Mott conduction in doped iii-nitride semiconductors

N-polar AIGaN grown on C-face 4H-SiC substrates via MBE with controlled growth conditions achieves ultralow resistivity and Mott conduction, addressing doping challenges in high Al content AIGaN for advanced semiconductor devices.

WO2025255569A1PCT designated stage Publication Date: 2025-12-11THE RGT UNIV OF MICHIGAN
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
PCT/US2025/032858
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-06-07
Filing Date
2025-06-09
Publication Date
2025-12-11

AI Technical Summary

Technical Problem

Attaining high levels of doping in high Al content AIGaN to facilitate ultralow resistivity is challenging due to increased activation energy, self-compensation, and mobility constraints, exacerbated by growth on substrates with high dislocation densities and poor surface morphology, particularly in N-polar AIGaN.

Method used

Growth of N-polar AIGaN on C-face 4H-SiC substrates using molecular beam epitaxy (MBE) under nitrogen-rich conditions, with controlled growth temperature and N2 flow rate, achieving a high aluminum content and efficient Si doping to induce Mott conduction.

Benefits of technology

Realizes highly conductive AIGaN epilayers with a resistivity of 1 mΩ cm, the lowest reported, and a Mott transition to semi-metallic conductivity, suitable for high electron mobility transistors and other devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2025032858_11122025_PF_FP_ABST
    Figure US2025032858_11122025_PF_FP_ABST
Patent Text Reader

Abstract

A device includes a substrate and a heterostructure supported by the substrate, the heterostructure including a III-nitride-based buffer layer and a III-nitride-based conductive structure supported by the III-nitride-based buffer layer. The III-nitride-based conductive structure is doped to a level such that the III-nitride-based conductive structure exhibits Mott conduction.
Need to check novelty before this filing date? Find Prior Art

Description

MOTT CONDUCTION IN DOPED lll-NITRIDE SEMICONDUCTORSCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of U.S. provisional application entitled “Mott Conduction in Doped Ill-Nitride Semiconductors,” filed June 7, 2024, and assigned Serial No. 63 / 657,498, the entire disclosure of which is hereby expressly incorporated by reference.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under Contract No. W911 NF-23- 1-0142 awarded by U.S. Army Research Office. The government has certain rights in the invention.BACKGROUND OF THE DISCLOSUREField of the Disclosure

[0003] The disclosure relates generally to Ill-nitride heterostructures.Brief Description of Related Technology

[0004] Recent technological interest in Ill-nitride semiconductors has been centered around ultra-wide bandgap materials, owing to the unique advantages that these materials offer both in terms of high operational voltages, stability at high temperatures, high frequency operation, and the potential to achieve high-performance optoelectronic devices in the deep ultraviolet (UV) regime. In this context, AlxGai-xN has positioned itself as the material of choice with its direct tunable ultrawide-bandgap, large breakdown field, low intrinsic carrier concentration, and the ability to be doped both n and p-type.

[0005] In degenerately doped semiconductors, the Mott transition takes place beyond a certain value of doping, at which electrons directly hop between orbital states located at adjacent dopant sites. Such a transition to semi-metallic conductivity has been commonly observed in Si, GaAs, and GaN. However, attaining such high levels of doping to facilitateultralow resistivity in high Al content AIGaN has been a significant challenge. AIGaN with Al content greater than 60% may exhibit a resistivity as high as 105Q cm.

[0006] Silicon (Si) is the commonly used n-type dopant for lll-N materials, functioning as a substitutional shallow donor at the group-ill site. In high aluminum content AIGaN, the challenge in doping arises from a combination of several factors. The activation energy increases significantly, rising from approximately 15 meV in GaN to > 100 meV in AIN. Additionally, Si exhibits self-compensation behavior at elevated doping levels in high aluminum content AIGaN, resulting in decreased electron concentration and increased resistivity. Furthermore, the mobility in AIGaN is constrained by alloy scattering, a limitation not observed in GaN. These problems are further exacerbated in connection with N-polar AIGaN, as the growth on commonly used substrates like sapphire and silicon is plagued with the presence of very high density of dislocations, due to the large lattice mismatch, the formation of inversion domains, and poor surface morphology. Thus, as the properties of spontaneous and piezoelectric polarization are leveraged to facilitate the creation of high Al content AIGaN channel high electron mobility transistor (HEMT) and other devices, a significant challenge lies in improving the selective regrowth of heavily doped AIGaN sourcedrain ohmic contacts to attain high current density and avoid Ohmic heating, particularly when employing a higher Al mole fraction in the channel.SUMMARY OF THE DISCLOSURE

[0007] In accordance with one aspect of the disclosure, a device includes a substrate and a heterostructure supported by the substrate. The heterostructure includes a Ill-nitride-based buffer layer and a Ill-nitride-based conductive structure supported by the Ill-nitride-based buffer layer. The Ill-nitride-based conductive structure is doped to a level such that the Ill- nitride-based conductive structure exhibits Mott conduction.

