Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

III-Nitride Short Period Superlattices for Deep UV Light Emitters

III-Nitride Short Period Superlattices for Deep UV Light Emitters applied sciences Review III-Nitride Short Period Superlattices for Deep UV Light Emitters Sergey A. Nikishin Nano Tech Center, Department Electrical and Computer Engineering, Texas Tech University, Lubbock, TX 79423, USA; sergey.a.nikishin@ttu.edu; Tel.: +1-806-834-8807 Received: 17 October 2018; Accepted: 20 November 2018; Published: 23 November 2018 Featured Application: Advanced infrared, visible, and ultraviolet light emitters. Abstract: III-Nitride short period superlattices (SPSLs), whose period does not exceed ~2 nm (~8 monolayers), have a few unique properties allowing engineering of light-emitting devices emitting in deep UV range of wavelengths with significant reduction of dislocation density in the active layer. Such SPSLs can be grown using both molecular beam epitaxy and metal organic chemical vapor deposition approaches. Of the two growth methods, the former is discussed in more detail in this review. The electrical and optical properties of such SPSLs, as well as the design and fabrication of deep UV light-emitting devices based on these materials, are described and discussed. Keywords: III-nitrides; short period superlattices; light emitters 1. Introduction The invention of semiconductor double heterostructure laser [1,2] and the concept of a semiconductor superlattice (SL) [3] can be considered as the foundation of modern semiconductor p–n junction-based light emitters, lasers, and light-emitting diodes. Double heterostructure (DHS) is a semiconductor “sandwich” where a layer of narrow band gap semiconductor (referred to as an active layer or a well, depending on its thickness) is placed between n-type and p-type layers (cladding layers) of the wide bandgap semiconductors. Under forward bias, the electrons and holes are injected from the cladding layers into an active layer, and well confined there. Confinement of these injected carriers leads to very effective radiative recombination in direct bandgap semiconductors. The SL periodic layered structure of different crystalline semiconductors, allows for engineering of a bandgap of active (referred as a well) and cladding (referred as a barrier) layers when their thickness is of about a few nanometers. Note, these structures must be grown into two-dimensional (2D) growth mode, in order to get a flat/abrupt interface between all layers. The SLs based on the direct bandgap semiconductors in different combinations with p–n DHS are used for design and fabrication of light emitters operating in very wide range of wavelengths, ensuring the needs of optical communication, medicine, security, lighting, and agriculture [4–11]. We will discuss only p–n junction-based structures in this review. III-Nitrides, AlN (E = 6.2 eV, [12,13]), GaN (E = 3.4 eV, [13]), InN (E = 0.7 eV, [14–16]), and their g g g AlGaInN, AlGaN, AlInN, InGaN alloys, are direct bandgap semiconductors which would significantly influence the development of new optoelectronic and light-emitting devices throughout the 21st century [17–19]. Most III-nitride (III-N) light-emitting and laser diodes contain different SLs with a period exceeding 4 nm [20–23]. These SLs, in addition to the bandgap engineering, allow for reducing the dislocation density propagating into an active layer of III-N light-emitting diodes (LEDs) grown on the heavily lattice-mismatched substrates, like silicon and sapphire [24–26]. Strain engineering in an active layer of LEDs allows for modification of an internal quantum efficiency of radiative carrier Appl. Sci. 2018, 8, 2362; doi:10.3390/app8122362 www.mdpi.com/journal/applsci Appl. Sci. 2018, 8, 2362 2 of 17 recombination [27,28]. The design and fabrication of such light emitters, based on “long period” SLs, are well described in many books and reviews [29–36]. III-N LEDs based on very short period superlattices (SPSLs, sometimes referenced as digital alloys, DA), periodic structures of GaN/AlN, AlGaN/AlN, AlGaInN/AlN, InGaN/GaN, InN/GaN, and InAlN/GaN having a few monolayer thick wells and barriers, and a period not exceeding 2 nm, are very attractive for the design and fabrication of a new generation of light emitters. The main important difference between SPSLs and SL with a long period, is that carriers tunneling through barriers between quantum wells (QWs) in SPSLs already affect energy levels, and even lead to the formation of minibands (at least in the conduction band). Bandgap behavior of the InGaN/GaN, InN/GaN, and InAlN/GaN SPSLs, and their applications in visible and infrared light emitters, are well summarized in recent publications [37–40]. The GaN/AlN, AlGaN/AlN, and AlGa(In)N/AlN (C < 0.02 mol fraction) SPSLs are very attractive for deep ultraviolet light emitters [41–48]. One of In the attractive features of AlGa(In)N/AlN SPSLs relates to the formation of very sharp heterointerfaces over the entire range of compositions, which makes it possible to obtain well/barrier thicknesses comparable to the interatomic distance, and to make tunneling the main carrier transport mechanism. For InGaN/GaN structures, for example, the roughness of heterointerfaces increases if the composition of In approaches ~20%, which is important for practical applications, since carrier tunneling is difficult. This review aims to summarize the most significant efforts demonstrated in this field since 2002, when the first LED based on AlGa(In)N/AlN SPSLs operating at 280 nm, was demonstrated [41,42]. 2. Growth and Structural Characterization Most of III-N SPSLs for deep UV light emitters were grown using both molecular beam epitaxy (MBE) [41–43] and metal organic chemical vapor deposition (MOCVD) [37,44] methods on (0001) sapphire, Si (111), and (0001) GaN/sapphire template substrates. The detailed analysis of deep UV LED efficiency (internal and external), grown on different substrates, can be found in ref. [44]. One well-known advantage of MBE over MOCVD is the in situ monitoring of the growth process using the reflection high energy electron diffraction (RHEED) [49]. Analysis of the RHEED patterns in real time allows for controlling, at the monolayer (ML) scale, the structural properties of any substrate at the onset of epitaxy, the nucleation process and growth mode of the epitaxial layer, the growth rate of III-N compounds, and the composition of their alloys, by monitoring the period of RHEED intensity oscillations during deposition [50–52]. This statement can be illustrated by a few RHEED patterns. The evolution of RHEED patterns illustrating the onset of gas source MBE (GSMBE) with ammonia on bulk (0001) AlN substrate is shown in Figure 1. As seen in Figure 1a, there are two types of reflections indicated separately by black solid and white dashed arrows. It was shown that this complex RHEED pattern can be attributable to the presence of Al O surface islands [53]. The well-defined (00), (01), 2 3 and (–01) reflections indicated by solid arrows can be attributed to the (1  1) AlN (0001) surface reconstruction at low temperatures. The weak additional reflections, indicated by dashed arrows, arise from formation of crystalline Al O islands on the AlN surface. These islands cannot be removed by 2 3 baking of the AlN substrate at high temperatures, up to ~1100 C [53]. However, nitridation of such a surface, by exposing it to the flux of ammonia for a few minutes at a substrate temperature of ~800 C, yields formation of a pure (1  1) surface structure on AlN (0001), as shown in Figure 1b. This (1  1) surface reconstruction was stable up to 900 C. At this temperature, AlN, Al Ga N, and SPSL of 0.6 0.4 AlN (3 ML)/Al Ga N (3 ML), with a total of 100 pairs, were successfully grown. The entire SPSL 0.08 0.92 was grown in the 2D mode, and formation of a (2  2) surface reconstruction is shown in Figure 1c. The surface was very flat, with the root mean square (rms) roughness of less than 1 nm, as measured by 1  1 m scans, using atomic force microscopy. Figure 2 shows the evolution of the RHEED patterns illustrating 2D!3D!2D growth mode transitions during ammonia GSMBE of Al Ga N (barrier)/Al Ga N (well) structure on 0.55 0.45 0.45 0.55 Al Ga N/AlN buffer grown on (0001) sapphire substrate [54,55]. 0.55 0.45 Appl. Sci. 2018, 8, x FOR PEER REVIEW 3 of 17 (a) (b) (c) Figure 1. Evolution of reflection high energy electron diffraction (RHEED) patterns for different stages at onset of gas source molecular beam epitaxy (GSMBE). (a) The surface of (0001) AlN substrate at low temperatures. The (00), (01), and (–01) reflections from the (1 × 1) AlN surface and reflections from crystalline Al2O3 islands are indicated by arrows; (b) (1 × 1) surface reconstruction of AlN exposed to ammonia; (c) (2 × 2) surface structure after deposition of about 20 pairs of short period superlattices (SPSLs). All the barrier layers were grown in 2D growth mode with 1 × 1 surface reconstruction, as shown Appl. Sci. 2018, 8, 2362 3 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 3 of 17 in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 °C. With the ammonia flux reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increa se was (a) (b) (c) attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest CL intensity and longest PL lifetime for these structures are due to formation of quantum well Figure Figure 1. 