oxide semiconductor material in a sealed system; evacuating the air from the sealed system; introducing hydrogen gas into the sealed system; and annealing the oxide semiconductor material in the sealed system at a second elevated temperature for a second duration to diffuse the hydrogen gas into the oxide semiconductor material, thereby doping the oxide semiconductor material n-type.

[0013] In certain embodiments, the first duration is at least about 1 hour. In certain embodiments, the first duration is in the range of about 1 hour to about 5 hours. In certain embodiments, the first duration is in the range of about 1 hour to about 2 hours. In certain embodiments, the first temperature is in the range of about 310° C. to about 1000° C.. In certain embodiments, the first temperature is in the range of about 350° C. to about 1000° C.. In certain embodiments, the first temperature is in the range of about 700° C. to about 950° C.. In certain embodiments, the first temperature is about 950° C.. In certain embodiments, the second duration is at least about 1 hour. In certain embodiments, the second duration is in the range of about 1 hour to about 5 hours. In certain embodiments, the second duration is in the range of about 1 hour to about 2 hours. In certain embodiments, the second duration is about 2 hours. In certain embodiments, the second temperature is in the range of about 310° C. to about 1000° C. In certain embodiments, the second temperature is in the range of about 350° C. to about 1000° C. In certain embodiments, the second temperature is in the range of about 700° C. to about 950° C. In certain embodiments, the second temperature is about 950° C. In certain embodiments, the sealed system is at a pressure of less than 1 atm during the annealing. In certain embodiments, the sealed system is at a pressure of about 400 Torr to about 700 Torr during the annealing. In certain embodiments, the sealed system is at a pressure of about 500 Torr to about 650 Torr during the annealing. In certain embodiments, the sealed system is at a pressure of about 580 Torr during the annealing. In certain embodiments, the oxide semiconductor material includes Ga2O3. In certain embodiments, the second duration is about 2 hours, the second temperature is about 950° C., the oxide semiconductor material includes Ga2O3, and the sealed system is at a pressure of about 580 Torr during the annealing. In one particular embodiment, the oxide semiconductor material includes Ga2O3, the first duration is about 2 hours, the first temperature is about 950° C., the second duration is about 2 hours, the second temperature is about 950° C., and the sealed system is at a pressure of about 580 Torr during annealing.

[0014] In some embodiments of the methods described herein, instead of annealing or molecular hydrogen diffusion, a hydrogen plasma is used in the plasma reactor at a lower temperature or any range of temperatures, at any range of pressures, and for any duration.

[0015] Further provided are products of any of the methods described herein. The use of hydrogen to dope oxide semiconductor materials n-type or p-type without additional dopants is also provided.

[0016] The patent or patent application file may contain one or more drawings and / or one or more photographs executed in color. Copies of this patent or patent application publication with color drawings and / or photographs may be obtained from the U.S. Patent and Trademark Office upon request and payment of the necessary fee. [Brief description of the drawings]

[0017] [Figure 1] Figure 1 shows the theoretical ideal performance limits of β-Ga2O3 power devices compared to other leading semiconductors. [Figure 2A] 2A-2D are schematic diagrams showing the incorporation of hydrogen into β-Ga2O3. Figure 2A shows how hydrogen molecules come into contact with a surface at elevated temperature and dissociate heterogeneously. The electron cloud of H2 is attracted to the gallium and the protons are attracted to the oxygen. [Figure 2B] Figures 2A-D are schematic diagrams showing the incorporation of hydrogen into β-Ga2O3. Figure 2B shows how at high temperatures, protons and hydrides are attracted to oxygen and gallium atoms, respectively, on the crystal surface and diffuse into the bulk crystal. Protons are attracted toward negatively charged gallium vacancies. [Figure 2C] 2A-2D are schematic diagrams showing the incorporation of hydrogen into β-Ga2O3, and Fig. 2C shows a Ga vacancy decorated with two hydrogens obtained from DFT calculations. [Figure 2D]2A-2D are schematic diagrams showing the incorporation of hydrogen into β-Ga2O3, and Fig. 2D shows a Ga vacancy decorated with four hydrogens obtained from DFT calculations. [Figure 3A] Figure 3 shows the sheet resistance (Fig. 3A) and sheet number (Fig. 3B) of β-Ga2O3 samples after several treatments. First set 1, 2, 3, 4, 5: 1-as-grown β-Ga2O3 single crystal; 2-annealed in H2; 3-annealed in H2 after 4 days; 4-annealed for the second time in H2; 5-annealed for the second time in H2 after 4 days (H diffusion at 700 °C for 1 h resulted in p-type conductivity that decayed over time); Second set a, b, c: a-as-grown β-Ga2O3 single crystal; b-annealed in H2 immediately after annealing; c-annealed in H2 4 days after annealing (H diffusion at 950 °C for 2 h resulted in stable p-type conductivity over time by allowing H+ diffusion deeper into the crystal). Black squares - annealing at 700 °C for 1 h; red circles - annealing at 950 °C for 2 h. [Figure 3B] Figure 3 shows the sheet resistance (Fig. 3A) and sheet number (Fig. 3B) of β-Ga2O3 samples after several treatments. First set 1, 2, 3, 4, 5: 1-as-grown β-Ga2O3 single crystal; 2-annealed in H2; 3-annealed in H2 after 4 days; 4-annealed for the second time in H2; 5-annealed for the second time in H2 after 4 days (H diffusion at 700 °C for 1 h resulted in p-type conductivity that decayed over time); Second set a, b, c: a-as-grown β-Ga2O3 single crystal; b-annealed in H2 immediately after annealing; c-annealed in H2 4 days after annealing (H diffusion at 950 °C for 2 h resulted in stable p-type conductivity over time by allowing H+ diffusion deeper into the crystal). Black squares - annealing at 700 °C for 1 h; red circles - annealing at 950 °C for 2 h. [Figure 3C]Figure 3C shows the sheet resistance of the samples, and Figure 3D shows the sheet number of the samples.Third set 1, 2, 3: 1-as-grown β-Ga2O3 single crystal; 2-annealed in O2; 3-annealed in O2 followed by H2 diffusion (black squares) (O annealing increased the resistivity and subsequent H diffusion gave high n-type conductivity that was stable over time);Fourth set a, b: a-as-grown β-Ga2O3 single crystal; b-annealed in Ga followed by H2 diffusion (red circles) (annealing in Ga and subsequent H diffusion did not induce significant changes).