[0008] In accordance with another aspect of the disclosure, a method of fabricating a heterostructure includes providing a substrate, forming a Ill-nitride-based buffer layer of the heterostructure, the Ill-nitride-based buffer layer being supported by the substrate, and implementing a molecular beam epitaxy (MBE) procedure to grow a Ill-nitride-based structure of the heterostructure such that the Ill-nitride-based structure is supported by the Ill-nitride-based buffer layer. The MBE procedure is implemented at a growth temperature that falls in a range from about 200 degrees Celsius to about 950 degrees Celsius with a nitrogen flow rate that falls in a range from about 0.3 seem to about 2.0 seem.

[0009] In connection with any one of the aforementioned aspects, the devices and / or methods described herein may alternatively or additionally include or involve any combination of one or more of the following aspects or features. The Ill-nitride-based conductive structure has a dopant concentration falling in a range from about 2 x 1019cm'3to about 5 x 1020cm3. The level at which the Ill-nitride-based conductive structure is doped is sufficient to establish a charge carrier concentration in the Ill-nitride-based conductive structure on the order of 1020cm3. The Ill-nitride-based conductive structure is configured as a source / drain ohmic contact. The Ill-nitride-based conductive structure has an aluminum concentration at or above about 50%. The Ill-nitride-based conductive structure has an aluminum concentration at or above about 60%. The Ill-nitride-based conductive structure is doped n-type. The Ill-nitride-based conductive structure is doped with silicon. The Ill- nitride-based conductive structure has a nitrogen (N)-polar orientation. The Ill-nitride-based conductive structure is in contact with the Ill-nitride-based buffer layer. The Ill-nitride-based buffer layer is in contact with the substrate. The substrate includes silicon carbide. The substrate has a carbon (C)-face. The Il-nitride-based buffer layer is in contact with the C- face of the substrate. The Ill-nitride-based buffer layer includes AIN. The Ill-nitride-based conductive structure includes AIGaN. The MBE procedure is implemented at a growth temperature that falls in a range from about 650 degrees Celsius to about 700 degrees Celsius. The MBE procedure has a lll / V ratio of that falls in a range from about 1 to about 3. The MBE procedure has an aluminum flux that falls in a range from about 4 x 10'8mbar to about 2 x 10'7mbar and a gallium flux that falls in a range from about 4 x 10'8mbar to about 5 x 10'7mbar. The MBE procedure has a nitrogen flow rate that falls in a range from about 0.5 seem to about 0.7 seem. The MBE procedure is implemented under nitrogen (N)-rich conditions. The substrate includes silicon carbide, the substrate has a carbon (C)-face, and the Ill-nitride-based buffer layer is in contact with the C-face of the substrate.BRIEF DESCRIPTION OF THE DRAWING FIGURES

[0010] For a more complete understanding of the disclosure, reference should be made to the following detailed description and accompanying drawing figures, in which like reference numerals identify like elements in the figures.

[0011] Figure 1 depicts (a) a schematic representation of an epitaxially grown 200 nm AIGaN heterostructure on AIN / SiC in accordance with one example, and (b) an AFM image of the AIGaN epilayer, with an inset depicting a streaky RHEED pattern observed in- situ after the growth of AIGaN, from which the r.m.s. roughness of the epilayer extracted to be 0.25 nm over a 1 pm x 1 pm scanning area.

[0012] Figure 2 depicts graphical plots of (a) photoluminescence spectroscopy results showing the room-temperature emission for AIGaN samples A and B, with the band edge transition peak at 260 nm (Pi) and a clear asymmetric broadening observed for sample B and C, and (b) photoluminescence spectra of AIGaN samples C, D and E, showing the presence of a distinct secondary peak P2at about 280 nm for samples D and E.

[0013] Figure 3 depicts graphical plots showing a comparative analysis of (a) electron concentration, (b) mobility, and (c) resistivity of example AIGaN epilayers grown under specified conditions (i.e., Condition A and Condition B described herein) as a function of Si cell temperature (Note: Six samples were grown for each of these two conditions (A1 ,....,A6) and (C1 ,....,C6), with the Si cell temperatures varied from 1200 °C to 1360 °C).

[0014] Figure 4 depicts graphical plots of temperature dependent Hall measurements showing the independence of (a) electron concentration (b) mobility and (c) sheet resistance with varying temperatures for the optimized growth conditions (AIGaN sample C with TSi cell = 1330 °C) indicating the attainment of Mott transition induced metallic conductivity state for heavily doped AIGaN:Si films, as well as of (d) comparison of the free electron concentration and resistivity of an example Si-doped AIGaN film and some other reports in the literature.

[0015] Figure 5 is a cross-sectional, schematic view of a high-electron mobility transistor (HEMT) device having AIGaN ohmic contacts sufficiently heavily doped to exhibit Mott conduction in accordance with one example.

[0016] Figure 6 is a flow diagram of a method of fabricating a heterostructure having a Ill- nitride-based conductive structure in accordance with one example.