1. E Evolution volution of of refle reflection ction high energy electron high energy electron diffract diffraction ion (RHEED) (RHEED) pa patterns tterns for for diff differ erent stages ent stages (QW)/QD regions in AlxGa1−xN/AlyGa1−yN (0.3 < x < 0.45, 0.53 < y ≤ 1) QW structures [57,58]. The at at onset onset of g of gas as sou sour rce ce m molecular olecular beam beam epita epitaxy xy (G (GSMBE). SMBE). ( (a a) The surface ) The surface of of (0001) (0001) AlN AlN sub substrate at strate at approach described in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and low low temperatures. The (00 temperatures. The (00), ), (0(01), 1), and (–01) and (–01) refrleflection ections from s from the (1 the × 1) (1  AlN 1) AlN surface and reflections from surface and reflections successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm fr cry om stalline crystalline Al2O3 is Al laO ndsislands are indiar cae teindicated d by arrows by; ( arr b) (1 ows; × 1) (b surface reconst ) (1  1) surface ruction of reconstr AlN uction exposed of AlN to 2 3 from PAMBE-grown 2 ML thick GaN QDs was also observed [60]. exposed ammonia; ( to c ammonia; ) (2 × 2) surface (c) (2 structure after deposition  2) surface structure after ofdeposition about 20 pairs of of about short period su 20 pairs of short perlattice period s superlattices (SPSLs). (SPSLs). All the barrier layers were grown in 2D growth mode with 1 × 1 surface reconstruction, as shown in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 °C. With the ammonia flux reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth (a) (b) (c) conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm Ga)/NH Ga)/NH3 flux ratios: ( flux ratios: a (a ) 2D-grown barrier at 20 sccm of ammonia; ( ) 2D-grown barrier at 20 sccm of ammonia; (b b) 3D- ) 3D-gr grown well at own well at 5.5 sccm o 5.5 sccm off and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increase was All the barrier layers were grown in 2D growth mode with 1 1 surface reconstruction, as shown Analyzing the state-of-the art results mentioned above and discussed in literature within last attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same two to three years, we can conclude that one of the main current trends aimed at improving internal CL intensity and longest PL lifetime for these structures are due to formation of quantum well ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a low (QW)/QD regions in AlxGa1−xN/AlyGa1−yN (0.3 < x < 0.45, 0.53 < y ≤ 1) QW structures [57,58]. The greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 C. With the ammonia flux defective and highly efficient active layer in such structures. Future research should focus on finding approach described in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior the optimal ratio and distribution of QDs in the active layer, as well as on the development of the successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth growth of LED structures on relatively inexpensive templates or bulk AlN substrates. Of course, it is from PAMBE-grown 2 ML thick GaN QDs was also observed [60]. conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments much more effective to design and develop such an active layer using MBE, which provides in situ are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increase was attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest CL intensity and longest PL lifetime for these structures are due to formation of quantum well (QW)/QD regions in Al Ga N/Al Ga N (0.3 < x < 0.45, 0.53 < y  1) QW structur es [57,58]. The approach described x 1x y 1y (a) (b) (c) in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm from PAMBE-grown Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + 2 ML thick GaN QDs was also observed [60]. Ga)/NH3 flux ratios: (a) 2D-grown barrier at 20 sccm of ammonia; (b) 3D-grown well at 5.5 sccm of Analyzing the state-of-the art results mentioned above and discussed in literature within last ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. two to three years, we can conclude that one of the main current trends aimed at improving internal quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a Analyzing the state-of-the art results mentioned above and discussed in literature within last low two to three defective yea and rs, we ca highly n concl efficient ude that one of active layer the ma in suchin str current trends ai uctures. Futuremed resear at i ch mshould proving interna focus on l finding the optimal ratio and distribution of QDs in the active layer, as well as on the development of quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a low the defect growth ive and of high LEDly stef ructur ficien es t act onir ve elatively layer in inexpensive such structures. templates Future or rese bulk arch AlN sho substrates. uld focus on Of fcourse, inding it is much more effective to design and develop such an active layer using MBE, which provides in the optimal ratio and distribution of QDs in the active layer, as well as on the development of the growth of LED structures on relatively inexpensive templates or bulk AlN substrates. Of course, it is much more effective to design and develop such an active layer using MBE, which provides in situ Appl. Sci. 2018, 8, 2362 4 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 4 of 17 situ monitoring of the growth process. Despite the fact that the MOCVD process dominates in the monitoring of the growth process. Despite the fact that the MOCVD process dominates in the industrial growth of such structures, the results obtained using the MPE should make it possible to industrial growth of such structures, the results obtained using the MPE should make it possible to identify the most important structural and morphological factors influencing the radiative efficiency of identify the most important structural and morphological factors influencing the radiative efficiency the active layer of the LED. These new concepts can then be transferred to the MOCVD processes. of the active layer of the LED. These new concepts can then be transferred to the MOCVD processes. High resolution X-ray diffraction (HR-XRD) of AlN/AlGaN, AlN/GaN, and AlGaN/InGaN were High resolution X-ray diffraction (HR-XRD) of AlN/AlGaN, AlN/GaN, and AlGaN/InGaN were carried out by many researchers [61–65] in order to estimate strain and dislocation density. carried out by many researchers [61–65] in order to estimate strain and dislocation density. Note, HR-XRD measurements, in conjunction with Raman measurements, allow for estimation Note, HR-XRD measurements, in conjunction with Raman measurements, allow for estimation of the residual strain in SPSLs more precisely [66–69]. Usually, HR-XRD studies are carried out of the residual strain in SPSLs more precisely [66–69]. Usually, HR-XRD studies are carried out using using a high-resolution diffractometer in double- and triple-axis alignment. A long range 2-! scan a high-resolution diffractometer in double- and triple-axis alignment. A long range 2θ-ω scan of the of the (0002) reflection is shown in Figure 3 for a typical AlN/Al Ga N SPSL grown on (0001) (0002) reflection is shown in Figure 3 for a typical AlN/AlxGa x 1-x 1N x SPSL grown on (0001) Al Ga N/AlN/sapphire template. Individual peaks corresponding to the AlN and Al Ga N Al0.4 0.4Ga0.6 0.6 N/AlN/sapphire template. Individual peaks corresponding to the AlN and Al0.4Ga0.6 0.4 N buffe 0.6 r buffer layers, and the 0th, 1, and 2 satellites of the SPSL are well defined in Figure 3. The average layers, and the 0th, ±1, and ±2 satellites of the SPSL are well defined in Figure 3. The average SPSL SPSL composition, 0.68, was determined from the 2 position of the 0th peak [61]. From the position of composition, 0.68, was determined from the 2θ position of the 0th peak [61]. From the position of the the 0th and 1 satellite peaks, the average period of the SPSL was determined to be 2.236 nm. Using 0th and ±1 satellite peaks, the average period of the SPSL was determined to be 2.236 nm. Using the the experimentally determined period, and assuming well composition of Al Ga N and pure AlN experimentally determined period, and assuming well composition of Al0.08 0.08Ga0. 0.92 92N and pure AlN barriers, the well and the barrier thicknesses are found to be 0.808 nm and 1.428 nm, respectively [61]. barriers, the well and the barrier thicknesses are found to be 0.808 nm and 1.428 nm, respectively [61]. Simulations based on the experimentally determined SPSL parameters yield an excellent fit to the Simulations based on the experimentally determined SPSL parameters yield an excellent fit to the experimental data, as shown in Figure 3. The deviation of the well and barrier thicknesses from their experimental data, as shown in Figure 3. The deviation of the well and barrier thicknesses from their integer lattice parameters (integer ML multiples) can be attributed to many factors, including residual integer lattice parameters (integer ML multiples) can be attributed to many factors, including residual strain in the SPSL, formation of interfacial layers, interface roughness, composition fluctuations in the strain in the SPSL, formation of interfacial layers, interface roughness, composition fluctuations in well and barrier, stacking faults (SFs), and inversion domain boundaries (IDBs). The detailed analysis the well and barrier, stacking faults (SFs), and inversion domain boundaries (IDBs). The detailed of significance of all these factors was conducted in reference [61]. analysis of significance of all these factors was conducted in reference [61]. Figure 3. A long range 2-! scan of (0002) reflection for AlN/Al Ga N (0.07 < x < 0.09) obtained 1x Figure 3. A long range 2θ-ω scan of (0002) reflection for AlN/AlxGa1−xN (0.07 < x < 0.09) obtained using using a hybrid X-ray mirror. Black line—data [61], red line—simulations (courtesy of Dr. A. Chandolu). a hybrid X-ray mirror. Black line—data [61], red line—simulations (courtesy of Dr. A. Chandolu). Crystalline microstructure of any semiconductor is a very important factor influencing the Crystalline microstructure of any semiconductor is a very important factor influencing the performance of all light emitters. Cross-sectional structure of SPSLs should be investigated by performance of all light emitters. Cross-sectional structure of SPSLs should be investigated by transmission electron microscopy (TEM), in order to get a nanoscale resolution. Two TEM cross-sections transmission electron microscopy (TEM), in order to get a nanoscale resolution. Two TEM cross- of AlN/AlGaN SPSL, grown by GSMBE, are shown in Figure 4. Although the growth conditions sections of AlN/AlGaN SPSL, grown by GSMBE, are shown in Figure 4. Although the growth (substrate temperature, flux ratio, growth time) were the same, the crystalline quality of these SPSLs conditions (substrate temperature, flux ratio, growth time) were the same, the crystalline quality of were very different. It is clearly seen that SPSL grown directly on sapphire substrate contains a very these SPSLs were very different. It is clearly seen that SPSL grown directly on sapphire substrate high density of inversion domain boundaries (IDBs). These domains start to grow from substrate/layer contains a very high density of inversion domain boundaries (IDBs). These domains start to grow interface, mostly due to incomplete nitridation of sapphire at the onset of epitaxial growth [45,70]. from substrate/layer interface, mostly due to incomplete nitridation of sapphire at the onset of It was shown [71] that IDBs dominate the light emission process in GaN containing these defects. epitaxial growth [45,70]. It was shown [71] that IDBs dominate the light emission process in GaN A similar result was reported for MBE grown AlGaN/GaN SLs [72]. However, SPSLs with high density containing these defects. A similar result was reported for MBE grown AlGaN/GaN SLs [72]. of IDBs have inferior electrical properties [45] and cannot be used in the preparation of light-emitting However, SPSLs with high density of IDBs have inferior electrical properties [45] and cannot be used devices. The most detailed impact of sapphire nitridation on the formation of inverse domains in AlN in the preparation of light-emitting devices. The most detailed impact of sapphire nitridation on the layers grown by MOCVD was discussed in a recent paper [70]. formation of inverse domains in AlN layers grown by MOCVD was discussed in a recent paper [70]. Appl. Sci. 2018, 8, 2362 5 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 17 (a) (b) Figure Figure 4. 