[0018] [Figure 3D] Figure 3C shows the sheet resistance of the samples, and Figure 3D shows the sheet number.Third set 1, 2, 3: 1-as-grown β-Ga2O3 single crystal; 2-annealed in O2; 3-annealed in O2 followed by H2 diffusion (black squares) (O annealing increased the resistivity and subsequent H diffusion gave high n-type conductivity that was stable over time);Fourth set a, b: a-as-grown β-Ga2O3 single crystal; b-annealed in Ga followed by H2 diffusion (red circles) (annealing in Ga and subsequent H diffusion did not induce significant changes). [Figure 4A] Figures 4A-C show the thermally stimulated luminescence emission (Figure 4A) of samples annealed at 950 °C for 2 h in different environments. Data points for annealed samples were normalized to 0-1. Data points for as-grown samples were normalized to 0-0.5 to minimize noise (no glow peak). Peaks were fitted with a Gaussian function. The two peaks seen at low temperature after H2 diffusion and after H diffusion followed by O annealing were associated with shallow acceptors and shallow donors induced in the sample, respectively, and were used to calculate the ionization energy. [Figure 4B] FIG. 4B shows a flat band diagram illustrating the donor and acceptor states of the sample after direct hydrogen diffusion. [Figure 4C]FIG. 4C shows hydrogen diffusion after filling the oxygen vacancies. [Figure 5A] Figures 5A-5D are evidence of the formation of shallow acceptors and shallow donors from hydrogen doping of Ga2O3. Figure 5A shows the defect parameters S and W as a function of penetration depth measured by Doppler broadening of positron annihilation spectroscopy (DBPAS), where S and W are defined as the fraction of positron annihilation by valence and core electrons, respectively. The lower x-axis represents positron energy and the upper x-axis represents penetration depth. The graph shows that H2 diffuses about 500 nm into the crystal. [Figure 5B] Positron annihilation lifetime spectroscopy (PALS) data for the as-grown sample (FIG. 5B), and the sample annealed in H2 (950 °C for 2 h) (FIG. 5C), and the sample annealed in O2 followed by H2 (950 °C for 2 h) (FIG. 5D). EP = positron incident energy, Zmean = positron incident depth, t = positron lifetime, I = intensity of lifetime component. Graphs b, c, and d show the two positron lifetime components and their intensities for each sample. [Figure 5C] Positron annihilation lifetime spectroscopy (PALS) data for the as-grown sample (FIG. 5B), and the sample annealed in H2 (950 °C for 2 h) (FIG. 5C), and the sample annealed in O2 followed by H2 (950 °C for 2 h) (FIG. 5D). EP = positron incident energy, Zmean = positron incident depth, t = positron lifetime, I = intensity of lifetime component. Graphs b, c, and d show the two positron lifetime components and their intensities for each sample. [Figure 5D] Positron annihilation lifetime spectroscopy (PALS) data for the as-grown sample (FIG. 5B), and the sample annealed in H2 (950 °C for 2 h) (FIG. 5C), and the sample annealed in O2 followed by H2 (950 °C for 2 h) (FIG. 5D). EP = positron incident energy, Zmean = positron incident depth, t = positron lifetime, I = intensity of lifetime component. Graphs b, c, and d show the two positron lifetime components and their intensities for each sample. [Figure 6A]6A-6B are schematic diagrams of the thermoluminescence process in the case of a donor (FIG. 6A) and an acceptor (FIG. 6B). [Figure 6B] 6A-6B are schematic diagrams of the thermoluminescence process in the case of a donor (FIG. 6A) and an acceptor (FIG. 6B). [Figure 7A] 7A-7B are calculations of ionization energies using the initial rise method. Linear fits of ln(I) versus 1 / T are shown for a β-Ga2O3 sample annealed in oxygen followed by hydrogen diffusion (FIG. 7A) and a β-Ga2O3 sample without hydrogen diffusion (FIG. 7B). [Figure 7B] 7A-7B are calculations of ionization energies using the initial rise method. Linear fits of ln(I) versus 1 / T are shown for a β-Ga2O3 sample annealed in oxygen followed by hydrogen diffusion (FIG. 7A) and a β-Ga2O3 sample without hydrogen diffusion (FIG. 7B). [Figure 8] FIG. 8 shows positron lifetime spectra at Ep=6 for the H2-diffused sample and the sample annealed in O2 followed by H2-diffusion. [Figure 9A] 9A-9D are temperature dependent transport properties of n-type and p-type H-treated β-Ga2O3 samples. Figure 9A shows the sheet resistance. [Figure 9B] 9A-9D are temperature dependent transport properties of n-type and p-type H-treated β-Ga2O3 samples, and FIG. 9B shows the sheet number. [Figure 9C] Figures 9A-9D show temperature-dependent transport properties of n-type and p-type H-treated β-Ga2O3 samples, and Figure 9C shows the logarithm of the number of sheets plotted as a function of 1000 / T. [Figure 9D] 9A-9D are temperature dependent transport properties of n-type and p-type H-treated β-Ga2O3 samples. Fig. 9D shows the temperature dependence of n-type mobility. [Figure 10] Figure 10 shows X-ray diffraction (XRD) measurements of the as-grown sample, the H-diffused p-type sample, and the O-annealed and H-diffused n-type sample. [Figure 11A] 11A-11C show the TSL glow curves of the as-grown sample (FIG. 11A) and the p-type H-diffused sample (FIG. 11B), as well as the location of these levels in the bandgap (FIG. 11C). [Figure 11B] 11A-11C show the TSL glow curves of the as-grown sample (FIG. 11A) and the p-type H-diffused sample (FIG. 11B), as well as the location of these levels in the bandgap (FIG. 11C). [Figure 11C] 11A-11C show the TSL glow curves of the as-grown sample (FIG. 11A) and the p-type H-diffused sample (FIG. 11B), as well as the location of these levels in the bandgap (FIG. 11C). [Figure 12A] 12A-B are TSL glow curves of the as-grown sample (FIG. 12A) and the n-type O annealed and H diffused sample (FIG. 12B), showing deep traps in the sample. [Figure 12B] 12A-B are TSL glow curves of the as-grown sample (FIG. 12A) and the n-type O annealed and H diffused sample (FIG. 12B), showing deep traps in the sample. [Figure 12C] FIG. 12C shows a diagram illustrating the location of these levels in the bandgap. [Figure 13A] 13A-13D are contour plots of emission intensity as a function of temperature and wavelength for the as-grown sample (FIG. 13A), the H-diffused sample (FIG. 13B), and the H-diffused sample following O annealing (FIG. 13C). Figure 13D shows green emission from the H-diffused p-type sample. [Figure 13B] 13A-13D are contour plots of emission intensity as a function of temperature and wavelength for the as-grown sample (FIG. 13A), the H-diffused sample (FIG. 13B), and the H-diffused sample following O annealing (FIG. 13C). Figure 13D shows green emission from the H-diffused p-type sample. [Figure 13C] 13A-13D are contour plots of emission intensity as a function of temperature and wavelength for the as-grown sample (FIG. 13A), the H-diffused sample (FIG. 13B), and the H-diffused sample following O annealing (FIG. 13C). Figure 13D shows green emission from the H-diffused p-type sample. [Figure 13D] 13A-13D are contour plots of emission intensity as a function of temperature and wavelength for the as-grown sample (FIG. 13A), the H-diffused sample (FIG. 13B), and the H-diffused sample following O annealing (FIG. 13C). Figure 13D shows green emission from the H-diffused p-type sample. [Figure 14A] 14A-B are photoluminescence emission and mechanism. Figure 14A shows the PL emission of the as-grown sample and the p-type H-diffused sample. [Figure 14B] Figures 14A-B show the emission and mechanism of photoluminescence. Figure 14B shows two diagrams, the right diagram showing the mechanism of 380 nm UV emission and the left diagram showing the mechanism of broadband emission (blue and green). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Throughout this disclosure, various publications, patents, and published patent specifications are referenced by an identifying citation. The disclosures of these publications, patents, and published patent specifications are incorporated by reference in their entireties into this disclosure in order to more fully describe the state of the art to which this invention pertains.