[0017] Figure 7 depicts (a) a schematic view of a transfer length method (TLM) arrangement for measuring the contact resistivity of a doped AIGaN layer in accordance with one example, (b) a plan view of a layout pattern of the TLM arrangement, (c) a graphical plot of l-V curves for example doped AIGaN layers in the TLM arrangement, and (d) a graphical plot of contact resistance for the example doped AIGaN layers in the TLM arrangement.

[0018] The embodiments of the disclosed systems and methods may assume various forms. Specific embodiments are illustrated in the drawing and hereafter described with the understanding that the disclosure is intended to be illustrative. The disclosure is not intended to limit the invention to the specific embodiments described and illustrated herein.DETAILED DESCRIPTION OF THE DISCLOSURE

[0019] Devices and heterostructures having a Ill-nitride-based conductive structure doped to a level to attain Mott conduction are described. Such semi-metallic conduction allow the Ill-nitride-based conductive structures to be configured as an ohmic contact, such as a source / drain ohmic contact of a high electron mobility transistor (HEMT) device. As described herein, the Ill-nitride-based conductive structure may be composed of, or otherwise include, nitrogen (N)-polar AIGaN with a high aluminum content (e.g., about 50% or more). Methods for fabricating such devices are also described.

[0020] In some cases, the heterostructures may be supported by a 4H-silicon carbide (SiC) substrate with a C-face orientation, which are well-suited for the epitaxy of N-polar AIN and AIGaN. The Al content of N-polar AIGaN on 4H-SiC substrates may be tuned or increased in connection with high levels of Si doping. The disclosed devices and methods support a reduction in the resistivity for Al-rich AIGaN, for example, for achieving regrown contact access layers in AIGaN channel HEMT devices with sufficiently high conductivity.

[0021] Described herein are a number of examples that exhibit the attainment of semimetallic conductivity in Si-doped N-polar AIGaN grown epitaxially on C-face 4H-SiC substrates with high Al content (about 60 %). By controlling the growth temperature and the lll / V ratio of the growth procedure, a growth window was found in which a highly conductive AIGaN epilayer is obtained with atomically smooth surface morphology. The example growth conditions realize AIGaN epilayers with distinct photoluminescence characteristics, a notable reduction in defect peaks, and a simultaneous enhancement in the emission intensity. Hall measurements in the standard Van der Pauw configuration reveal the highest measured electron concentration of about 1 .8 x 1 O20cm'3, with a high mobility of 34 cm2 / V-s, leading to a room temperature resistivity of only 1 mfi cm, which is the lowest reported resistivity in Alo.6Gao.4N to date. Furthermore, temperature dependent Hall measurements were performed to elucidate the charge carrier transport characteristics in these films, in which a clear signature of Mott conduction was observed for these highly conductive films, with the vanishing of the dopant activation energy.

[0022] The conductive Ill-nitride-based structures of the disclosed devices may be grown under nitrogen (N)-rich conditions. As described herein, the N-rich conditions may be established despite a lll / V ratio higher than 1 - e.g., greater than about 2. Such higher lll / N flux ratios do not necessarily result in metal rich conditions at the growth front. Indeed, N- rich conditions can also be maintained in connection with such high lll / N flux ratios at the growth front under certain conditions. For instance, at relatively high temperatures, metalatoms will be desorbed from surfaces much faster than N atoms, leading to N-rich conditions at the growth front.

[0023] The conductive Ill-nitride-based structures of the disclosed devices may be grown via molecular beam epitaxy (MBE). Compared to metal-organic chemical vapor deposition (MOCVD), MBE offers several advantages for the epitaxy of AIGaN and other Ill-nitride- based materials, including simple and more controllable growth chemistry, reduced impurity incorporation, efficient dopant incorporation, easier polarity control, and atomically sharp interfaces.

[0024] Although described herein in connection with examples involving N-polar AIGaN conductive structures, the disclosed devices, heterostructures, and methods may be directed to other Ill-nitride-based layers. A wide variety of Ill-nitrides and Ill-nitride alloys may be grown. For instance, the disclosed methods may be directed to forming a heterostructure with an N-polar layer of InGaN, ScAIN, YAIN, or AIBN.

[0025] Although described herein in connection with examples involving C-face 4H-SiC substrates, the disclosed devices, heterostructures, and methods may utilize other substrates and substrate materials. For instance, other SiC substrates may be used, including those having a different polytype or structure. In some cases, the substrate may be composed of, or otherwise include, SiC having a 6H structure. The substrate may alternatively or additionally be composed of, or otherwise include, other materials, such as sapphire, silicon, and aluminum nitride.

[0026] Although described herein in connection with examples involving AIN buffer layers, the disclosed devices, heterostructures, and methods may utilize other buffer layer materials. For instance, the buffer layers may alternatively or additionally be composed of other Ill-nitride materials, such as BN, GaN, or AIGaN.

[0027] Although described herein in connection with source / drain contacts for HEMT devices, the conductive Ill-nitride-based structures and heterostructures described herein are useful in connection with a wide variety of electronic and optoelectronic devices. For instance, the heterostructures may be used in various emissive and non-emissive optoelectronic devices, such as detectors. Moreover, other types of emissive devices may also be formed, including, for instance, various types of laser devices. Additionally, the heterostructures may be used in light modulation devices, such as an optical modulator.