4. TEM TEM cross- cross-s sect ection ion of of AlN/A AlN/AlGaN lGaN S SPSLs PSLs (cou (courtesy rtesy of Dr. S. of Dr. S. N. N. G G. . C Chu) hu) ((a a) grow ) grown n direct directly ly on on bar bare (0001) sapphire; ( e (0001) sapphire; (bb ) ) grown on ~50 nm thick AlN buf grown on ~50 nm thick AlN bufferflayer er layer. The white scale bars are 20 nm . The white scale bars are 20 nm long. The long. The [0001 [0001] direction ] direis ction shown is sh by own by black a black arrows.rrows. 3. Bandgap of AlN/AlGa(In)N SPSLs 3. Bandgap of AlN/AlGa(In)N SPSLs The bandgap structure of III-Ns SPSLs convenient for the deep UV light emitters can be simulated The bandgap structure of III-Ns SPSLs convenient for the deep UV light emitters can be using different software [73–75]. The BESST (Bandgap Engineering Superlattice Simulation Tool) simulated using different software [73–75]. The BESST (Bandgap Engineering Superlattice Simulation commercially available package from STR Inc. [76] was used to analyze the results of different Tool) commercially available package from STR Inc. [76] was used to analyze the results of different teams [77,78], as well as most of our experimental results. This software is suitable for modelling teams [77,78], as well as most of our experimental results. This software is suitable for modelling optoelectronic devices utilizing SPSLs as essential units of their heterostructure designs. The BESST optoelectronic devices utilizing SPSLs as essential units of their heterostructure designs. The BESST calculates the SPSL electron and hole minibands using a tight-binding approach and numerical solution calculates the SPSL electron and hole minibands using a tight-binding approach and numerical of the Schrödinger equation with account of complex valence band structure of III-nitride compounds. solution of the Schrödinger equation with account of complex valence band structure of III-nitride Coupled solution of the Poisson equation for electric potential, accounting for polarization charges compounds. Coupled solution of the Poisson equation for electric potential, accounting for at the heterostructure interfaces, and discrete drift-diffusion transport equations, allows building up polarization charges at the heterostructure interfaces, and discrete drift-diffusion transport the band diagram of a device at an arbitrary bias and calculating the corresponding electron and equations, allows building up the band diagram of a device at an arbitrary bias and calculating the hole currents, as well as the radiative recombination rate and emission spectrum. Field-dependent corresponding electron and hole currents, as well as the radiative recombination rate and emission mobilities of electrons and holes used in the transport equations are found, self-consistently, with the spectrum. Field-dependent mobilities of electrons and holes used in the transport equations are simulated minibands of SPSLs. found, self-consistently, with the simulated minibands of SPSLs. Fourier-transform infrared optical reflectance (FTIR) [79–82], photoluminescence (PL) [83–85], Fourier-transform infrared optical reflectance (FTIR) [79–82], photoluminescence (PL) [83–85], and cathodoluminescence (CL) [41,42,86] are widely used to estimate the effective bandgaps of and cathodoluminescence (CL) [41,42,86] are widely used to estimate the effective bandgaps of AlN/Ga(Al,In)N SPSLs. At room temperature, these methods are mostly qualitative, although still AlN/Ga(Al,In)N SPSLs. At room temperature, these methods are mostly qualitative, although still very useful for express control and adjustment of light-emitter properties during fabrication. The room very useful for express control and adjustment of light-emitter properties during fabrication. The temperature FTIR and CL were successfully used to facilitate fabrication of the first deep UV LEDs room temperature FTIR and CL were successfully used to facilitate fabrication of the first deep UV operating at 280 nm using undoped and n- and p-type AlN/Al Ga N SPSLs [41,42]. x 1x LEDs operating at 280 nm using undoped and n- and p-type AlN/AlxGa1−xN SPSLs [41,42]. As an example, the experimental FTIR and CL effective bandgaps of AlN/Al Ga N SPSLs 0.08 0.92 As an example, the experimental FTIR and CL effective bandgaps of AlN/Al0.08Ga0.92N SPSLs and and simulations of these structures obtained using two different approaches [46,76] are shown in simulations of these structures obtained using two different approaches [46,76] are shown in Figure Figure 5. One can see that both simulations provide the slope of the optical energy gap dependence 5. One can see that both simulations provide the slope of the optical energy gap dependence on the on the SPSL period (actually, on the AlN barrier “effective” width), similar to that obtained by CL, SPSL period (actually, on the AlN barrier “effective” width), similar to that obtained by CL, whereas whereas FTIR data demonstrate a different slope. This may have originated from the fact that FTIR FTIR data demonstrate a different slope. This may have originated from the fact that FTIR measures measures the spectral dependence of light absorption, which may involve higher electron minibands the spectral dependence of light absorption, which may involve higher electron minibands and lower and lower hole minibands or levels, whereas luminescence occurs mainly from the ground state hole minibands or levels, whereas luminescence occurs mainly from the ground state minibands, due minibands, due to their dominant occupation. Simulations by BESST show that SPSLs, regarded here, to their dominant occupation. Simulations by BESST show that SPSLs, regarded here, possess two possess two different electron minibands and up to three heavy- and light-hole minibands, which can different electron minibands and up to three heavy- and light-hole minibands, which can be be considered as single energy levels at large SPSL periods because of miniband narrowing. Hence, considered as single energy levels at large SPSL periods because of miniband narrowing. Hence, contribution of the extra minibands to the light absorption may explain the difference in the optical contribution of the extra minibands to the light absorption may explain the difference in the optical energy gap determination by FTIR and CL. energy gap determination by FTIR and CL. Of course, the Stokes shift, which is related to the excited state configuration in the well material, Of course, the Stokes shift, which is related to the excited state configuration in the well material, is a factor in all radiative recombination processes and, therefore, should also yield to red shift of is a factor in all radiative recombination processes and, therefore, should also yield to red shift of the the CL’s estimated bandgap. Note that scatter in the experimental data shown in Figure 5 can be CL’s estimated bandgap. Note that scatter in the experimental data shown in Figure 5 can be attributed to monolayer level uncertainty in the well and barrier thickness across the wafer, as well to attributed to monolayer level uncertainty in the well and barrier thickness across the wafer, as well local composition fluctuations in the well alloy [61,65]. to local composition fluctuations in the well alloy [61,65]. Appl. Sci. 2018, 8, 2362 6 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 17 Figure 5. Experimental optical reflectance and cathodoluminescence (CL)-obtained effective bandgaps Figure 5. Experimental optical reflectance and cathodoluminescence (CL)-obtained effective of AlN/Al Ga N SPSLs vs its period, grown with two nominal well thicknesses (2 and 3 0.08 0.98 bandgaps of AlN/Al0.08Ga0.98N SPSLs vs its period, grown with two nominal well thicknesses (2 and 3 monolayers (MLs), blue and green symbols, respectively) [45]. Theoretical simulations based on monolayers (MLs), blue and green symbols, respectively) [45]. Theoretical simulations based on two two approaches [46,76] are shown by continuous dashed and solid curves, respectively. approaches [46,76] are shown by continuous dashed and solid curves, respectively. 