[0020] Provided herein are methods to induce bipolar (n-type and p-type) conductivity in wide bandgap oxide semiconductors and to control their transport properties simply by controlling the incorporation of hydrogen into the lattice. These methods allow high carrier density to be achieved while maintaining good mobility in oxide semiconductors by using hydrogen instead of traditional doping methods. Shallow donor type of semiconductors can be created by modifying cation vacancies with hydrogen. It is shown herein that hydrogen can act as a shallow donor to induce very high n-type conductivity in oxide semiconductors or as a shallow acceptor to induce p-type conductivity in oxide semiconductors.

[0021] Wide bandgap oxide semiconductors are generally difficult to dope in either direction, especially while maintaining high carrier concentration and good mobility. The present disclosure alleviates these challenges by providing wide bandgap oxide semiconductors that can be doped either n-type or p-type and still have high carrier concentration and good mobility. Without being bound by theory, it is believed that this method of inserting hydrogen ions into cation vacancies does not introduce disorder into the crystal structure, and therefore does not degrade carrier concentration and mobility. This is because hydrogen ions are very small.

[0022] Using the methods described herein, p-type Ga2O3 has been developed. Additionally, using the methods described herein, significantly more conductive n-type Ga2O3 has been developed. As can be seen from FIG. 1 and Table 1, Ga2O3 has advantageous properties for use in high power and optoelectronic devices. However, achieving p-type is critical for these applications. FIG. 1 shows the theoretical ideal performance limits of β-Ga2O3 power devices compared to other leading semiconductors. Table 1 below shows the material properties of leading semiconductors and β-Ga2O3.

[0023] [Table 1]

[0024] However, as with other wide bandgap oxide semiconductors, doping Ga2O3 is difficult. Doping typically reduces the mobility of the semiconductor, and increasing carrier concentration by doping usually comes at the expense of mobility. In fact, the only n-type Ga2O3 reported to date does not have the very high electrical conductivity described herein, which is important for transparent conductor applications, and p-type Ga2O3 has not yet been reported. The present disclosure provides for the fabrication of both p-type Ga2O3 and highly conductive n-type Ga2O3. For example, the n-type Ga2O3 of the present disclosure can be thinned to a thickness of about several hundred nanometers, with a thickness of about 10 μm. 16 cm -2 and a carrier concentration of about 100 cm at room temperature. 2 In view of its high conductivity, the n-type Ga2O3 described herein is particularly useful as a transparent conductive oxide (TCO) in optoelectronic devices such as photovoltaic devices, light emitting diodes (LEDs), or light emitting devices, or in high power devices.

[0025] The present disclosure provides a different donor type, shallow donor, in semiconductors by modifying cation vacancies with hydrogen. In the examples herein, it is shown that hydrogen can act as a shallow donor and induce very high n-type conductivity in oxides. In the examples herein, it is also shown that hydrogen can act as a shallow acceptor and induce p-type conductivity in oxides. Thus, provided herein are oxide semiconductor materials in which cation vacancies are filled with hydrogen. In some embodiments, the oxide semiconductor materials have a (HV Ca ) 1- In some embodiments, the oxide semiconductor material has a p-type conductivity. Ca ) 1+The complex includes a compound in which Ca is the cation and the oxide semiconductor material has n-type conductivity. Ga2O3 is described herein for illustrative purposes, but the disclosure is in no way limited to Ga2O3. Rather, any oxide semiconductor material can be doped p-type or n-type with hydrogen as described herein.

[0026] In general, bipolar doping according to the present disclosure involves exposing an oxide semiconductor material to hydrogen in a sealed system, and then annealing the sealed system at an elevated temperature for a duration to allow hydrogen to diffuse into the oxide semiconductor material. If this hydrogen diffusion process is performed without first annealing the oxide semiconductor material in air or otherwise in the presence of oxygen, the resulting oxide semiconductor material can be doped p-type. However, if the hydrogen diffusion process is performed after first annealing the oxide semiconductor material in air or otherwise in the presence of oxygen, the resulting oxide semiconductor material can be doped n-type. Without being bound by theory, it is believed that hydrogen fills the cation vacancies in the oxide semiconductor material, but may also fill the oxygen vacancies. If oxygen annealing is performed before hydrogen diffusion, the oxygen vacancies in the lattice are filled first, which allows more hydrogen atoms to be directed into the cation vacancies during the subsequent hydrogen diffusion. In other words, the absence of oxygen vacancies after oxygen annealing means that H + This means that the only available traps for n are cation vacancies, which will therefore be filled to a higher degree, resulting in n-type conductivity. This doping can also be achieved by increasing the amount of hydrogen diffused (by increasing the diffusion time or H pressure) instead of a prior O anneal. Figures 2A-2D show the incorporation of hydrogen.

[0027] Hydrogen is typically introduced into the sealed system at a pressure of less than 1 atm, and the sealed system may be maintained at a pressure of less than 1 atm, such as from about 400 Torr to about 700 Torr, or from about 500 Torr to about 650 Torr, during annealing. In one non-limiting example, the sealed system is at a pressure of about 580 Torr during annealing. Air may be evacuated from the sealed system to create a vacuum prior to introduction of hydrogen. The sealed system may be of any scale, for example an ampoule or a large vacuum chamber. The sealed system may be inserted into an oven or furnace, or may itself include a heating element. The size and configuration of the sealed system are not particularly limited.

[0028] As noted above, the hydrogen diffusion process involves annealing at an elevated temperature for a duration. The duration may be at least about 1 hour. In some embodiments, the duration is within a range of about 1 hour to about 5 hours, or about 1 hour to about 2 hours. In some embodiments, the duration is greater than about 2 hours. For best results, the ideal duration may depend on other variables, such as the temperature of the annealing. For example, for lower temperatures, a longer duration for diffusion may be desirable.

[0029] The elevated temperature for the hydrogen diffusion annealing step is generally above 300°C. In some embodiments, the elevated temperature is in the range of about 310°C to about 1000°C, or about 350°C to about 1000°C, or about 700°C to about 950°C. If too low a temperature is used, proper hydrogen diffusion may not work. For example, at 300°C, the method may not work well to produce p-type Ga2O3 well. However, the elevated temperature required for proper diffusion of hydrogen into the lattice may depend on the length of time for diffusion and the pressure at which the sealed system is exposed to the hydrogen. For example, if a shorter length of time is used, increasing the temperature may give better results. Also, if a hydrogen plasma is used instead of a gas, doping may be performed at a much lower temperature.