[0028] With reference to Figure 1 , a Veeco GENxplor MBE system was used for the epitaxial growth of a number of example N-polar films. The system is equipped with a radio frequency (RF) nitrogen plasma source (with a purity of 99.9999%) and Knudsen effusioncells for Ga (99.99999% purity), Al (99.99995% purity), and Si (99.999%). Before introducing a C-face 4H-SiC substrate 100 into the MBE chamber, a standard solvent cleaning process was performed with acetone, isopropyl alcohol (IPA), and deionized (DI) water. Subsequently, the substrates 100 were baked and outgassed at 200 °C and 600 °C in the MBE load-lock and preparation chamber, respectively. Si-doped AIGaN epilayers 102 of these examples were 200 nm thick and grown on top of a 200 nm AIN buffer layer 104 (shown in Figure 1 , part a).

[0029] The AIN buffer layer 104 may be useful in the growth of the N-polar AIGaN films 102. For Si doping of high Al content AIGaN, the challenges in attaining a satisfactory level of doping may arise significantly from residual oxygen impurities. This is not only due to their involvement in the deep donor level (DX) transition but also because, when combined with a high concentration of dislocations, oxygen can boost the formation of cation vacancies, serving as compensation centers. AIN, having a lattice mismatch of lower than 1% with the underlying SiC substrate, not only produces an epitaxially smooth growth front for the subsequent growth of AIGaN, but also minimizes the incorporation of O impurities from the substrate in the epilayer, which eventually helps in attaining highly conductive Si doped AIGaN epilayers. Further details regarding the growth conditions of example AIN epilayers are set forth in Hu et al., "Heteroepitaxy of N-polar AIN on C-face 4H-SiC: Structural and optical properties," APL Mater. 11 , 121111 (2023), the entire disclosure of which is hereby incorporated by reference.

[0030] As described herein, highly conductive Si doped AIGaN epilayers have been realized via a growth window involving epitaxy parameters such as growth temperature and the N2flow rate. Higher growth temperatures typically result in a smooth surface morphology, as described in Mondal et al., "Tunable bandgap and Si-doping in N-polar AIGaN on C-face 4H-SiC via molecular beam epitaxy," Appl. Phys. Lett. 123, 182106 (2023), the entire disclosure of which is hereby incorporated by reference. On the other hand, the surface morphology deteriorates at lower growth temperatures owing to the low adatom mobility of a N-polar surface. However, a lower growth temperature results in fewer Al vacancies and Si-complexes, which may be useful for enhancing Si doping efficiency in AIGaN films, as elaborated upon later.

[0031] The disclosed devices and methods may adjust or specify the N2flow rate of the growth procedure to achieve a smooth surface morphology at a reduced growth temperature of, for instance, 680 °C, measured from the pyrometer reading. A higher growth rate, as obtained by increasing the N2flow, may also help in reducing the resistivity of Si doped AIGaN films, as has been recently reported.

[0032] Figure 1 , part b, shows an AFM image of an example AIGaN epilayer grown via the disclosed methods. The r.m.s. roughness was extracted to be 0.25 nm for a 1 pm x 1 pm scanning area. The in-situ RHEED observed during the epitaxy of AIGaN is narrow and streaky, indicative of a smooth surface morphology. Furthermore, the N-polar orientation of the epitaxially grown AIGaN films was confirmed by polarity sensitive wet-etching.

[0033] Obtaining a high conductivity level in AIGaN with a substantial Al content poses a formidable challenge. Not only does the activation energy increase significantly, but AIGaN with high Al content is also susceptible to high dislocation densities and compensation arising from acceptor-like defects, particularly Al vacancies. Room temperature photoluminescence experiments were carried out to investigate the effects of the growth parameters in suppressing these defect states in the epitaxially grown AIGaN films. Excitation of the example films was carried out using a 193 nm ArF excimer laser, and the photoluminescence emission was gathered and examined using a Horiba iHR550 spectrometer equipped with a UV-sensitive Symphony II CCD detector.

[0034] Table I summarizes the growth conditions (temperature, N2flow, and the metal fluxes) of the AIGaN films being analyzed using photoluminescence spectroscopy.Table I: Key growth parameters of samples being investigated in the present work

[0035] Figure 2, part a, shows the comparison of room temperature PL characteristics of three sample films A, B, and C. Sample A was grown under a metal rich regime (0.3 seem N2flow) at a relatively higher temperature. For sample B, the N2flow was kept the same, however, the total metal flux and the growth temperature were reduced, and sample C was grown at similar growth temperature as Sample B but with a higher N2flow of about 0.6 seem.