4. AlN/AlGa(In)N SPSL Doping and Ohmic Contacts 4. AlN/AlGa(In)N SPSL Doping and Ohmic Contacts The efficiency of deep UV LEDs is very sensitive to the doping level of p- and n-type emitters. The efficiency of deep UV LEDs is very sensitive to the doping level of p- and n-type emitters. There are no significant issues with n-type doping of III-Ns compounds and their alloys, since Si, There are no significant issues with n-type doping of III-Ns compounds and their alloys, since Si, a a common n-type dopant, behaves as a shallow donor, even in wide bandgap AlGaN alloys [87]. common n-type dopant, behaves as a shallow donor, even in wide bandgap AlGaN alloys [87]. Although the activation energy of Si significantly increases in AlN [88–90], the resistivity of Si-doped Although the activation energy of Si significantly increases in AlN [88–90], the resistivity of Si-doped AlN/Al Ga (In)N (0.05 < x < 0.1) SPSLs is very low, 0.015–0.040 Wcm, and the electron concentration x 1x 19 3 AlN/AlxGa1−x(In)N (0.05 < x < 0.1) SPSLs is very low, 0.015–0.040 Ω·cm, and the electron concentration exceeds 10 cm [42,91]. Such n-type SPSLs can also be used as contact layers. Using a Ti/Al/Ti/Au 19 −3  5 2 exceeds 10 cm [42,91]. Such n-type SPSLs can also be used as contact layers. Using a Ti/Al/Ti/Au stack annealed at 700 C, specific contact resistance of the order of 10 Wcm was obtained for –5 2 stack annealed at 700 °C, specific contact resistance of the order of 10 Ω·cm was obtained for AlN/AlGa(In)N SPSL with ~5.1 eV bandgap [92]. AlN/AlGa(In)N SPSL with ~5.1 eV bandgap [92]. Unfortunately, for III-Ns compounds and their alloys, there is only one convenient p-type dopant, Unfortunately, for III-Ns compounds and their alloys, there is only one convenient p-type Mg, the experimentally determined activation energy of which varies from ~120 to ~220 meV in dopant, Mg, the experimentally determined activation energy of which varies from ~120 to ~220 meV GaN [93,94], and reaches more than 500 meV in AlN [94,95]. A detailed analysis of the Mg activation in GaN [93,94], and reaches more than 500 meV in AlN [94,95]. A detailed analysis of the Mg energy in p-AlGaN epitaxial layers over the entire composition range was recently published [96]. activation energy in p-AlGaN epitaxial layers over the entire composition range was recently It should be noted that if Mg-doped AlGa(In)N is grown using MOCVD, then Mg activation is required. published [96]. It should be noted that if Mg-doped AlGa(In)N is grown using MOCVD, then Mg This can be done by both rapid thermal annealing at elevated temperatures [97,98] and holding the activation is required. This can be done by both rapid thermal annealing at elevated temperatures sample under an electron beam irradiation [98–101]. It was also shown that Mg–O co-doping reduces [97,98] and holding the sample under an electron beam irradiation [98–101]. It was also shown that acceptor activation energy in GaN [102,103], as in AlN [104]. However, this method of doping is not Mg–O co-doping reduces acceptor activation energy in GaN [102,103], as in AlN [104]. However, this widely used since oxygen can react with aluminum and gallium, forming undesirable oxides of these method of doping is not widely used since oxygen can react with aluminum and gallium, forming metals, especially when used in molecular beam epitaxy. undesirable oxides of these metals, especially when used in molecular beam epitaxy. The hole density at room temperature in Al Ga N:Mg alloys with Mg concentration at x 1x 19 20 3 19 The hole density at room temperature in AlxGa1−xN:Mg alloys with Mg concentration at ~10 – ~10 –10 cm is shown in Figure 6a. Note that all Mg concentrations were obtained using secondary 20 –3 10 cm is shown in Figure 6a. Note that all Mg concentrations were obtained using secondary ion ion mass spectrometry (SIMS) [25,103]. mass spectrometry (SIMS) [25,103]. A significant decrease in the hole concentration with an increase in the Al content in A significant decrease in the hole concentration with an increase in the Al content in AlGaN is AlGaN is consistent with an increase in the activation energy of Mg in the wide bandgap layers. consistent with an increase in the activation energy of Mg in the wide bandgap layers. For For AlN/Al Ga (In)N (0.03 < x < 0.08) SPSLs with 2–3 ML thick wells, and periods of the 1x AlN/AlxGa1−x(In)N (0.03 < x < 0.08) SPSLs with 2–3 ML thick wells, and periods of the order of 6–8 order of 6–8 MLs, the average AlN concentration in SPSLs can be changed in the range of y ave 19 3 MLs, the average AlN concentration in SPSLs can be changed in the range of yave = 0.5–0.8. The = 0.5–0.8. The average Mg concentration in these SPSLs is usually at the level of 10 cm . 19 −3 18 3 average Mg concentration in these SPSLs is usually at the level of 10 cm . The concentration of holes The concentration of holes can be at the level of 10 cm , even in SPSLs with high average AlN 18 −3 can be at the level of 10 cm , even in SPSLs with high average AlN content, as is seen in Figure 6a. content, as is seen in Figure 6a. Such structures were obtained using both MBE and MOCVD Such structures were obtained using both MBE and MOCVD methods [25,41,42,45,91,105–110]. methods [25,41,42,45,91,105–110]. Figure 6b shows the results of temperature-dependent Hall characterization of three SPSLs and Figure 6b shows the results of temperature-dependent Hall characterization of three SPSLs and 19 −3 19 −3 19 3 19 3 one AlGaN layer. The Mg concentration was ~10 cm in SPSLs, and ~3 × 10 cm in AlGaN. Note one AlGaN layer. The Mg concentration was ~10 cm in SPSLs, and ~3  10 cm in AlGaN. that the composition of the well was in the range 0.03 < x < 0.08 which is very similar to the Al concentration in the AlGaN thick layer. Average AlN concentrations of these SPSLs were yave = 0.65, Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 17 0.55, and 0.45. The average SPSL compositions, yave, were obtained using X-ray diffraction [61]. The hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in Appl. Sci. 2018, 8, 2362 7 of 17 SPSLs, and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 17 Mg-doped Al0.05Ga0.95N layer. Similar experimental results were reported by many different teams Note that the composition of the well was in the range 0.03 < x < 0.08 which is very similar to the 0.55, and 0.45. The average SPSL compositions, yave, were obtained using X-ray diffraction [61]. The [45,105–115]. Al concentration in the AlGaN thick layer. Average AlN concentrations of these SPSLs were y = ave hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. 0.65, 0.55, and 0.45. The average SPSL compositions, y , were obtained using X-ray diffraction [61]. ave Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to The hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to SPSLs, and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in SPSLs, Mg-doped Al0.05Ga0.95N layer. Similar experimental results were reported by many different teams and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly Mg-doped [45,105–115]. Al Ga N layer. Similar experimental results were reported by many different teams [45,105–115]. 0.05 0.95 (a) (b) Figure 6. Hole concentration in AlxGa1-xN alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on average Al concentration in epitaxial layers (black symbols) and in SPSLs (red symbols and green box); (b) The temperature dependence of the hole concentration in Al0.05Ga0.95N epitaxial layer (~0.5 μm thick, filled circles) and in AlN/AlxGa1−xN SPSLs (open symbols). Lines 1, 2, and 3 are the results of the applied BESST simulator [76]. (a) (b) The valence band edge profile in AlN(~0.90 nm)/Al0.03Ga0.97N(~0.52 nm) SPSL, computed by the Figure 6. Hole concentration in AlxGa1-xN alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on Figure 6. Hole concentration in Al Ga N alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on 1x BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in averag average e Al Al con concentration centration in in epitaxial epitaxial lay layers ers (black (blacsymbols) k symbols) and and in SPSLs in SPSLs (red s (red symbols ymbol and s gr and een gree box); n this SPSL corresponds to the zero-energy horizontal line shown in Figure 7a. The acceptor levels in box); ( (b) The b) The t temperatur emperature dependence of th e dependence of the holee hole con concentration centration in Al in Al Ga 0.05Ga N0.95 epitaxial N epitax layer ial lay (~0.5 er (~0.5 m 0.05 0.95 Al0.03Ga0.97N-well and AlN-barrier layers are indicated by the dashed lines. μ thick, m thick filled , filcir led cles) circand les) and in AlN/Al in AlN/A Ga lxGa1 N−xN SPSLs SPSL(open s (open symbol symbols).sLines ). Lines 1, 1, 2, 2, and and 3 are the res 3 are the results ulof ts x 1x It is clearly seen that the acceptor levels in the quantum wells are above the Fermi level. Thus, the applied BESST simulator [76]. of the applied BESST simulator [76]. these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor The valence band edge profile in AlN(~0.90 nm)/Al Ga N(~0.52 nm) SPSL, computed by the 0.03 0.97 energy leve The valenc ls, yi e b eld aind ed ng a part ge pro ial act file in iv AlN(~0.90 nm)/A ation of acceptors in these l0.03Ga0.97N(~0 lay .5e2 rs. The d nm) SPSL, computed by the egree of activation BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in depends on temperature, creating hole tunneling from barriers to wells, which results in the this SPSL corresponds to the zero-energy horizontal line shown in Figure 7a. The acceptor levels in exponential dependences o this SPSL corresponds to tf h hole concent e zero-energy horizont ration, as sho al lw ine shown n in Figure in Fig 6. ure 7a. The acceptor levels in Al Ga N-well and AlN-barrier layers are indicated by the dashed lines. Al0.03 0.03 Ga0.97 0.97 N-well and AlN-barrier layers are indicated by the dashed lines. It is clearly seen that the acceptor levels in the quantum wells are above the Fermi level. Thus, these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor energy levels, yielding a partial activation of acceptors in these layers. The degree of activation depends on temperature, creating hole tunneling from barriers to wells, which results in the exponential dependences of hole concentration, as shown in Figure 6. (a) (b) Figure 7. Valence band layout in AlN/Al0.03Ga0.97N SPSL, computed for room temperature (a) and Figure 7. Valence band layout in AlN/Al Ga N SPSL, computed for room temperature (a) and 0.03 0.97 band d band diagram iagram and heavy and heavy (H (HH), H), lig light ht (LH) (LH), , and sp and sp lit-off lit-of ( f S (SH) H) hole hole concen concentrations trations in thi in this s sam sample ple (b (). b ). The acceptor level posit The acceptor level positions ions were wereest estimated imated by negle by neglecting cting the effe the effect ct of ofthe hole the hole confin confinement ement in the in the qu quantum antum wells wells. . It is clearly seen that the (a) ( acceptor levels in the quantum wells arebabove ) the Fermi level. Thus, these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor energy Figure 7. Valence band layout in AlN/Al0.03Ga0.97N SPSL, computed for room temperature (a) and levels, yielding a partial activation of acceptors in these layers. The degree of activation depends band diagram and heavy (HH), light (LH), and split-off (SH) hole concentrations in this sample (b). The acceptor level positions were estimated by neglecting the effect of the hole confinement in the quantum wells. Appl. Sci. 2018, 8, x FOR PEER REVIEW 8 of 17 Figure 7b shows the distributions