[0030] As shown in the examples herein, the oxide semiconductor β-Ga2O3 becomes conductive when diffused at an elevated temperature of 700°C and a pressure of 580 Torr for 1 hour, but then becomes resistive again after about 4 days due to a reversal of diffusion. However, when hydrogen is diffused at an elevated temperature of 950°C and a pressure of 580 Torr for 2 hours, the conductivity becomes permanent. Thus, the present disclosure can be used to provide permanent conductivity in oxide semiconductor materials, but can also be used to generate temporary conductivity.

[0031] When doping an oxide semiconductor material to n-type, compared to the method for doping an oxide semiconductor to p-type, only one additional step is required: annealing the oxide semiconductor in air (or otherwise in the presence of oxygen) before exposing the oxide semiconductor material to hydrogen gas in a sealed system, followed by annealing at an elevated temperature. Thus, when doping an oxide semiconductor material to n-type, the method includes annealing the oxide semiconductor material in air or oxygen at a first temperature for a first duration, followed by exposing the oxide semiconductor to hydrogen gas in a sealed system (i.e., placing the annealed oxide semiconductor material in a sealed system, evacuating the air from the sealed system, and introducing hydrogen gas into the sealed system), and then annealing the oxide semiconductor with hydrogen gas in the sealed system at a second temperature for a second duration. In this method, the second temperature is equivalent to the elevated temperature discussed above, and the second duration is equivalent to the duration discussed above (for doping an oxide semiconductor to p-type). However, the first temperature and the first duration may be different or the same as the second temperature and the second duration, respectively. In some embodiments, the first duration is at least about 1 hour. In some embodiments, the first duration is in a range of about 1 hour to about 5 hours, or about 1 hour to about 2 hours. In some embodiments, the first temperature is in a range of about 310° C. to about 1000° C., or about 350° C. to about 1000° C., or about 700° C. to about 950° C. In a particular embodiment, the first temperature is about 950° C. The pressure of the sealed system during the hydrogen annealing may be comparable to that of the method for p-doping an oxide semiconductor.

[0032] The products of the methods described herein are highly advantageous, for example, because they can be transparent conductive oxides with high carrier concentration and good mobility. As noted above, the methods are suitable for producing stable p-type GaO and for at least about 10 16 cm -2 or at least about 100 cm at room temperature. 2In one embodiment, the composition was used to fabricate n-type Ga2O3, which may have a mobility of 100 / VS. Each of these compositions may be used in optoelectronic devices. Additionally, both p-type Ga2O3 and n-type Ga2O3 may be used to create pn junctions in devices, such as optoelectronic devices. Thus, stable p-type Ga2O3, as well as a junction with a mobility of at least about 10 16 cm -2 and / or at least about 100 cm at room temperature 2 In addition to n-type Ga2O3, which may have a mobility of / VS, Ga2O3-based devices are also provided herein having a pn junction formed from p-type Ga2O3 and n-type Ga2O3 (which may or may not be the highly conductive Ga2O3 described herein). Non-limiting examples of optoelectronic devices in which the doped oxide semiconductor materials described herein may be used include photodiodes, phototransistors, photomultipliers, optical isolators, integrated optical circuit elements, photoresistors, charge-coupled imaging devices, light-emitting diodes (LEDs), organic light-emitting diodes (OLEDs), solar-blind UV detectors, gas sensors, photodetectors, transistors, or solar cells. Non-limiting examples of power devices in which the doped oxide semiconductor materials described herein may be used include power diodes, power switches, and power transistors.

[0033] Although Ga2O3 is described herein for purposes of example, the method is not limited to use with Ga2O3. The method described herein may be used for bipolar doping of other wide band gap oxide semiconductors, and may in fact be simpler than most conventional doping methods. The method may be implemented for doping of any oxide semiconductor. Without being bound by theory, it is believed that the method may be used for any oxide semiconductor because oxygen and hydrogen easily form bonds.

[0034] The bipolar doping described herein, using only hydrogen diffusion and without further doping with other elements, is very different from conventional doping, which is typically achieved by doping with other elements during or after growth. According to the present disclosure, hydrogen diffusion can be used to dope and induce high conductivity. Hydrogen doping can result in higher carrier density and good mobility. Hydrogen doping can generate shallow donor type in the semiconductor. By filling the cation vacancies with hydrogen ions, the electron mobility in the semiconductor can be improved, which is useful in switching and high RF. As shown in the examples herein, this method can be used to fabricate p-type Ga2O3 and even highly conductive n-type Ga2O3. The present disclosure provides cheaper power electronics, gas sensors, MOSFETs, etc., by using gallium oxide.

[0035] Working Example Hydrogen-induced p-type and n-type conductivity in ultrawide bandgap oxides Future technologies for optoelectronics and high power devices rely on the development of wide bandgap materials with exceptional transport properties. However, bipolar doping (n-type and p-type doping) is a major challenge and has hindered the development of many wide bandgap oxide-based devices. Standard chemical doping with elements at substitutional or interstitial sites often reduces carrier mobility, and there is always a trade-off between increasing the maximum achievable carrier density and maintaining good mobility in the oxide. In these examples, we demonstrate how to generate p-type and n-type conductivity in ultra-wide bandgap oxides with bandgap energies of about 5 eV by hydrogen diffusion without further doping. As shown herein, it is possible to switch between p-type and n-type conductivity by controlling H incorporation into the β-Ga2O3 lattice. A nine order of magnitude increase in n-type conductivity (20 meV donor ionization energy) is achieved with a resistivity of 10 -4 Ω cm, sheet carrier concentration 10 16 cm -2, and a mobility of 100 cm at room temperature 2 / VS. The development of stable p-type Ga2O3 (acceptor ionization energy 42 meV) by simply modifying the incorporation of hydrogen into the lattice is further shown below. Density functional theory calculations were used to investigate the incorporation of hydrogen into the Ga2O3 lattice, supporting the interpretation of the experimental results. These examples demonstrate a useful approach to engineer the transport properties of oxide semiconductors and can be used to develop Ga2O3 devices that could greatly advance optoelectronics and high power electronics.

[0036] Wide band gap energy has become a key parameter for the development of future high power transistors and many optoelectronic devices, and wide band gap oxides such as ZnO have been shown to exhibit exceptional properties. However, their use in many applications has been hindered by the lack of conductivity control or the difficulty of achieving high carrier densities with good mobility. Bipolar doping (realization of both n-type and p-type) is one of the major challenges in wide band gap materials, which is important for the development of most devices. Furthermore, substitutional doping of elements, the most common method to obtain charge carriers, often induces disorder, which reduces carrier mobility. These examples describe the induction of p-type and n-type conductivity in ultra-wide band gap oxides without further substitutional doping by controlled incorporation of hydrogen (H) into the lattice, with room temperature electron mobilities of 100 cm. 2 V -1 S -1 Sheet electron density at 10 16 cm -2have demonstrated. Such high electron density and good mobility in an oxide semiconductor is remarkable. The examples of the present invention are carried out with Ga2O3, which is an important material for high power transistors due to its wide band gap (about 4.5-5 eV) and high breakdown field strength of 8 MV / cm. As a transparent semiconducting oxide, Ga2O3 also has applications as transparent contacts in, for example, photovoltaic devices, liquid crystal displays, and light emitting diodes. β-Ga2O3 is the most stable polymorph of the Ga2O3 phase and has a monoclinic structure with space group C2 / m. In its defect-free crystalline form, it behaves as an insulator. However, due to the presence of oxygen vacancies (V O ) is due to the existence of inherent n-type conductivity, V O This has been reported even though recent calculations have confirmed that β-Ga2O3 is a deep donor and therefore unlikely to induce n-type conductivity. For successful device integration, it is essential to construct a conductive β-Ga2O3 film, but many attempts have been unsuccessful. Currently, only one conductivity type (n-type) is achieved by doping β-Ga2O3 with Sn, Ge, or Si during growth. No significant success has been achieved so far for p-type conductivity by doping or annealing.