[0036] It can be clearly seen that the spectral linewidth of Sample A is 2.3 times larger than Sample C grown, which exhibits a narrow linewidth of only 200 meV. This asymmetric linewidth broadening towards the lower energy side can be attributed to the formation of Ga rich clusters in Sample A and B. The linewidth broadening further increases due to increased Ga-clusters at a lower growth T under the M-rich regime. However, increasing the N2 flow at the lower growth temperature, as in Sample C, leads to a reduction of Ga-cluster formation, as is evidenced by the decreasing PL linewidth. The variations in alloy composition were linked to the reduced mobility of Al atoms due to the stronger Al-N bond in AIGaN, leading to Ga-rich inter-domain regions. Thus, by minimizing such clusters, high spectral purity is obtained for the epitaxially grown films.

[0037] Figure 2, part b, shows a comparison of room temperature PL characteristics of three sample films (Samples C, D and E) grown with increasing N2 flow. It can be observed that for an increased N2flow beyond 0.6 seem, as in Sample D and Sample E, a distinct secondary peak (P2) at about 280 nm is observed apart from the band edge transition peak at 260 nm (P1). This may be attributable to the formation of Al vacancies as has been observed for high Al content AIGaN samples grown under N-rich growth conditions. Upon fitting the PL data, it is found that the integrated intensity ratio (P1 / P2) is 7.7 for Sample C, implying that such a defect related emission is hardly noticeable. However, for Sample D it is 1 .51 , and for Sample E it further drops to 0.86. Furthermore, the overall integrated intensity of the band-edge photoluminescence emission is 20 times higher for Sample C compared with Sample E. A significantly high radiative recombination rate is achieved, thereby implying the reduction of significant defects or dislocations in the epitaxially grown AIGaN films under the specified conditions. This indicates that the band-edge peak starts to dominate as the N2flow rate is reduced. However, a sufficiently high N2 flow is useful to minimize the formation of Ga-rich clusters as has been previously observed in photoluminescence studies of high Al content (e.g., x greater than 0.5) AIGaN films.

[0038] Reducing Al vacancies may impact the doping concentration of the films. A decrease in group Ill-vacancies leads to reduced self-compensation in highly silicon-doped AIGaN epilayers. To analyze the impact of growth conditions on the Si doping limits of the samples, two sets of growth conditions were considered. Condition A involved a high metal flux and low N2 flow (similar to Sample A). Condition B involved a low metal flux and high N2 flow (similar to Sample C), as described in Table I. Six samples were grown for each of these two conditions (A1 ,...., A6) and (C1 ,....,C6), while the Si cell temperatures were varied from 1200 °C to 1360 °C. Ti (10 nm) / Au (100 nm) electrodes were patterned by photolithography in the standard van der Pauw configuration, and Hall measurements werecarried out to determine the carrier concentration, mobility, and resistivity of the AIGaN epilayers in each of these cases.

[0039] As can be seen in Figure 3, part a, the carrier concentration steadily increases with increasing Si cell temperature, indicating proportionate carrier incorporation and activation in the AIGaN film. Even though a similar carrier concentration is achieved for a Si cell temperature of 1200 °C for both of the growth regimes, the rate of increase of carrier concentration with increasing Si cell temperature is much more pronounced in Condition B compared to Condition A. The highest electron concentration achieved was about 1 .8 x 1 O20cm'3in Condition B, indicating highly efficient doping incorporation and activation in the optimized epitaxial conditions. On the high doping side, the formation of metal vacancy- related point defects along with the growth temperature determines the limit for the maximum achievable carrier concentration. The growth conditions suppressing such metal vacancies (as discussed in connection with Figure 2), and lower growth temperature (e.g., relative to Condition A) thus avoid the self-compensation of Si by the formation of DX centers. However, upon exceeding the upper doping limit with silicon (Si), the surplus Si incorporated creates an acceptor state, compensating for the remaining Si donors. This emphasizes the role of self-compensation at higher Si levels, ultimately defining an upper doping limit for silicon in this context. As a result, heavily Si-doped AIGaN demonstrates elevated resistivity attributed to a substantial decrease in electron concentration. A similar “knee-behavior” is clearly observed for the investigated samples at a Si cell temperature of about 1330 °C, above which the carrier concentration drops sharply for both samples. This drop in carrier concentration is primarily attributed to the formation of Vm-nSim complexes at elevated Si doping levels.

[0040] Figure 3, part b, shows the mobility of the sample films in the two growth regimes as a function of Si cell temperature. The highest recorded value of mobility for Condition B is 45 cm2 / V-s at a corresponding electron concentration of 2 x 1019cm-3at a Si cell temperature of 1200 °C. For Condition A, the mobility is only about 20 cm2 / V s. Even for the highest recorded electron concentration of 1 .8 x 102° cm'3, the mobility is sustained at 34 cm2 / V-s.

[0041] Carbon has been identified as a species with a negative charge state (CN) in n-type lll-Ns that can act as a compensator and often result in a mobility collapse. The example growth conditions with a higher N2 flow are effective in suppressing the incorporation of CN in AIGaN epilayers, thereby resulting in enhanced mobility values. As seen in the plot of resistivity as a function of Si cell temperature in Figure 3, part c, the lowest recorded room temperature resistivity value is 1 mfi cm. This value is about 3 times lower than the lowest reported resistivity in AIGaN with similar Al content reported to date. The lowest resistivitywas recorded for Si cell temperature of about 1330 °C, beyond which the resistivity increases sharply, consistent with the observed “knee-behavior” in the electron concentration.