of the heavy, light, and split-off holes in the same SPSL obtained by self-consistent solution of the Poisson and Schrödinger equations [76], accounting for the complex valence band structure of III-nitrides [116]. It is clearly seen that the total hole density in such an SPSL mostly depends on the heavy hole concentration. The contribution of the split-off holes can be neglected, due to the fact that this subband in the SPSL is far below the heavy and light subbands. Note, the holes in the SPSL are accumulated in the wells, partially penetrating into the AlN Appl. Sci. 2018, 8, 2362 8 of 17 barriers. One could conclude that the efficiency of Mg activation is a very weak function of temperature in wide bandgap AlN/AlxGa1−x(In)N SPSL (Eg from ~4.2 to ~5.3 eV) [107,108]. Thus, these SPSLs are on temperature, creating hole tunneling from barriers to wells, which results in the exponential very attractive for fabrication of transparent low resistive ohmic contacts for deep UV light emitters. dependences of hole concentration, as shown in Figure 6. Various contact metallization schemes have been applied for Mg-doped p-GaN, including Au, Figure 7b shows the distributions of the heavy, light, and split-off holes in the same SPSL obtained Ni, Ti, Pd, Pt, Au/Ni, Au/Pt, Au/Cr, Au/Pd, Au/Mg/Au, Au/Pt/Pd, Au/Cr/Ni, Au/Pt/Ni, Au/Ni/Pt, by self-consistent solution of the Poisson and Schrödinger equations [76], accounting for the complex Au-Zn/Ni, Si/Ni/Mg/Ni, etc. [117]. Oxidized Au/Ni is the most common ohmic contact to p-GaN, due valence band structure of III-nitrides [116]. It is clearly seen that the total hole density in such an to the relatively low specific contact resistance (ρc) and simplicity of fabrication [118]. It was shown SPSL mostly depends on the heavy hole concentration. The contribution of the split-off holes can be that formation of crystalline NiO and Ni3Ga4 play an essential role in the reduction of the specific neglected, due to the fact that this subband in the SPSL is far below the heavy and light subbands. contact resistance of p-GaN [118–121]. While the formation of NiO yields a reduction of the specific Note, the holes in the SPSL are accumulated in the wells, partially penetrating into the AlN barriers. contact resistance, the formation of Ni3Ga4 results in the eventual increase in the contact resistance One could conclude that the efficiency of Mg activation is a very weak function of temperature in −6 −6 2 [119–121]. The lowest specific contact resistance ρc ~9.2 × 10 and 2 × 10 Ω cm at 150 °C were wide bandgap AlN/Al Ga (In)N SPSL (E from ~4.2 to ~5.3 eV) [107,108]. Thus, these SPSLs are x 1x g achieved for p-type Mg-doped GaN and AlxGa1−xN layers [121,122]. very attractive for fabrication of transparent low resistive ohmic contacts for deep UV light emitters. Similar approach was successfully used for fabrication of ohmic contacts on p-type Various contact metallization schemes have been applied for Mg-doped p-GaN, including Au, Ni, AlN/AlxGa1−xN SPSLs [108]. Figure 8 shows the temperature dependence of the specific contact Ti, Pd, Pt, Au/Ni, Au/Pt, Au/Cr, Au/Pd, Au/Mg/Au, Au/Pt/Pd, Au/Cr/Ni, Au/Pt/Ni, Au/Ni/Pt, resistances, ρc, for three SPSLs with average AlN content of yave ~0.7 and for the ~300 nm thick Mg- Au-Zn/Ni, Si/Ni/Mg/Ni, etc. [117]. Oxidized Au/Ni is the most common ohmic contact to p-GaN, 19 −3 doped p-Al0.03Ga0.97N epitaxial layer. The Mg concentration in all samples was ~10 cm and room due to the relatively low specific contact resistance (r ) and simplicity of fabrication [118]. It was −5 2 temperature ρc was ~4 × 10 Ω·cm . shown that formation of crystalline NiO and Ni Ga play an essential role in the reduction of the 3 4 It was shown that the temperature dependence of the specific contact resistance is primarily specific contact resistance of p-GaN [118–121]. While the formation of NiO yields a reduction of the controlled by the activation energy of Mg acceptors in AlxGa1−xN [123]. The ρc of metal/SPSL ohmic specific contact resistance, the formation of Ni Ga results in the eventual increase in the contact 3 4 6 6 2 contact, as seen in Figure 8, also follows the temperature dependence of hole concentration in these resistance [119–121]. The lowest specific contact resistance r ~9.2  10 and 2  10 W cm at structure  s, which are very similar to that shown in Figure 6b, since the density of acceptors activated 150 C were achieved for p-type Mg-doped GaN and Al Ga N layers [121,122]. x 1x in AlN barriers is the primary factor influencing the hole concentration in the well, producing a Similar approach was successfully used for fabrication of ohmic contacts on p-type weaker temperature dependence compared to bulk materials. Thus, a weak dependence of ρc on AlN/Al Ga N SPSLs [108]. Figure 8 shows the temperature dependence of the specific contact x 1x temperature is expected, even when thermionic mechanism of current injection dominates in the resistances, r , for three SPSLs with average AlN content of y ~0.7 and for the ~300 nm thick c ave metal/SPSL ohmic contact. This expectation agrees with the results shown in Figure 8. 19 3 Mg-doped p-Al Ga N epitaxial layer. The Mg concentration in all samples was ~10 cm and 0.03 0.97 5 2 room temperature r was ~4  10 Wcm . Figure 8. Specific contact resistances of Mg-doped AlN/Al Ga N SPSLs (y ~0.7) and Al Ga N x 1x ave 0.03 0.97 Figure 8. Specific contact resistances of Mg-doped AlN/AlxGa1−xN SPSLs (yave ~0.7) and Al0.03Ga0.97N epitaxial layer at different temperatures. epitaxial layer at different temperatures. It was shown that the temperature dependence of the specific contact resistance is primarily 5. Deep UV Light Emitters Based on AlN/AlGa(In)N SPSLs controlled by the activation energy of Mg acceptors in Al Ga N [123]. The r of metal/SPSL ohmic x 1x c contact, as seen in Figure 8, also follows the temperature dependence of hole concentration in these The ability to change the bandgap in SPSL by several hundred meV, by changing the thickness structures, which are very similar to that shown in Figure 6b, since the density of acceptors activated in of the well by 1 ML, opened up a very simple path to the implementation of the DHS LEDs. Since AlN barriers is the primary factor influencing the hole concentration in the well, producing a weaker temperature dependence compared to bulk materials. Thus, a weak dependence of r on temperature is expected, even when thermionic mechanism of current injection dominates in the metal/SPSL ohmic contact. This expectation agrees with the results shown in Figure 8. Appl. Sci. 2018, 8, 2362 9 of 17 5. Deep UV Light Emitters Based on AlN/AlGa(In)N SPSLs Appl. Sci. 2018, 8, x FOR PEER REVIEW 9 of 17 The ability to change the bandgap in SPSL by several hundred meV, by changing the thickness of the well by 1 ML, opened up a very simple path to the implementation of the DHS LEDs. Since SPSL with particular barrier/well thicknesses has its effective bandgap, we can consider them to form SPSL with particular barrier/well thicknesses has its effective bandgap, we can consider them to form a heterostructure in the sense that an SPSL-based active layer with a smaller effective bandgap is a heterostructure in the sense that an SPSL-based active layer with a smaller effective bandgap is sandwiched between n- and p-type SPSL- based cladding layers with a larger bandgap. sandwiched between n- and p-type SPSL- based cladding layers with a larger bandgap. A schematic cross-section of a typical DHS for a sub-300 nm LED is shown in Figure 9. The A schematic cross-section of a typical DHS for a sub-300 nm LED is shown in Figure 9. The structure structure consists of six main parts: (1) thin, <100 nm, AlN buffer; (2) thick, ~200 nm, Al0.7–0.8Ga0.3−0.2N consists of six main parts: (1) thin, <100 nm, AlN buffer; (2) thick, ~200 nm, Al Ga N buffer 0.7–0.8 0.3–0.2 buffer for reduction of dislocation density in a device structure (this layer can be replaced by for reduction of dislocation density in a device structure (this layer can be replaced by AlN/GaN AlN/GaN SPSL—“dislocation absorber”); (3) n-type AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(2 ML) SPSL SPSL—“dislocation absorber”); (3) n-type AlN(5 ML)/Al Ga (In)N(2 ML) SPSL cladding 0.92–0.97 0.08–0.03 cladding layer of ~400 nm; (4) undoped AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(3 ML) SPSL active region layer of ~400 nm; (4) undoped AlN(5 ML)/Al Ga (In)N(3 ML) SPSL active region (5 pairs); 0.92–0.97 0.08–0.03 (5 pairs); (5) p-type AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(2 ML) SPSL cladding layer of ~200 nm; (6) thin, (5) p-type AlN(5 ML)/Al Ga (In)N(2 ML) SPSL cladding layer of ~200 nm; (6) thin, ~2–5 nm, 0.92–0.97 0.08–0.03 ~2–5 nm, p-type contact layer with the composition of SPSL’s well material or pure GaN:Mg. As can p-type contact layer with the composition of SPSL’s well material or pure GaN:Mg. As can be seen be seen from the TEM cross-section of this device, shown in the same Figure 9, the barrier/well from the TEM cross-section of this device, shown in the same Figure 9, the barrier/well interfaces are interfaces are quite sharp, and the thickness of the well in the active region is greater than in the quite sharp, and the thickness of the well in the active region is greater than in the adjacent n- and adjacent n- and p-type barriers. p-type barriers. Figure 9. A not-in-scale schematic cross-section (on the left) of a double heterostructure (DHS) Figure 9. A not-in-scale schematic cross-section (on the left) of a double heterostructure (DHS) light- light-emitting diode (LED) and TEM cross-section (on the right) of the active region, and adjacent n- emitting diode (LED) and TEM cross-section (on the right) of the active region, and adjacent n- and and p-type emitters of this diode. p-type emitters of this diode. The current–voltage (I–V) characteristic of a typical 260 nm DHS mesa-LED is shown in Figure 10a. The current–voltage (I–V) characteristic of a typical 260 nm DHS mesa-LED is shown in Figure The LED turns on at ~6.0 V, and has a relatively high differential series resistance R of 115 5 W under ser 10a. The LED turns on at ~6.0 V, and has a relatively high differential series resistance Rser of 115 ± 5 forward bias from 8 V to 12 V. The R of mesa diodes is the sum of the contact (R ), spreading (R ), ser cont sp Ω under forward bias from 8 V to 12 V. The Rser of mesa diodes is the sum of the contact (Rcont), and vertical (R ) resistances. The contact resistance of a 160 m diameter diode, obtained from ver spreading (Rsp), and vertical (Rver) resistances. The contact resistance of a 160 μm diameter diode, specific contact resistance, is R ~90 W. The estimated spreading resistance of this LED is R ~20 W, cont sp obtained from specific contact resistance, is Rcont ~90 Ω. The estimated spreading resistance of this taking into account the resistivity of n-type AlGaN buffer layer (not shown in the inset of Figure 10a). LED is Rsp ~20 Ω, taking into account the resistivity of n-type AlGaN buffer layer (not shown in the The resistance of the etched part of the mesa, corresponding to transport across the SPSL layers, is inset of Figure 10a). The resistance of the etched part of the mesa, corresponding to transport across R  r  (h/A) where r is the perpendicular (along the growth direction) resistivity of the SPSL, ver 1 1 the SPSL layers, is Rver ≈ ρ1 × (h/A) where ρ1 is the perpendicular (along the growth direction) resistivity h is the height of the mesa (~300 nm, with the thickness of p-SL ~200 nm), and A is the contact area. of the SPSL, h is the height of the mesa (~300 nm, with the thickness of p-SL ~200 nm), and A is the Finally, we obtain R ~5 W, resulting in r ~50 W cm for a p-type SPSL. Comparing this to the in-plane ver 1 contact area. Finally, we obtain Rver ~5 Ω, resulting in ρ1 ~50 Ω cm for a p-type SPSL. Comparing this conductivity (r ) of the p-type SPSL obtained from Hall measurements, r ~4 W cm, we obtain the 2 2 to the in-plane conductivity (ρ2) of the p-type SPSL obtained from Hall measurements, ρ2 ~4 Ω cm, conductivity anisotropy r /r ~13. These simple considerations indicate relatively low r , considering 1 2 1 we obtain the conductivity anisotropy ρ1/ρ2 ~13. These simple considerations indicate relatively low the high AlN fraction in these SPSLs, and underscore the importance of reducing the contact resistance ρ1, considering the high AlN fraction in these SPSLs, and underscore the importance of reducing the to p-type materials. However, it is also important to lower R by optimization of the buffer layer sp contact resistance to p-type materials. However, it is also important to lower Rsp by optimization of thickness and its resistivity. Similar I–V characteristics were reported by different teams for LEDs the buffer layer thickness and its resistivity. Similar I–V characteristics were reported by different emitting near the same deep UV range of the wavelengths [42–46,91–128]. teams for LEDs emitting near the same deep UV range of the wavelengths [42–46,91–128]. The typical electroluminescence spectra of DHS LEDs based on AlN/AlGa(In)N SPSLs are shown The typical electroluminescence spectra of DHS LEDs based on AlN/AlGa(In)N SPSLs are in Figure 10b. All these spectra were obtained on LEDs with a similar geometry, and with the same shown in Figure 10b. All these spectra were obtained on LEDs with a similar geometry, and with the currents. A significant decrease in EQE is evident for devices emitting in ever deeper UV. The external same currents. A significant decrease in EQE is evident for devices emitting in ever deeper UV. The external quantum efficiency (EQE) of such simple LEDs does not exceed 0.1%. The absence of the electron blocking layer is the main drawback [129] of the structure shown in Figure 9. In order to increase the efficiency of such LED structures, it was proposed to insert a few monolayers-thick AlN electron blocking layer between the active layer and the p-emitter [35,36]. Appl. Sci. 2018, 8, 2362 10 of 17 quantum efficiency (EQE) of such simple LEDs does not exceed 0.1%. The absence of the electron blocking layer is the main drawback [129] of the structure shown in Figure 9. In order to increase Appl. Sci. 2018, 8, x FOR PEER REVIEW 10 of 17 the efficiency of such LED structures, it was proposed to insert a few monolayers-thick AlN electron blocking layer between the active layer and the p-emitter [35,36]. Summarizing, it should be noted that several new approaches to the use of SPSLs in combination Summarizing, it should be noted that several new approaches to the use of SPSLs in combination with QDs in the active layer of the LED have been proposed [60,130–132]. The most promising, from with QDs in the active layer of the LED have been proposed [60,130–132]. The most promising, my point of view, is the idea set forth in the ref. [60], where, experimentally, deep UV emission at 219 from my point of view, is the idea set forth in the ref. [60], where, experimentally, deep UV emission at nm from ultrathin MBE-grown GaN/AlN quantum structure was demonstrated. I believe in the 219 nm from ultrathin MBE-grown GaN/AlN quantum structure was demonstrated. I believe in the success of this approach, since we observed a similar effect almost 10 years ago [54,55,57,58]. success of this approach, since we observed a similar effect almost 10 years ago [54,55,57,58]. (a) (b) Figure 10. Figure 10. Electrical and opt Electrical and optical ical char characteristics acteristics of A of lN/A AlN/AlGa(In)N lGa(In)N SPSL-b SPSL-based ased DHS me DHS smesa-LEDs. a-LEDs. (a) Current–voltage characterist (a) Current–voltage characteristic ic of a typical LED of a typical LED operating at 262 nm wa operating at 262 nm wavelength. velength. Note Note that light that light em emission ission wa wass observed observed w with ith forward forward dc curr dc current above 2 mA; ( ent above 2 mA; (b) Electr b) oluminescence Electroluminescence spectra of spectra of similar similar AlN/AlGa(In)N AlN/AlGa(In)N SPSL- SPSL-based DHS based D mesa-LEDs HS mesa-LED operating s operat at 255, ing at 255, 280, 280, and 290 and 290 nm. nm. 6. Conclusions 6. Conclusions It was demonstrated that AlN/Al Ga N p- and n-type SPSLs with average AlN content up to x 1x It was demonstrated that AlN/AlxGa1−xN p- and n-type SPSLs with average AlN content up to 18 3 y ~0.7 and bandgap over 5.1 eV could have hole and electron concentrations exceeding 10 cm . ave 18 −3 yave ~0.7 and bandgap over 5.1 eV could have hole and electron concentrations exceeding 10 cm . 5 2 Low-resistance ohmic contacts with specific contact resistance below ~4  10 Wcm can be formed −5 2 Low-resistance ohmic contacts with specific contact resistance below ~4 × 10 Ω·cm can be formed on these SPSLs. The LEDs based on SPSLs with emission wavelengths from 290 to 232 nm were on these SPSLs. The LEDs based on SPSLs with emission wavelengths from 290 to 232 nm were demonstrated by different teams. demonstrated by different teams. Funding: This paper received no external funding. Funding: This paper received no external funding. Acknowledgments: I would like to acknowledge all my colleagues involved in this research. I thank S. Yu. Acknowledgments: I would like to acknowledge all my colleagues involved in this research. I thank S. Yu. Karpov for very helpful discussions. Karpov for very helpful discussions. Conflicts of Interest: The author declares no conflict of interest. Conflicts of Interest: The author declares no conflict of interest. References References 1. Alferov, Z.I. Nobel Lecture: The double heterostructure concept and its applications in physics, electronics, 1. Alferov, Z.I. Nobel Lecture: The double heterostructure concept and its applications in physics, electronics, and technology. Rev. Mod. Phys. 2001, 73, 767–782. [CrossRef] and technology. Rev. Mod. Phys. 2001, 73, 767–782, doi:10.1103/RevModPhys.73.767. 2. Kroemer, H. Nobel Lecture: Quasielectric fields and band offsets: Teaching electrons new tricks. 2. Kroemer, H. Nobel Lecture: Quasielectric fields and band offsets: Teaching electrons new tricks. Rev. Mod. Rev. Mod. Phys. 2001, 73, 783–793. [CrossRef] Phys. 2001, 73, 783–793, doi:10.1103/RevModPhys.73.883. 3. Esaki, L.; Tsu, R. Superlattice and negative differential conductivity in semiconductors. IBM J. Res. Dev. 1970, 3. Esaki, L.; Tsu, R. Superlattice and negative differential conductivity in semiconductors. IBM J. Res. Dev. 14, 61–65. [CrossRef] 1970, 14, 61–65, doi:10.1147/rd.141.0061. 4. Casey, H.C., Jr.; Panish, M.B. Heterostructure Lasers; Academic Press: New York, NY, USA, 1978; pp. 156–300. 4. Casey, H.C., Jr.; Panish, M.B. Heterostructure Lasers; Academic Press: New York, NY, USA, 1978; pp. 156– ISBN 0-12-163102-8. 300, ISBN 0-12-163102-8. 5. Ivanov, S.V.; Kopev, P.S.; Ledentson, N.N.; Meltser, B.Y.; Nadtochy, M.Y.; Ustinov, V.M.; Vasilev, A.M. 5. Ivanov, S.V.; Kopev, P.S.; Ledentson, N.N.; Meltser, B.Y.; Nadtochy, M.Y.; Ustinov, V.M.; Vasilev, A.M. Ultrathin layers and short-period superlattices in semiconductor structures. Phys. Stat. Sol. 1990, 118, Ultrathin layers and short-period superlattices in semiconductor structures. Phys. Stat. Sol. 1990, 118, 169– 169–177. [CrossRef] 177, doi:10.1002/pssa.2211180119. 6. Chang, M.-H.; Das, D.; Varde, P.V.; Pecht, M. Light emitting diodes reliability review. Microelectron. Reliab. 2012, 52, 762–782, doi:10.1016/j.microrel.2011.07.063. 7. Opel, D.R.; Hagstrom, E.; Pace, A.K.; Sisto, K.; Hirano-Ali, S.A.; Desai, S.; Swan, J. Light emitting diodes, a brief review and clinical experience. J. Clin. Aesthet. Dermatol. 2015, 8, 36–44, PMC4479368. Appl. Sci. 2018, 8, 2362 11 of 17 6. Chang, M.-H.; Das, D.; Varde, P.V.; Pecht, M. Light emitting diodes reliability review. Microelectron. Reliab. 2012, 52, 762–782. [CrossRef] 7. Opel, D.R.; Hagstrom, E.; Pace, A.K.; Sisto, K.; Hirano-Ali, S.A.; Desai, S.; Swan, J. Light emitting diodes, a brief review and clinical experience. J. Clin. Aesthet. Dermatol. 2015, 8, 36–44, PMCID:PMC4479368. [PubMed] 8. Yadav, A.; Titkov, I.E.; Sokolovskii, G.S.; Karpov, S.Y.; Dudelev, V.V.; Soboleva, K.K.; Strassburg, M.; Pietzonka, I.; Lugauer, H.-J.; Rafailov, E.U. AlGaInP red-emitting light emitting diode under extremely high pulsed pumping. Proc. SPIE 2016, 9768. [CrossRef] 9. Karpov, S.Y. Light-emitting diodes for solid-state lighting: Searching room for improvements. Proc. SPIE 2016, 9768. [CrossRef] 10. Gupta, S.D. (Ed.) Light Emitting Diodes for Agriculture; Springer: Singapore, 2017. [CrossRef] 11. Schubert, E.F. Light-Emitting Diodes, 3rd ed.; Rensselaer Polytechnic Institute Troy: New York, NY, USA, 2018; pp. 17-1–17-15, 19-6–19-16, 21-2–21-8. ISBN 978-0-9863826-6-6. 