[0037] Hydrogen is known to have a strong effect on the electrical conductivity of transparent conductive oxides and semiconductors, where it can induce shallow donors and passivate deep compensatory defects. In β-Ga2O3, monoatomic H has low formation energy and can occupy both interstitial and substitutional sites and act as a shallow donor. The complex crystal structure of β-Ga2O3 allows the interstitial hydrogen (H ii + ) can form bonds with oxygen in many configurations, creating energetically similar electronic states. i acts only as a shallow donor, and substitutional hydrogen H Ois believed to have a low formation energy only under oxygen-deficient conditions. Despite these theoretical predictions regarding the possibility of n-type conductivity due to H incorporation at various positions, no significant experimental success has been reported. i Or H O By controlling the incorporation of H into the cation vacancy sites rather than as a cation donor or acceptor, H donors and H acceptors are generated in Ga2O3. Cation vacancies are the electrically compensating acceptors in transparent conductive oxides and semiconductors, including β-Ga2O3. Although cation vacancies have high formation energies in some oxide semiconductors (e.g., SnO2, In2O3), previous first-principles calculations have shown that their formation energies in β-Ga2O3 are very low, so that H-decorated V Ga A high probability of formation was shown to be possible after incorporation of H into the crystal.

[0038] To gain insight into the hydrogen incorporation process in crystals, it is important to understand the interaction of H2 with the surface of metal oxide semiconductors. H incorporation into crystals at high temperatures occurs in two steps. First, H2 dissociates and becomes surface-bound, followed by diffusion into the bulk crystal. Depending on the nature of the material, H2 can follow either homolytic or heterolytic dissociation pathways. In the case of homolytic cleavage, the H2 molecule dissociates to form two H atoms that become bound to oxygen on the crystal surface. On the other hand, in heterolytic cleavage, H2 dissociates to form a proton and a hydride, which in this case become bound to oxygen and metal atoms, respectively. The redox capacity of the metal determines the type of dissociation that is most likely to occur. Density functional theory (DFT) predicts that H2 tends to dissociate heterolytically on the surfaces of non-reducible oxides (e.g., MgO, γ-Al2O3), whereas it follows a homolytic pathway on the surfaces of reducible oxides (e.g., CeO2). β-Ga2O3 was found to be non-reducible by DFT. Therefore, without being bound by theory, H2 most likely follows heterolytic dissociation, as shown in Figure 2A. The adsorbed protons and hydrides diffuse into the bulk crystal at high temperatures. As shown in Figure 2B, the protons are transported to the negatively charged V Ga while the hydride is attracted to the positively charged or neutral V O Such incorporation of hydrogen into different vacancy sites, as shown by these examples, can have a notable effect on the electrical properties of Ga2O3 and semiconducting oxides in general.

[0039] A series of experiments were performed to incorporate H into undoped β-Ga2O3 single crystals. The experiments were performed on samples grown by the Edge-defined Film-fed Growth (EFG) method proceeding parallel to the (010) plane and cut into 5 × 5 × 0.5 mm pieces. The as-grown samples were highly resistive, but after H2 diffusion in a sealed ampoule filled with H2 only at 580 Torr, they showed an increase in carrier density and p-type conductivity. H2 diffusion at 700 °C for 1 h caused an unstable conductivity that decayed with time. However, H2 diffusion at 950 °C for 2 h caused a larger increase in carrier density and a stable p-type conductivity with time. Other procedures were also performed to incorporate H2 at different sites in undoped β-Ga2O3. One sample was annealed in flowing O2 and another sample was annealed with Ga in a sealed ampoule at 950 °C for 2 h. This process is believed to fill the respective (anion or cation) vacancies. Hydrogen was then diffused into the crystal at 580 Torr for 2 hours at 950 °C in a sealed ampoule. O annealing followed by H diffusion resulted in high n-type conductivity (stable over time) and electron mobility of 100 cm. 2 / Vs has a remarkable sheet carrier density of about 10 16 cm -2 In contrast, annealing in Ga followed by H diffusion did not result in a significant increase in conductivity. These results are summarized in Table 2 below and Figures 3A-3D.

[0040] [Table 2]

[0041] Without being bound by theory, the realization of p-type and n-type conductivity after H2 diffusion can be explained as follows: Ga vacancies are transformed into deep acceptors (V Ga ) 3- When hydrogen diffuses into the crystal, the protons adsorbed on the surface (Fig. 2A-B) act as (V Ga ) 3-, where it stabilizes the negative charge and thereby degrades the acceptor state. As a result, H-modified gallium vacancies (V Ga -2H) 1- At lower temperatures (e.g., 700 °C), protons are less likely to diffuse deep inside the bulk crystal. This results in a decrease in conductivity over time due to back diffusion at room temperature. However, samples exposed to H2 at higher temperatures and for longer periods show a decrease in H + Since the p-type conductivity of the GaN-GaN-GaN-GaN is increased, the p-type conductivity of the GaN- ...

[0042] V O The samples exposed to H2 after filling (i.e., after annealing in O2) showed high n-type conductivity. O Because of the absence of V, more H atoms are Ga and acts as a donor (V Ga -4H) 1+ This causes the formation of V (as shown in FIG. 2D). O The absence of H + The only available trap for V Ga This means that it will be filled to a greater extent. Ga The increase in carrier concentration due to the subsequent H diffusion after filling the lattice is negligible, so the H i The contribution of n-type conductivity from H is not significant. It is further confirmed that H-decorated cation vacancies are the main contributor to the induced n-conductivity in the samples.

[0043] H incorporation into Ga vacancies is investigated using density functional theory implemented in the Vienna ab-initio Simulation Package (VASP). These calculations were performed for a 1 × 4 × 2 supercell of β-Ga2O3 containing a total of 160 atoms in the defect-free structure. The Brillouin zone was sampled using a 2 × 2 × 2 Monkhorst-Pack k-point mesh centered at the Γ point. Ga was created by removing tetrahedrally coordinated Ga ions from the cell, since this vacancy structure was identified as more favorable. A net charge of -3 was imparted to the structure. H with a charge of +1 + Ions were inserted into the resulting vacancy structure to reduce the net charge of the cell while the total number of electrons in the system remained constant. Details of the algorithm and binding energy calculations for each configuration are given below. The results are presented in Table 2(d) above. + The binding energy of the ions to the Ga vacancy is -4.4 eV. DFT calculations show that as N (the number of H ions) is increased up to at least N = 4, the reaction remains exothermic, but the bond strength per H atom decreases. + The energy gained by the addition of ions is only -0.8 eV, which is the initial H + This is much smaller than the -4.4 eV obtained with the addition of H ions. If this trend is expected to continue, this would suggest that the + Ion to V Ga These calculations therefore show that it is not advantageous to accommodate a single V Ga It can accommodate 4 H + ions, which have been shown to change the net charge of the complex from 3- (for N=0) to 1+ (for N=4), as well as (V Ga -4H) 1+ (Figure 2D) is H i + From these calculations, it is confirmed that (V Ga -4H) 1+ This supports the interpretation of electrical transport measurements that is the dominant donor in the treated highly conductive n-type samples.