[0042] Dislocations also play a role in determining the resistivity of a sample, with an equivalent acceptor concentration equal to their density. The growth of N-polar AIGaN with an AIN buffer on SiC offers less than 1 % lattice mismatch, resulting in highly reduced dislocation densities. Moreover, the observed low variation in carrier concentration (n) and resistivity (p) originates from the semi-metallic state owing to the high doping concentration of the disclosed films. In the above-referenced publication, significant yet controllable variation in carrier concentration (n) and resistivity (p) were observed, but the doping concentration employed in that study was not as high as that used in the disclosed devices and methods. For instance, the doping concentration of the AIGaN films disclosed herein may fall in a range from about 2 x 1019cm-3to about 5 x 1 O20cm3. There exists an AIN / AIGaN interface within the heterostructure that could potentially influence the depicted charge transport characteristics in this manuscript. However, given the substantially higher carrier concentration of the disclosed films, it can be deduced that the interface plays a less significant role in modulating the charge transport properties.

[0043] To gain further insight into the carrier transport mechanism of the lowest resistivity high Al content AIGaN epilayers (Condition B, TSiceii = 1330 °C), temperature-dependent Hall effect measurements were performed in a Quantum Design physical property measurement system (PPMS), and the sample was configured in the standard van der Pauw geometry. The sample was held at each temperature for 30 minutes prior to measurement to allow for adequate environmental equilibration; starting at 10 K. The temperature was increased at a rate of 5 K / min between measurements. Sheet resistance and Hall effect measurements were performed using a Keithley 2450 to source and sense the signals and Keithley 3706A- S to switch between the various van der Pauw configurations. The source current ranged between 100 pA and 1 mA for each sample. The magnetic field was swept from +0.32 T to - 0.32 T in increments of 2 mT for the Hall effect measurements.

[0044] At high Si doping levels, transition to a metallic state was observed for the AIGaN films, in which the activation energy (Ea) falls to about 0 meV. As the conduction does not rely on thermal activation of electrons, the conductivity remains relatively independent of temperature. The incorporation of Si almost reaches the measured electron concentration, due to the Mott transition induced metallic conductivity state. From the Arrhenius plot in Figure 4, part a, the carrier concentration was constant at about 1 .8 x 102° cm'3, i.e., theslope (Ea) is found to be 0 meV, thus indicating a clear transition to the metallic state. Moreover, at high doping levels, as Mott conduction dominates, it strongly reduces the temperature dependence of transport, so mobility stays relatively constant. A similar trend was observed in the example films, with the mobility being constant at 34 cm2 / V-s throughout the entire temperature range, as seen in Figure 4, part b. Figure 4, part c, shows the sheet resistance of the example AIGaN film to be exceptionally low at about 48 Q / sq. Moreover, the sheet resistance does not increase with decreases in temperature, as is generally expected for moderately doped AIGaN films.

[0045] A benchmarking of the free electron concentration and resistivity of the example AIGaN films is presented with other reports in literature in Figure 4, part d. This comparison includes all Si-doped AIGaN films, either M-polar or N-polar, albeit with a similar Al content. It is evident that the lowest resistivity value of 1 mfi cm achieved via the example films of the disclosed devices exceeds the best results documented in the literature for high Al content (about 60%) AIGaN, considering both carrier concentration and resistivity.

[0046] Described above are examples of heterostructures and fabrication methods that realize high Al content, highly conductive N-polar AIGaN films on 4H-SiC substrates. The epitaxially grown films show smooth surface morphology, and optimization of growth conditions have led to the identification of a growth window which helps minimize the formation of both Ga-rich clusters and Al-vacancies. Due to the highly efficient doping, Mott transition to the semi-metallic state was achieved, with an exceptionally high free electron concentration and the lowest recorded resistivity to date.

[0047] The highly conductive high Al content AIGaN epilayers described herein may be used in a variety of contexts. For instance, the epilayers may be used as regrown contact regions in AIGaN channel HEMT devices, such as high-power RF and millimeter-wave HEMTs devices. The epilayers may thus be useful in enhancing the breakdown voltages for power switching applications due to the ultrawide bandgap of AIGaN.

[0048] Figure 5 depicts a HEMT device 500 having conductive AIGaN structures 502 as drain / source ohmic contacts in accordance with one example. In this case, the Al content is about 0.5. The AIGaN structures 502 may be highly doped (n++) and configured for Mott conduction as described herein.

[0049] The HEMT device 500 includes a composite substrate 504 having a base substrate and a buffer layer supported by the base substrate. In this example, the base substrate is composed of sapphire, and the buffer layer is composed of AIN. An AIGaN barrier layer 506is supported by the buffer layer, and includes the drain / source ohmic contacts 502. The drain / source ohmic contacts 502 may be re-grown as shown.

[0050] The HEMT device 500 includes a high Al content AIGaN channel layer 508 between the source / drain ohmic contacts 502 as shown. The channel layer 508 may have a higher Al content than the other AIGaN regions of the device 500.