12. Pantha, B.N.; Dahal, R.; Nakarmi, M.L.; Nepal, N.; Li, J.; Lin, J.Y.; Jiang, H.X.; Paduano, Q.S.; Weyburne, D. Correlation between optoelectronic and structural properties and epilayer thickness of AlN. Appl. Phys. Lett. 2007, 90. [CrossRef] 13. Zhou, C.; Ghods, A.; Saravade, V.G.; Patel, P.V.; Yunghans, K.L.; Ferguson, C.; Feng, Y.; Kucukgok, B.; Lu, N.; Ferguson, I.T. Review—The current and emerging applications of the III-Nitrides. ECS J. Sol. Stat. Sci. Technol. 2017, 6, Q-149–Q-156. [CrossRef] 14. Davydov, V.Y.; Klochikhin, A.A.; Emtsev, V.V.; Kurdyukov, D.A.; Ivanov, S.V.; Vekshin, V.V.; Bechstedt, F.; Furthmüller, J.; Aderhold, J.; Graul, J.; et al. Band gap of hexagonal InN and InGaN alloys. Phys. Stat. Sol. 2002, 234, 787–795. [CrossRef] 15. Wu, J.; Walukiewicz, W.; Shan, W.; Yu, K.M.; Ager, J.W., III; Li, S.X.; Haller, E.E.; Lu, H.; Schaff, W.J. Temperature dependence of the fundamental band gap of InN. J. Appl. Phys. Lett. 2003, 94, 4457–4460. [CrossRef] 16. Andreev, B.A.; Kudryavtsev, K.E.; Yablonskiy, A.N.; Lobanov, D.N.; Bu shuykin, P.A.; Krasilnikova, L.V.; Skorokhodov, E.V.; Yunin, P.A.; Novikov, A.V.; Davydov, V.Y.; Krasilnik, Z.F. Towards the indium nitride laser: Obtaining infrared stimulated emission from planar monocrystalline InN structures. Sci. Rep. 2018, 8. [CrossRef] [PubMed] 17. Akasaki, I. Nobel Lecture: Fascinated journeys into blue light. Rev. Mod. Phys. 2015, 87, 1119–1131. [CrossRef] 18. Amano, H. Nobel Lecture: Growth of GaN on sapphire via low-temperature deposited buffer layer and realization of p-type GaN by Mg doping followed by low-energy electron beam irradiation. Rev. Mod. Phys. 2015, 87, 1133–1138. [CrossRef] 19. Nakamura, S. Nobel Lecture: Background story of the invention of efficient blue InGaN light emitting diodes. Rev. Mod. Phys. 2015, 87, 1139–1151. [CrossRef] 20. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Holtz, M.; Nikishin, S.; Temkin, H. AlGaInN light-emitting diodes grown on Si (111). Appl. Phys. Lett. 2002, 80, 3682–3684. [CrossRef] 21. Sizov, V.S.; Tsatsulnikov, A.F.; Sakharov, A.V.; Lundin, W.V.; Zavarin, E.E.; Cherkashin, N.A.; Hÿtch, M.J.; Nikolaev, A.E.; Mintairov, A.M.; He, Y.; et al. The use of short_period InGaN/GaN superlattices in blue-region light-emitting diodes. Semiconductors 2010, 44, 924–930. [CrossRef] 22. Lähnemann, J.; Ajay, A.; Den Hertog, M.I.; Monroy, E. Near-infrared intersubband photodetection in GaN/AlN nanowires. Nano Lett. 2017, 17, 6954–6960. [CrossRef] [PubMed] 23. Mehta, K.; Liu, Y.-S.; Wang, J.; Jeong, H.; Detchprohm, T.; Park, Y.J.; Alugubelli, S.R.; Wang, S.; Ponce, F.A.; Shen, S.-C.; et al. Lateral current spreading in III-N ultraviolet vertical-cavity surface-emitting lasers using modulation-doped short period superlattices. IEEE J. Quant. Electron. 2018, 54. [CrossRef] 24. Nikishin, S.A.; Faleev, N.N.; Antipov, V.G.; Francoeur, S.; Grave de Peralta, L.; Seryogin, G.A.; Temkin, H.; Prokofyeva, T.I.; Holtz, H.; Chu, S.N.G. High quality GaN grown on Si(111) by gas source molecular beam epitaxy with ammonia. Appl. Phys. Lett. 1999, 75, 2073–2075. [CrossRef] 25. Nikishin, S.; Kipshidze, G.; Kuryatkov, V.; Zubrilov, A.; Choi, K.; Gherasoiu, Iu.; Grave de Peralta, L.; Prokofyeva, T.; Holtz, M.; Asomoza, R.; et al. Gas source molecular beam epitaxy of high quality AlGaN on Si and sapphire. Mat. Res. Soc. Symp. 2001, 639. [CrossRef] Appl. Sci. 2018, 8, 2362 12 of 17 26. Xiao, M.; Zhang, J.; Hao, Y. High-quality Ga-rich AlGaN grown on trapezoidal patterned GaN template using super-short period AlN/GaN superlattices for rapid coalescence. J. Cryst. Growth 2017, 464, 153–158. [CrossRef] 27. Aumer, M.E.; LeBoeuf, S.F.; Bedairb, S.M.; Smith, M.; Lin, J.Y.; Jiang, H.X. Effects of tensile and compressive strain on the luminescence properties of AlInGaN/InGaN quantum well structures. Appl. Phys. Lett. 2000, 77, 821–823. [CrossRef] 28. Son, J.H.; Lee, J.-L. Strain engineering for the solution of efficiency droop in InGaN/GaN light-emitting diodes. Opt. Express 2010, 18, 5466–5471. [CrossRef] [PubMed] 29. Nakamura, S.; Fasol, G. The Blue Laser Diode; Springer: Berglin/Heidelberg, Germany; New York, NY, USA, 1997; pp. 147–300. ISBN 978-3-662-03464-4 or 978-3-662-03462-0. 30. Morkoç, H. Handbook of Nitride Semiconductors and Devices. V.3: GaN-Based Optical and Electronic Devices; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2009; pp. 4–334. ISBN 978-3-527-40839-9. 31. Kneissl, M.; Rass, J. III-Nitride Ultraviolet Emitters Technology and Applications; Springer Series in Materials Science; Springer: Berglin/Heidelberg, Germany, 2016; Volume 227, pp. 3–12, 90–104, ISBN 978-3-319-24098-5. 32. Amano, H. Progress and prospect of growth of wide-band-gap group III Nitrides. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 1–9. ISBN 978-981-10-3754-2. 33. Egawa, T.; Oda, O. LEDs based on hetero epitaxial GaN on Si substrates. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 1–9. ISBN 978-981-10-3754-2. 34. Tadatomo, K.; Okagawa, H.; Ohuchi, Y.; Tsunekawa, T.; Kudo, H.; Sudo, Y.; Kato, M.; Taguchi, T. High output power near-ultraviolet and violet light-emitting diodes fabricated on patterned sapphire substrates using metalorganic vapor phase epitaxy. J. Light Vis. Env. 2003, 27, 140–145. [CrossRef] 35. Kneissl, M.; Rass, J.; Schade, L.; Schwarz, U.T. Growth and Optical Properties of GaN-Based Non- and Semipolar LEDs. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 93–128. ISBN 978-981-10-3754-2. 36. Hirayama, H.; Kamata, N.; Tsubaki, K. AlGaN-Based Deep-Ultraviolet Light-Emitting Diodes. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 93–128. ISBN 978-981-10-3754-2. 37. Staszczak, G.; Gorczyca, I.; Grzank, E.; Smalc-Koziorowska, J.; Targowski, G.; Czernecki, R.; Siekacz, M.; Grzanka, S.; Skierbiszewski, C.; Schulz, T.; et al. Bandgap behavior of InGaN/GaN short period superlattices grown by metal-organic vapor phase epitaxy. Phys. Stat. Sol. 2017, 254, 1600710. [CrossRef] 38. Dimitrakopulos, G.P.; Vasileiadis, I.G.; Bazioti, C.; Smalc-Koziorowska, J.; Kret, S.; Dimakis, E.; Florini, N.; Kehagias, T.; Suski, T.; Karakostas, T.; et al. Compositional and strain analysis of In(Ga)N/GaN short period superlattices. J. Appl. Phys. 2018, 123, 024304. [CrossRef] 39. Gorczyca, I.; Suski, T.; Christensen, N.E.; Svane, A. Theoretical study of nitride short period superlattices. J. Phys. Condens. Matter 2018, 30, 063001. [CrossRef] [PubMed] 40. Lundin, W.V.; Nikolaev, A.E.; Sakharov, A.V.; Zavarin, E.E.; Valkovskiy, G.A.; Yagovkina, M.A.; Usov, S.O.; Kryzhanovskaya, N.V.; Sizov, V.S.; Brunkov, P.N.; et al. Single quantum well deep-green LEDs with buried InGaN/GaN short-period superlattice. J. Cryst. Growth 2011, 315, 267–271. [CrossRef] 41. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Nikishin, S.; Holtz, M.; Chu, S.N.G.; Temkin, H. Deep Ultraviolet AlGaInN-Based Light-Emitting Diodes on Si(111) and Sapphire. Phys. Stat. Sol. 2002, 192, 286–291. [CrossRef] 42. Kipshidze, G.; Kuryatkov, V.; Zhu, K.; Borisov, B.; Holtz, M.; Nikishin, S.; Temkin, H. AlN/AlGaInN superlattice light emitting diodes at 280 nm. J. Appl. Phys. 2003, 93, 1363–1366. [CrossRef] 43. Jmerik, V.N.; Lutsenko, E.V.; Ivanov, S.V. Plasma-assisted molecular beam epitaxy of AlGaN heterostructures for deep-ultraviolet optically pumped lasers. Phys. Stat. Sol. 2013, 210, 439–450. [CrossRef] 44. Shatalov, M.; Sun, W.; Jain, R.; Lunev, A.; Hu, X.; Dobrinsky, A.; Bilenko, Yu.; Yang, J.; Garrett, G.A.; Rodak, L.E.; et al. High power AlGaN ultraviolet light emitters. Semicond. Sci. Technol. 2014, 29, 084007. [CrossRef] 45. Nikishin, S.A.; Holtz, M.; Temkin, H. Digital alloys AlN/AlGaN for deep UV light emitting diodes. Jpn. J. Appl. Phys. 2005, 44, 7221–7226. [CrossRef] Appl. Sci. 2018, 8, 2362 13 of 17 46. Sun, W.; Tan, C.-K.; Tansu, N. AlN/GaN Digital alloy for mid- and deep-ultraviolet optoelectronics. Sci. Rep. 2017, 7, 11826–11833. [CrossRef] [PubMed] 47. Gorczyca, I.; Suski, T.; Strak, P.; Staszczak, G.; Christensen, N.E. Band gap engineering of In(Ga)N/GaN short period superlattices. Sci. Rep. 2017, 7, 16055. [CrossRef] [PubMed] 48. Kaneko, M.; Ueta, S.; Horita, M.; Kimoto, T.; Suda, J. Deep-ultraviolet light emission from 4H-AlN/4H-GaN short-period superlattice grown on 4H-SiC (11–20). Appl. Phys. Lett. 2018, 112, 012106. [CrossRef] 49. Neave, J.H.; Joyce, B.A.; Dobson, P.J.; Norton, N. Dynamics of films growth of GaAs by MBE from RHEED observations. Appl. Phys. A 1983, 31, 1–8. [CrossRef] 50. Yang, Z.; Li, L.K.; Wang, W.I. High-quality GaN and AlN grown by gas-source molecular beam epitaxy using ammonia as the nitrogen source. J. Vac. Sci. Technol. B 1996, 14, 2354–2356. [CrossRef] 51. Grandjean, N.; Massies, J. GaN and Al Ga N molecular beam epitaxy monitored by reflection high-energy x 1x electron diffraction. Appl. Phys. Lett. 1997, 71, 1816–1818. [CrossRef] 52. Daudin, B.; Widmann, F. Layer-by-layer grown of AlN and GaN by molecular beam epitaxy. J. Cryst. Growth 1997, 182, 1–5. [CrossRef] 53. Nikishin, S.A.; Borisov, B.A.; Chandolu, A.; Kuryatkov, V.V.; Temkin, H.; Holtz, M.; Mokhov, E.N.; Makarov, Y.; Helava, H. Short period superlattices of AlN/Al Ga N grown on AlN substrates. Appl. Phys. Lett. 2004, 0.08 0.92 85, 4355–4357. [CrossRef] 54. Borisov, B.; Nikishin, S.; Kuryatkov, V.; Temkin, H. Enhanced deep ultraviolet luminescence from AlGaN quantum wells grown in the three dimensional mode. Appl. Phys. Lett. 2005, 87, 191902. [CrossRef] 55. Borisov, B.A.; Nikishin, S.A.; Kuryatkov, V.V.; Kuchinski, V.I.; Holtz, M.; Temkin, H. Enhanced radiative recombination in AlGaN quantum wells grown by molecular-beam epitaxy. Semiconductors 2006, 40, 454–458. [CrossRef] 56. Panish, M.B.; Temkin, H. Gas Source Molecular Beam Epitaxy. Growth and Properties of Phosphorus Containing III–V Heterostructures; Springer Series in Materials Science; Springer: Berlin/Heidelberg, Germany, 1993; Volume 26, pp. 115–230. ISBN 3-540-56540-X (Berlin) or 0-387-56540-X (New York). 57. Nikishin, S.A.; Borisov, B.A.; Garrett, G.A.; Sarney, W.L.; Sampath, A.V.; Shen, H.; Wraback, M.; Holtz, M. Enhanced luminescence from Al Ga N/Al Ga N quantum wells grown by gas source molecular beam x 1x y 1y epitaxy with ammonia. Proc. SPIE 2007, 6473, 647306. 