[0044] Thermally stimulated luminescence (TSL) spectroscopy was performed on the samples to calculate the ionization energies of the donor and acceptor. Details of the method and measurements, data analysis, and calculation of the ionization energies are described below. Figure 4A shows the TSL emission for as-grown, p-type, and n-type H2-treated Ga2O3. The as-grown sample does not show a peak corresponding to a shallow level. The other two samples each show a peak at a lower temperature, indicating the formation of a shallow level. The peak formed at 107 K in the p-type H2-annealed sample (red curve in Figure 4A) is associated with the formation of a shallow acceptor with an ionization energy of 42 meV calculated using the early rise method (see Figure 7). The ionization energy of the donor appearing after H2 diffusion following O2 annealing (green curve in Figure 4A) was also calculated from the peak at 111 K using the early rise method and found to be 20 meV (see Figure 7). 4B-4C show the corresponding flat band diagrams and the corresponding donor and acceptor states, respectively.

[0045] To further understand the effect of hydrogen and confirm our interpretation of the origin of the conductivity, positron annihilation spectroscopy (PAS) measurements were performed. PAS is a well-established technique for detecting and characterizing cation vacancies in semiconductors. Both Doppler broadening of positron annihilation spectroscopy (DBPAS) and positron annihilation lifetime spectroscopy (PALS) were used. Details of these techniques and measurements are provided below.

[0046] FIG. 5A shows the dependence of the S and W parameters for the H2-diffused sample and the sample with H2-diffusion followed by O2 annealing. The high value of the S parameter in the very first part of the two curves is common to all DBPAS measurements and indicates the formation of positronium at the surface. The graph shows a large difference between the two samples in the first 500 nm, with the sample with H2-diffused following O2 annealing showing lower S values ​​and higher W values, which indicates a higher n-type conductivity. In DBPAS, the decrease in the S parameter indicates the suppression of positron trapping in cationic or neutral vacancies. Thus, these measurements confirm the reduction of negatively charged vacancies and neutral vacancies in the sample with H2-diffused following O2 annealing. This is due to the filling of Ga vacancies with four or more H ions, which leads to the formation of positive charge states and shallow donors, as shown by the large increase in n-type conductivity. This (HV Ga ) 1+ The complex has a positive charge state and is unable to trap the positron, resulting in a substantial drop in the S-parameter. On the other hand, in the case of H2 diffusion alone, hydrogen is V Ga This partially fills the negative charge state and provides a shallow acceptor, which provides p-type conductivity. Ga ) 1- The complex is still an active positron trap, resulting in a higher S-parameter.

[0047] PALS depth-resolved measurements reveal two major positron lifetime components in each sample. Figures 5B-5D show the lifetime components and their intensities as a function of depth for the as-grown, H2-diffused, and H2-diffused samples following O2 annealing. Clear differences in the intensities and magnitudes of the positron lifetime components are seen among the three samples. The large second lifetime component τ2 is due to the presence of V GaFor as-grown Ga2O3, τ2 is about 470 ps with an intensity of about 25–30% throughout the sample depth (Figure 5B). After H2 annealing, τ2 decreases to about 320 ps, ​​which indicates the presence of V Ga The intensity of τ2 has decreased to about 13%, which indicates that the associated defects are partially filled by hydrogen, due to the reduced positron trapping at these vacancies as a result of the reduced negative charge (Fig. 5C). After H2 diffusion followed by O2 annealing, almost all positrons are annihilated and the lifetime is close to the bulk lifetime. The intensity of τ2 has decreased to about 1%, which indicates the almost complete absence of positron trapping at defects, V Ga There is strong evidence that the relevant defects are filled by H2 and converted into positive charge state donors that cannot trap positrons. Thus, the DBPAS and PALS measurements unambiguously confirm our interpretation of the origin of n-type and p-type conductivity.

[0048] In summary, the development of stable p-type and n-type Ga2O3 has been demonstrated by controlling the diffusion and incorporation of H2 into the β-Ga2O3 lattice, while a facile method for doping and controlling the conductivity of wide band gap oxides has been developed with the realization of remarkably high carrier density and good mobility in oxide semiconductors, which is a major challenge for the common substitutional doping method.

[0049] Hydrogen incorporation and transport measurements High quality β-Ga2O3 samples grown by Edge-defined Film-fed Growth (EFG) method were obtained from Tamura Inc., Japan. Some samples (5 mm × 5 mm × 0.5 mm) were placed in a quartz ampoule with one end open and connected to a vacuum pump to pump out the air and evacuate the ampoule. The tube was then filled with H2 gas at a pressure of 580 Torr. After filling the tube with hydrogen, the open end was properly sealed. The ampoule was placed in an oven that allows precise control of temperature. The temperature was increased to the desired value in two steps and H2 was allowed to diffuse into the crystal for 1 or 2 hours. A few other samples of the same dimensions were first annealed at 950 °C in oxygen flow and then annealed in hydrogen following the same procedure, while other samples were first annealed in gallium and then annealed in hydrogen following the same procedure.

[0050] Hall effect measurements by Van der Pauw method were performed to characterize the electrical transport properties of the samples. Measurements were performed at room temperature (298 K) and at a constant magnetic field of 9300 G. Four indium contacts were arranged in a square on the surface of each sample and carefully adjusted to keep the contacts as small as possible. The current-voltage linearity was checked every time to ensure that the contacts were good and that the resistivity did not vary by more than 10% between different contacts.

[0051] Calculation of donor / acceptor ionization energies by thermoluminescence spectroscopy. Thermoluminescence (TL) is the emission of light from a material by thermal stimulation after irradiating the sample with ionizing radiation at low temperatures. It is a powerful technique to calculate the energy levels of defects that trap charge carriers (e.g. electrons / holes) at low temperatures. This phenomenon can be explained by the energy band theory of solids. In an ideal semiconductor, at low temperatures, most of the charge carriers (e.g. electrons / holes) reside in the valence band. Electrons can be excited to the conduction band (holes to the valence band) by excitation. Wide band gap materials often have structural defects that can trap charge carriers. Donor / acceptor states can also be considered as defects that trap charge carriers at low temperatures. A thermal stimulus can release the electrons / holes from these traps, which transfer their energy to the luminescence center. A schematic diagram of the TL process is shown in Figure 6A-6B.