[0051] Figure 6 depicts a method 600 of fabricating a heterostructure device having an N- polar Ill-nitride-based layer in accordance with one example. The method 600 may be used to manufacture any of the heterostructures described herein or another type of heterostructure. The method 600 may include additional, fewer, or alternative acts. For instance, the method 600 may or may not include one or more acts directed to forming other device structures (act 622).

[0052] The method 600 may begin with an act 602 in which a substrate is prepared or otherwise provided. The substrate may be, be formed from, or otherwise include 4H-SiC. Other SiC substrates may be used, including, for instance, 6H-SiC. Alternative or additional materials may be used, including, for instance, sapphire.

[0053] Preparation of the substrate may include an act 604 in which one or more chemical or thermal cleaning procedures are implemented. Alternatively or additionally, the act 602 include an act 604 in which the substrate is annealed (e.g., in the MBE chamber in which the layer(s) of the heterostructure are grown).

[0054] The method 600 includes an act 610 in which a buffer or template layer of the heterostructure is grown on the substrate. As described herein, the buffer layer may be composed of, or otherwise include, AIN. The act 610 may include implementation of an MBE procedure. The MBE procedure may be implemented in an act 612 in a common (the same) growth chamber used to grow the conductive Ill-nitride layer. The MBE procedure may or may not be implemented under nitrogen-rich conditions as described herein.

[0055] In an act 614, an MBE procedure is implemented to grow a conductive Ill-nitride- based layer. As described herein, the act 614 may include an act 616 in which the MBE procedure is implemented under N-rich conditions. For instance, a lll / V ratio falling in a range from about 1 to about 3 may be used.

[0056] The MBE procedure may be configured to dope the conductive Ill-nitride-based layer (e.g., with silicon) in an act 618. The Si cell temperature may be controlled to achieve a desired dopant concentration, as described herein.

[0057] The N2flow rate, Al flux, Ga flux (or other Group III), and other process parameters may be controlled in an act 620. For instance, the N2flow rate may be about 0.6 seem, and the Al and Ga fluxes may be on the order of 10'8mbar.

[0058] The N2flow rate may vary in other cases, including, e.g., cases involving different Al content and / or growth temperatures. For instance, the N2flow rate may fall in a range from about 0.5 seem to about 0.7 seem. In other cases, the N2flow rate falls in a range from about 0.5 seem to about 0.75 seem, or from about 0.3 seem to about 0.75 seem. In still other cases, the N2flow rate falls in a range from about 0.3 seem to about 2 seem. Alternatively or additionally, the metal flux levels may vary in other cases, e.g., cases involving different Al content and / or growth temperatures. For instance, in other cases the Al flux level may fall in a range from about 4 x 10-8mbar to about 2 x 10-7mbar, and the Ga flux level may fall in a range from about 4 x 10-8mbar to about 5 x 10-7mbar. As a result of these variations, the lll / V ratio of the MBE procedure may vary accordingly. For example, the lll / V ratio may fall in a range from about 1 to about 1 .2 in some cases.

[0059] The MBE procedure may be implemented at a growth temperature of about 680 degrees Celsius, as described herein. The growth temperature may vary in other cases, including, e.g., cases involving different Al content and / or different Ill-nitride alloys. For instance, the growth temperature may fall in a range from about 650 °C to about 750 °C, or in a range from about 650 °C to about 700 °C. In other cases, the growth temperature falls in a range from about 600 °C to about 800 °C. In still other cases, the growth temperature falls in a range from about 200 °C to about 950 °C.

[0060] The act 614 may be implemented after the formation of one or more Ill-nitride layers. For instance, an AIGaN layer may be grown after the buffer layer is grown in the act 610. After the AIGaN layer is patterned, the conductive Ill-nitride layer may be selectively formed in a re-growth procedure to form source / drain contact regions or other structures.

[0061] In some cases, the method 600 may include an act 622 directed to forming one or more structures of a device in which the heterostructure is integrated. For instance, one or more metal layers may be deposited and patterned to form source / drain contacts.

[0062] Figure 7, parts a, b, depict a transfer length method (TLM) arrangement for measuring the contact resistance of a doped AIGaN layer in accordance with one example. Test results shown in Figure 7, parts c, d, for the example AIGaN layers established a contact resistance of 0.09 Q mm, and a very low contact resistivity of 4.72 x 10'8Q cm2.

[0063] Described above are a number of examples of extremely highly doped, high Al content AIGaN epilayers useful for, e.g., regrown source / drain contact regions. Semi-metallicconductivity in Silicon (Si) doped N-polar AI0.eGa0.4N grown on C-face 4H-SiC substrates by Molecular Beam Epitaxy (MBE) was achieved. Under specified conditions, the AIGaN epilayer exhibited smooth surface morphology and a narrow photoluminescence spectroscopy linewidth, without the presence of any secondary peaks. A favorable growth window was identified, in which the free electron concentration reached as high as about 1 .8 x 1 O20cm'3as obtained from Hall measurements, with a high mobility of 34 cm2 / V-s, leading to a room temperature resistivity of only 1 mfi cm. Temperature-dependent Hall measurements established that the electron concentration, mobility, and sheet resistance do not depend on temperature, clearly indicating dopant Mott transition to a semi-metallic state, in which the activation energy (Ea) falls to 0 meV at this high value of Si doping for the AIGaN films. This achievement of semi-metallic conductivity in Si doped in N-polar high Al content AIGaN is useful in a wide variety of ultrawide bandgap electronic and optoelectronic devices and applications.