[0052] TL measurements were performed using a home-built spectrometer that recorded the emission as a function of temperature and wavelength. Measurements were performed from -190 °C to 250 °C. The sample was first placed in a light-tight compartment and irradiated with UV light using a pulsed xenon lamp at -190 °C for 30 min. The temperature was precisely regulated using liquid nitrogen, a water pump, and a digital temperature controller. After irradiation, the temperature of the sample was set to increase at a constant rate (600 °C / min) and emission spectra were recorded from 200 to 800 nm every 5 s. Glow curves representing the emission intensity as a function of temperature were constructed from the integral of the emission over wavelength at each temperature.

[0053] The donor / acceptor ionization energies were calculated by the initial rise method. Randall and Wilkins simplified the thermoluminescence model by assuming negligible retrapping, linear heating rates, and developed the well-known Randall-Wilkins linear expression for the TL intensity:

[0054]

number

[0055] where s is the frequency factor, which is taken as constant in the simplified model, T is the absolute temperature, k is the Boltzmann constant, ED is the donor / acceptor ionization energy, n0 is the total number of trapped electrons / holes at t=0, and β is a constant heating rate. The symmetric shape of the peaks for the samples indicates second order or higher kinetics where significant retrapping of charge carriers after detrapping occurs. A similar equation has been derived for second order kinetics where significant retrapping occurs:

[0056]

number

[0057] Initially, the intensity of the glow peak is dominated by the first exponential of equations (1) and (2), and the second half can be ignored. As a result, when ln(I) is plotted against the initial points of the glow peak, a straight line is obtained, and from its slope, the donor / acceptor ionization energy, ED, can be calculated. Linear fits of ln(I) versus 1 / T for n-type (FIG. 3A) and p-type (FIG. 3B) are shown in FIG. 7A-FIG. 7B, respectively. The donor ionization energy of the hydrogen-diffused β-Ga2O3 sample following annealing in oxygen (FIG. 7A) and the acceptor ionization energy of the hydrogen-diffused β-Ga2O3 sample (FIG. 7B) were found to be 20 meV and 42 meV, respectively.

[0058] A computational method for calculating hydrogen bond energies using density functional theory (DFT) Hydrogen incorporation into Ga vacancies was investigated using density functional theory implemented in the Vienna ab-initio Simulation Package (VASP). These calculations were performed for a 1 × 4 × 2 supercell of β-Ga2O3 containing a total of 160 atoms in the defect-free structure. A 2 × 2 × 2 Monkhorst-Pack k-point mesh centered at the Γ point was used to sample the Brillouin zone. The energy cutoff for plane waves was 400 eV. Pseudopotentials were used based on the Projector Augmented Wave method and the Perdew, Burke, and Ernzerhof (PBE) generalized gradient approximation (GGA) exchange-correlation functional. In general, calculations were continued until the maximum component of the force for any atom was less than 0.02 eV / Å, with one exception (the charged Ga vacancy) where such tight convergence was not possible. In this case, the maximum force was 0.024 eV / Å. Both a monopole correction (using a calculated dielectric constant of 4.16, slightly higher than previously reported) and an alignment correction were applied to the energy. Instead of averaging the potential to make the alignment correction, we simply shifted the density of states so that the deepest states in the material are aligned across the different structures, which has been shown to produce a similar correction. In both cases, the magnitude of this correction was less than 0.1 eV.

[0059] Ga vacancies were created by removing tetrahedrally coordinated Ga ions from the cell, since this vacancy structure was identified as being more favorable. A net charge of -3 was imparted to the structure. A hydrogen ion with a charge of +1 was inserted into the resulting vacancy structure to reduce the net charge of the cell while keeping the total number of electrons in the system constant. The resulting binding energies for each configuration were calculated via the following relationship:

[0060]

number

[0061] In the formula, E(V Ga N.H. (-3+N) ) is the energy of the system in which Ga vacancies are filled with N H+ ions, E(Bulk Ga2O3) is the energy of defect-free β-Ga2O3, and E(V Ga 3- ) is the energy of an isolated Ga vacancy in the 3-charge state, and E(H + ) is the energy of an isolated interstitial 1+H. In this definition, negative energies indicate exothermic or favorable reactions. A systematic search for the lowest energy H interstitial site was not performed, but multiple minimizations were performed in which H was moved randomly to find plausible structures. The structures found here are + The structure is one in which the ion is bonded to one of the three-coordinate oxygen ions.

[0062] Positron Annihilation Lifetime Spectroscopy (PALS) and Doppler Broadening of Positron Annihilation Spectroscopy (DBPAS) Doppler broadening in positron annihilation spectroscopy (DBPAS) was measured using a monoenergetic positron beam with variable energy. Positrons were generated from a high-intensity 22Na source and a tungsten moderator, and were focused at discrete energies E p E p The positrons were accelerated to the range of 0.05 to 35 keV. p allows a penetration depth of about 1.8 μm in Ga2O3. pDoppler broadening spectra, which represent the positron annihilation distribution for 1.09±0.01 keV, were acquired using a single crystal high purity germanium detector with an energy resolution of 1.09±0.01 keV at 511 keV. Doppler broadening spectroscopy was performed on the recorded 511 keV peak originating from electron-positron annihilation. The difference in the electron momentum distribution is represented by the Doppler shift of the annihilation photons. The low electron momentum part of the 511 keV peak is characterized by the so-called S parameter (shape parameter), which is defined as the number of annihilation events in the energy window of about 511±0.742 keV, which is exactly in the middle of the peak, normalized to the total number of events. It reflects the proportion of positron annihilation with valence electrons. On the other hand, the high electron momentum part of this peak is represented by the so-called W parameter (wing parameter), which is defined as the number of annihilation events from the energy windows of about 508-509 keV and about 513-514 keV, which are the tails of the spectrum, normalized to the total number of events. It reflects the rate of positron annihilation with inner shell electrons. Positron trapping at defects causes positron annihilation with valence electrons and an increase in the S parameter, and positron annihilation with inner shell electrons and a decrease in the W parameter. To be able to perform such an analysis, a very large number of statistics should be obtained. Here, each E p At least 1 to 1.5 × 10 5 Obtain a spectrum for the 511 keV peak consisting of a total count of 10 ... p The calculation was made as follows.

[0063] Positron annihilation lifetime spectroscopy (PALS) has been established as an effective method to study cation-vacancy related defects, distinguishing between their types and providing information on their concentrations. Positron lifetime experiments were performed at the Pulsed Monoenergetic Positron Spectroscopy beamline. Lifetime spectra were obtained for each positron energy E up to 16 keV. p FIG. 8 shows the E for two samples of H following annealing in H and O. pThe positron lifetime spectrum obtained at 100 kV is shown. A digital lifetime CrBr3 scintillation detector with home-built software was used, using an SPDevices ADQ14DC-2X with 14-bit vertical resolution and 2 GS / s horizontal resolution. It was optimized for room temperature measurements with a time resolution of 205 ps. The required resolution for the spectral analysis was determined by the positron incident energy E p and two Gaussian functions whose intensities vary with the appropriate relative shift. All spectra were measured with a gradient of at least 5·10 6 The positron lifetime spectrum is the sum of time-dependent exponential annihilations N(t) = Σ i I i / τ i ·exp(-t / τ i ) convolved with a Gaussian function representing the time resolution of the spectrometer, using the nonlinear least-squares based package PALSfit fitting software. An yttria-stabilized zirconia (YSZ) reference sample with a well-defined single-component positron lifetime t ≈ 181 ps was used as a correction spectrum to take into account unwanted background by subtracting extra components not related to the sample positron lifetime. The calculated positron lifetime components from this analysis and their intensities provide an indication of the defect type and density, respectively.

[0064] Temperature-dependent transport properties, XRD measurements, TSL measurements, and PL measurements 9A-9D show temperature dependent transport properties of n-type and p-type H-treated β-Ga2O3 samples. FIG. 9A shows the sheet resistance. FIG. 9B shows the sheet number. FIG. 9C shows the log of the sheet number plotted as a function of 1000 / T. FIG. 9D shows the temperature dependence of n-type mobility. The mobility is 100 cm at room temperature. 2 / VS, which was normalized to the maximum value at low temperature due to the noise of the cryostat system.

[0065] Figure 10 shows X-ray diffraction (XRD) measurements of the as-grown sample, the H-diffused p-type sample, and the O-annealed and H-diffused n-type sample. These measurements show that H-diffusion or annealing did not affect the crystallinity and orientation of the samples.

[0066] Further thermally stimulated luminescence (TSL) measurements were performed at higher temperatures to check for the presence of deep traps in the samples. The defect (trap) energy levels were calculated from the glow curves using the initial rise method. Figures 11A-11B show the TSL glow curves of the as-grown sample (Figure 11A) and the p-type H-diffused sample (Figure 11B), showing the deep traps in the samples. Figure 11C shows the location of these levels in the band gap. These measurements show how hydrogen in Ga vacancies lowers the energy levels of the Ga vacancies, by converting them from deep to shallow acceptors. These measurements also confirm how p-type conductivity is induced by lowering the acceptor levels.

[0067] Figures 12A-B show TSL glow curves for the as-grown sample (Figure 12A) and the n-type O annealed and H diffused sample (Figure 12B), showing the deep traps in the sample. Figure 12C shows a diagram showing the location of these levels in the bandgap. These measurements show how H diffusion after O annealing converts the deep acceptor levels into shallow donors.

[0068] 13A-13D show contour plots of emission intensity as a function of temperature and wavelength for the as-grown sample (FIG. 13A), the H-diffused sample (FIG. 13B), and the H-diffused sample following O annealing (FIG. 13C). Figure 13D shows green emission from the H-diffused p-type sample.

[0069] Photoluminescence (PL) measurements reveal a strong UV emission at 380 nm (3.26 eV) in the p-type H-diffused sample. This emission is attributed to the formation of shallow acceptors. Figure 14A shows the PL emission of the as-grown and p-type H-diffused samples. Figure 14B shows two diagrams, the right one showing the mechanism of the 380 nm UV emission and the left one showing the mechanism of the broadband emission (blue and green).

[0070] Certain embodiments of the compositions, devices, and methods disclosed herein are clarified in the above examples. It should be understood that these examples, while showing certain embodiments of the present invention, are given by way of illustration only. From the above discussion and these examples, one skilled in the art can ascertain the essential features of the present disclosure, and can make various changes and modifications to adapt the compositions, devices, and methods described herein to various uses and conditions without departing from the spirit and scope of the present disclosure. Various modifications may be made, and equivalents may be substituted for the elements of the present disclosure without departing from the essential scope of the present disclosure. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the present disclosure without departing from the essential scope of the present disclosure.

Claims

1. A composition comprising an oxide semiconductor material having cation vacancies filled with hydrogen, The oxide semiconductor material is Ｃａ ) １－ a complex, Ca representing said cation, said oxide semiconductor material having p-type conductivity, or The oxide semiconductor material is Ｃａ ) １＋ a complex, Ca represents the cation, and the oxide semiconductor material has n-type conductivity; The oxide semiconductor is Ga ２ O ３ Including, composition.

2. p-type Ga ２ O ３ A composition comprising: The p-type Ga ２ O ３ contains hydrogen atoms as dopants, and the hydrogen atoms are ２ O ３ present in the Ga vacancy, composition.

3. The oxide semiconductor material is n-type Ga ２ O ３ The composition of claim 1 comprising:

4. The n-type Ga ２ O ３ The composition of claim 3 , wherein the composition has a crystal structure with four hydrogen atoms in the Ga vacancies.

5. The n-type Ga ２ O ３ is at least about 10 １６ cm －２ and at least about 100 cm at room temperature. ２ The composition of claim 4 having both a mobilities of 1:1 and 1:

2.

6. The n-type Ga ２ O ３ The composition of claim 3, wherein the film has a thickness in the range of about 100 nm to about 900 nm.

7. The n-type Ga ２ O ３ But about 10 －４ The composition of claim 3 having a resistivity of Ω-cm.

8. 1. A method of bipolar doping comprising the steps of: partially filling cation vacancies in an oxide semiconductor material with hydrogen, thereby reducing the acceptor state of the vacancies to act as shallow acceptors, thereby doping the oxide semiconductor material p-type; or filling the cation vacancies with hydrogen and additional H ions, thereby doping the oxide semiconductor material n-type; 13. A method comprising:

9. 9. A method for p-doping an oxide semiconductor material, comprising the steps of: placing an oxide semiconductor material in a sealed system; evacuating air from said sealed system; introducing hydrogen gas into said sealed system; and annealing the oxide semiconductor material in the sealed system at an elevated temperature for a duration to diffuse the hydrogen gas into the oxide semiconductor material, thereby doping the oxide semiconductor material p-type; A method comprising:

10. The oxide semiconductor material is Ga ２ O ３ 10. The method of claim 9, comprising:

11. The oxide semiconductor material is Ga ２ O ３ 10. The method of claim 9, comprising: said elevated temperature is about 950° C., said duration is about 2 hours, and said sealed system is at a pressure of about 580 Torr during said annealing.

12. 9. A method for n-doping an oxide semiconductor material, comprising: annealing the oxide semiconductor material in air at a first temperature for a first duration; placing the oxide semiconductor material in a sealed system; evacuating air from said sealed system; introducing hydrogen gas into said sealed system; and annealing the oxide semiconductor material in the sealed system at a second elevated temperature for a second duration to diffuse the hydrogen gas into the oxide semiconductor material, thereby doping the oxide semiconductor material n-type; A method comprising:

13. The oxide semiconductor material is Ga ２ O ３ The method of claim 12, comprising:

14. The oxide semiconductor material is Ga ２ O ３ 13. The method of claim 12, comprising: said first duration is about 2 hours, said first temperature is about 950° C., said second duration is about 2 hours, said second temperature is about 950° C., and said sealed system is at a pressure of about 580 Torr during said annealing.

15. 10. The method of claim 8, wherein a hydrogen plasma is used in a plasma reactor to partially fill cation vacancies in the oxide semiconductor material with hydrogen or to fill the cation vacancies with hydrogen and additional H ions.

Images