[0064] The term "about" is used herein in a manner to include deviations from a specified value that would be understood by one of ordinary skill in the art to effectively be the same as the specified value due to, for instance, the absence of appreciable, detectable, or otherwise effective difference in operation, outcome, characteristic, or other aspect of the disclosed methods and devices.

[0065] The present disclosure has been described with reference to specific examples that are intended to be illustrative only and not to be limiting of the disclosure. Changes, additions and / or deletions may be made to the examples without departing from the spirit and scope of the disclosure.

[0066] The foregoing description is given for clearness of understanding only, and no unnecessary limitations should be understood therefrom.

Claims

What is Claimed is:

1. A device comprising: a substrate; and a heterostructure supported by the substrate, the heterostructure comprising: a Ill-nitride-based buffer layer; and a Ill-nitride-based conductive structure supported by the Ill-nitride-based buffer layer; wherein the Ill-nitride-based conductive structure is doped to a level such that the Ill- nitride-based conductive structure exhibits Mott conduction.

2. The device of claim 1 , wherein the Ill-nitride-based conductive structure has a dopant concentration falling in a range from about 2 x 1019cm'3to about 5 x 1 O20cm-3.

3. The device of claim 1 , wherein the level at which the Ill-nitride-based conductive structure is doped is sufficient to establish a charge carrier concentration in the Ill-nitride- based conductive structure on the order of 1020cm-3.

4. The device of claim 1 , wherein the Ill-nitride-based conductive structure is configured as a source / drain ohmic contact.

5. The device of claim 1 , wherein the Ill-nitride-based conductive structure has an aluminum concentration at or above about 50%.

6. The device of claim 1 , wherein the Ill-nitride-based conductive structure has an aluminum concentration at or above about 60%.

7. The device of claim 1 , wherein the Ill-nitride-based conductive structure is doped n- type.

8. The device of claim 1 , wherein the Ill-nitride-based conductive structure is doped with silicon.

9. The device of claim 1 , wherein the Ill-nitride-based conductive structure has a nitrogen (N)-polar orientation.

10. The device of claim 1 , wherein the Ill-nitride-based conductive structure is in contact with the Ill-nitride-based buffer layer.

11. The device of claim 1 , wherein the 11 l-nitride-based buffer layer is in contact with the substrate.

12. The device of claim 1 , wherein: the substrate comprises silicon carbide; the substrate has a carbon (C)-face; and the Ill-nitride-based buffer layer is in contact with the C-face of the substrate.

13. The device of claim 1 , wherein the Ill-nitride-based buffer layer comprises AIN.

14. The device of claim 1 , wherein the Ill-nitride-based conductive structure comprises AIGaN.

15. A method of fabricating a heterostructure, the method comprising: providing a substrate; forming a Ill-nitride-based buffer layer of the heterostructure, the Ill-nitride-based buffer layer being supported by the substrate; and implementing a molecular beam epitaxy (MBE) procedure to grow a Ill-nitride-based structure of the heterostructure such that the Ill-nitride-based structure is supported by the Ill-nitride-based buffer layer; wherein the MBE procedure is implemented at a growth temperature that falls in a range from about 200 degrees Celsius to about 950 degrees Celsius with a nitrogen flow rate that falls in a range from about 0.3 seem to about 2.0 seem.

16. The method of claim 15, wherein the MBE procedure is implemented at a growth temperature that falls in a range from about 650 degrees Celsius to about 700 degrees Celsius.

17. The method of claim 15, wherein the MBE procedure has a lll / V ratio of that falls in a range from about 1 to about 3.

18. The method of claim 15, wherein the MBE procedure has an aluminum flux that falls in a range from about 4 x 10-8mbar to about 2 x 10-7mbar and a gallium flux that falls in a range from about 4 x 10-8mbar to about 5 x 10-7mbar.

19. The method of claim 15, wherein the MBE procedure has a nitrogen flow rate that falls in a range from about 0.5 seem to about 0.7 seem.

20. The method of claim 15, wherein the MBE procedure is implemented under nitrogen (N)-rich conditions.

21. The method of claim 15, wherein: the substrate comprises silicon carbide; the substrate has a carbon (C)-face; and the Ill-nitride-based buffer layer is in contact with the C-face of the substrate.

Citation Information

Patent Citations

  • Heterojunction field effect transistor, method for producing heterojunction field effect transistor, and electronic device

    US20110284865A1

  • Heterostructure with Sacrificial Layer

    US20180374986A1

  • High-voltage p-channel FET based on iii-nitride heterostructures

    US20200144407A1