58. Nikishin, S.A.; Borisov, B.A.; Kuryatkov, V.V.; Holtz, M.; Garrett, G.A.; Sarney, W.L.; Sampath, A.V.; Shen, H.; Wraback, M. Correlations between the growth modes and luminescence properties of AlGaN quantum structures. Jpn. J. Appl. Phys. 2008, 47, 1556–1558. [CrossRef] 59. Islam, S.M.; Lee, K.; Verma, J.; Protasenko, V.; Rouvimov, S.; Bharadwaj, S.; Xing, H.G.; Jena, D. MBE-grown 232–270 nm deep-UV LEDs using monolayer thin binary GaN/AlN quantum heterostructures. Appl. Phys. Lett. 2017, 110, 041108. [CrossRef] 60. Islam, S.M.; Protasenko, V.; Lee, K.; Rouvimov, S.; Verma, J.; Xing, H.G.; Jena, D. Deep-UV emission at 219 nm from ultrathin MBE GaN/AlN quantum heterostructures. Appl. Phys. Lett. 2017, 111, 091104. [CrossRef] 61. Chandolu, A.; Nikishin, S.; Holtz, M.; Temkin, H. X-ray diffraction study of AlN/AlGaN short-period superlattices. J. Appl. Phys. 2007, 102, 114909. [CrossRef] 62. Kandaswamy, P.K.; Guillot, F.; Bellet-Amalric, E.; Monroy, E.; Nevou, L.; Tchernycheva, M.; Michon, A.; Julien, F.H.; Baumann, E.; Giorgetta, F.R.; et al. GaN/AlN short-period superlattices for intersubband optoelectronics: A systematic study of their epitaxial growth, design, and performance. J. Appl. Phys. 2008, 104, 093501. [CrossRef] 63. Kladko, V.; Kuchuk, A.; Lytvyn, P.; Yefanov, O.; Safriuk, N.; Belyaev, A.; Mazur, Y.I.; DeCuir, E.A., Jr.; Ware, M.E.; Salamo, G.J. Substrate effects on the strain relaxation in GaN/AlN short-period superlattices. Nanoscale Res. Lett. 2012, 7, 289. [CrossRef] [PubMed] 64. Noh, Y.-K.; Seo, J.-H.; Choi, H.-S.; Kim, M.-D.; Oh, J.-E. Thickness dependence of temperature-induced emission mechanism in InGaN/AlGaN short-period superlattices. J. Appl. Phys. 2012, 112, 043102. [CrossRef] 65. Kutt, R.N.; Shcheglov, M.P.; Ratnikov, V.V.; Yagovkina, M.A.; Davydov, V.Y.; Smirnov, A.N.; Rozhavskaya, M.M.; Zavarin, E.E.; Lundin, V.V. X-ray diffraction study of short-period AlN/GaN superlattices. Crystallogr. Rep. 2013, 58, 953–958. [CrossRef] 66. Pinquier, C.; Frandon, J.; Demangeot, F.; Smirnov, M.; Sarua, A.; Kuball, M.; Monroy, E.; Daudin, B. Behavior of phonons in short-period GaN-AlN superlattices. Phys. Stat. Sol. 2004, 1, 2706–2710. [CrossRef] Appl. Sci. 2018, 8, 2362 14 of 17 67. Davydov, V.Yu.; Roginskii, E.M.; Smirnov, A.N.; Kitaev, Y.E.; Yagovkina, M.A.; Kyutt, R.N.; Rozhavskaya, M.M.; Zavarin, E.E.; Lundin, W.V.; Smirnov, M.B. Lattice dynamics of short-period AlN/GaN short period superlattices: Theory and experiments. Phys. Stat. Sol. 2013, 210, 484–487. [CrossRef] 68. Pankin, D.V.; Smirnov, M.B.; Davydov, V.Y.; Smirnov, A.N.; Zavarin, E.E.; Lundin, W.V. Elastic Strains and Delocalized Optical Phonons in AlN/GaN Superlattices. Semiconductors 2016, 50, 1043–1048. [CrossRef] 69. Sarkar, K.; Datta, D.; Gosztola, D.J.; Shi, F.; Nicholls, A.; Stroscio, M.A.; Dutta, M. Raman analysis of phonon modes in a short period AlN/GaN superlattice. Superlattices Microstruct. 2018, 115, 116–122. [CrossRef] 70. Stolyarchuk, N.; Markurt, T.; Courville, A.; March, K.; Tottereau, O.; Vennéguès, P.; Albrecht, M. Impact of sapphire nitridation on formation of Al-polar inversion domains in N-polar AlN epitaxial layers. J. Appl. Phys. 2017, 122, 155303. [CrossRef] 71. Schuck, P.J.; Mason, M.D.; Grober, R.D.; Ambacher, O.; Lima, A.P.; Miskys, C.; Dimitrov, R.; Stutzmann, M. Spatially resolved photoluminescence of inversion domain boundaries in GaN-based lateral polarity heterostructures. Appl. Phys. Lett. 2001, 79, 952–954. [CrossRef] 72. Shubina, T.V.; Jmerik, V.N.; Tkachman, M.G.; Vekshin, V.A.; Ratnikov, V.V.; Toropov, A.A.; Sitnikova, A.A.; Ivanov, S.V.; Bergman, J.P.; Karlsson, F.; et al. Nanometric-scale fluctuations of intrinsic electric fields in GaN/AlGaN quantum wells with inversion domains. Phys. Stat. Sol. 2002, 234, 919–923. [CrossRef] 73. SiLENSe—Software Tool for Light Emitting Diode (LED) Bandgap Engineering. Available online: https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.str-soft. com%2Fproducts%2FSiLENSe%2Findex.htm&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=YpequeQnoIjnZ5cNklA7U5Zts7E%2FGL0Jz%2F1oAfZEdX4%3D& reserved=0 (accessed on 22 November 2018). 74. Crosslight Software Inc. Available online: https://na01.safelinks.protection.outlook.com/ ?url=https%3A%2F%2Fcrosslight.com%2F&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=vCC9BOBAXHLNXsh6KhW1CvooXSMJOC2KdKHyv77BAjI%3D& reserved=0 (accessed on 22 November 2018). 75. Synopsys, Inc. Available online: https://na01.safelinks.protection.outlook.com/?url= https%3A%2F%2Fwww.synopsys.com%2F&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=33YOp0jlN06RqaUg2AIjE6oOB2OH%2Bi%2BNnzafC3Y2lgM%3D& reserved=0 (accessed on 22 November 2018). 76. Bandgap Engineering Superlattice Simulation Tool (BESST)—Package for Simulation of Optoelectronic Devices Based on Group-III Nitride superlattices. Available online: http://www.str-soft.com/products/ BESST/index.htm (accessed on 22 November 2018). 77. Kozodoy, P.; Hansen, M.; DenBaars, S.P.; Mishra, U.K. Enhanced Mg doping efficiency in Al Ga N/GaN 0.2 0.8 superlattices. Appl. Phys. Lett. 1999, 74, 3681–3683. [CrossRef] 78. Bulashevich, K.A.; Karpov, S.Y. Heterojunctions between group-III nitride short period superlattices. Phys. Stat. Sol. 2005, 2, 2394–2398. [CrossRef] 79. Holtz, M.; Prokofyeva, T.; Seon, M.; Copeland, K.; Vanbuskirk, J.; Williams, S.; Nikishin, S.; Tretyakov, V.; Temkin, H. Composition dependence of the optical phonon energies in hexagonal Al Ga N. J. Appl. Phys. 1x 2001, 89, 7977–7982. [CrossRef] 80. Kuryatkov, V.; Chandolu, A.; Borisov, B.; Kipshidze, G.; Zhu, K.; Nikishin, S.; Temkin, H.; Holtz, M. Solar-blind ultraviolet photodetectors based on superlattices of AlN/AlGa(In)N. Appl. Phys. Lett. 2003, 82, 1323–1325. [CrossRef] 81. Herzog, J.B.; Mintairov, A.M.; Sun, K.; Cao, Y.; Jena, D.; Merz, J.L. Infrared reflectivity spectroscopy of optical phonons in short-period AlGaN/GaN superlattices. Proc. SPIE 2007, 6648, 664803. 82. Marx, G.; Engelbrecht, J.A.A.; Lee, M.E.; Wagener, M.C.; Henry, A. Comments on a peak of Al Ga N x 1x observed by infrared reflectance. Infrared Phys. Technol. 2016, 76, 493–499. [CrossRef] 83. Pendlebury, S.T.; Lynam, P.; Mowbray, D.J.; Parbrook, P.J.; Wood, D.A.; Lada, M.; O’Neill, J.P.; Cullis, A.G.; Skolnick, M.S. Optical characterization of AlGaN epitaxial layers and GaN/AlGaN quantum wells. Phys. Stat. Sol. 2001, 188, 871–875. [CrossRef] Appl. Sci. 2018, 8, 2362 15 of 17 84. Adelmann, C.; Sarigiannidou, E.; Jalabert, D.; Hori, Y.; Rouvière, J.-L.; Daudin, B.; Fanget, S.; Bru-Chevallier, C.; Shibata, T.; Tanaka, M. Growth and optical properties of GaN/AlN quantum wells. Appl. Phys. Lett. 2003, 82, 4154–4156. [CrossRef] 85. Paskov, P.; Monemar, B.; Kamyiama, S.; Amano, H.; Akasaki, I. Optical Properties of undoped, n-doped and p-doped GaN/AlN superlattices. Mater. Res. Soc. Symp. Proc. 2007, 955, 0955-I12-03. [CrossRef] 86. Da-Bing, L.; Wei-Guo, H.; Hideto, M.; Kazumasa, H.; Hang, S. Enhanced deep ultraviolet emission from Si-doped Al Ga N/AlN MQWs. Chin. Phys. B 2010, 19, 127891. [CrossRef] 1x 87. Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Nikishin, S.; Holtz, M.; Temkin, H. Si-doped Al Ga N (0.56  x  1) layers grown by molecular beam epitaxy with ammonia. Appl. Phys. Lett. 2005, x 1x 87, 132106–132108. [CrossRef] 88. Taniyasu, Y.; Kasu, M.; Toshiki Makimoto, T. Electrical conduction properties of n-type Si-doped AlN with 2 1 1 high electron mobility (>100 cm V s ). Appl. Phys. Lett. 2004, 85, 4672–4674. [CrossRef] 89. Contreras, S.; Konczewicz, L.; Messaoud, J.B.; Peyre, H.; Khalfioui, M.A.; Matta, S.; Leroux, M.; Damilano, B.; Brault, J. High temperature electrical transport study of Si-doped AlN. Superlattices Microstruct. 2016, 98, 253–258. [CrossRef] 90. Pampili, P.; Parbrook, P.J. Doping on III-nitride materials. Mat. Sci. Semicond. Proc. 2017, 62, 180–191. [CrossRef] 91. Nikishin, S.A.; Kuryatkov, V.V.; Chandolu, A.; Borisov, B.A.; Kipshidze, G.D.; Ahmad, I.; Holtz, M.; Temkin, H. Deep ultraviolet light emitting diodes based on short period superlattices of AlN/AlGa(In)N. Jpn. J. Appl. Phys. 2003, 42, L1362–L1365. [CrossRef] 92. Yun, J.; Choi, K.; Mathur, K.; Kuryatkov, V.; Borisov, B.; Kipshidze, G.; Nikishin, S.; Temkin, H. Low resistance ohmic contacts to digital alloys of n-AlGaN/AlN. IEEE Electron Device Lett. 2006, 27, 22–24. [CrossRef] 93. Kozodoy, P.; Xing, H.; DenBaars, S.P.; Mishra, U.K.; Saxler, A.; Perrin, R.; Elhamri, S.; Mitchel, W.C. Heavy doping effects in Mg-doped GaN. J. Appl. Phys. 2000, 87, 1832–1835. [CrossRef] 94. Jiang, X.-H.; Shi, J.-J.; Zhang, M.; Zhong, H.-X.; Huang, P.; Ding, Y.-M.; He, Y.-P.; Cao, X. Reduction of the Mg acceptor activation energy in GaN, AlN, Al Ga N and Mg d-doping (AlN) /(GaN) : The strain effect. 0.83 0.17 Ga 5 1 J. Phys. D Appl. Phys. 2015, 48, 475104. [CrossRef] 95. Nam, K.B.; Nakarmi, M.L.; Li, J.; Lin, J.Y.; Jiang, H.X. Mg acceptor level in AlN probed by deep ultraviolet photoluminescence. Appl. Phys. Lett. 2003, 83, 878–880. [CrossRef] 96. Liang, Y.-H.; Towe, E. Progress in efficient doping of high aluminum-containing group III-nitrides. Appl. Phys. Rev. 2018, 5, 011107. [CrossRef] 97. Ilegems, M.; Dingle, R. Luminescence of Mg- and Be- doped GaN. J. Appl. Phys. 1973, 44, 4234–4235. [CrossRef] 98. Gelhausen, O.; Phillips, M.R.; Goldys, E.M.; Paskova, T.; Monemar, B.; Strassburg, M.; Hoffmann, A. Dissociation of H-related defect complexes in Mg-doped GaN. Phys. Rev. B 2004, 69, 125210. [CrossRef] 99. Amano, H.; Kito, M.; Hiramatsu, K.; Akasaki, I. P-type conduction in Mg-doped GaN treated with low-energy electron beam irradiation (LEEBI). Jpn. J. Appl. Phys. 1989, 28, L2112–L2114. [CrossRef] 100. Chernyak, L.; Osinsky, A.; Fuflyigin, V.; Schubert, E.F. Electron beam-induced increase of electron diffusion length in -type GaN and AlGaN/GaN superlattices. Appl. Phys. Lett. 2000, 77, 875–877. [CrossRef] 101. Phillips, M.R.; Merklein, M.; Hardy, T.; Manning, T.J.; Wintrebert-Fouquet, M. A mechanism for Mg acceptor activation in GaN by low energy electron beam irradiation. Microsc. Microanal. 2010, 16, 816–817. [CrossRef] 102. Korotkov, R.Y.; Gregie, J.M.; Wessels, B.W. Electrical properties of p-type GaN:Mg codoped with oxygen. Appl. Phys. Lett. 2001, 78, 222–224. [CrossRef] 103. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Kudryavtsev, Yu.; Asomoza, R.; Nikishin, S.; Temkin, H. Mg and O codoping in p-type GaN and Al Ga N (0 < x < 0.08). Appl. Phys. Lett. 2002, 80, 2910–2912. [CrossRef] x 1x 104. Wu, R.Q.; Shen, L.; Yang, M.; Sha, Z.D.; Cai, Y.Q.; Feng, Y.P.; Huang, Z.G.; Wu, Q.Y. Enhancing hole concentration in AlN by Mg:O codoping: Ab initio study. Phys. Rev. B 2008, 77, 073203. [CrossRef] 105. Khan, M.A.; Adviarahan, V.; Zhang, J.P.; Chen, C.; Kuokstis, E.; Chitnis, A.; Shatalov, M.; Yang, J.W.; Simin, G. Stripe geometry ultraviolet light emitting diodes at 305 nanometers using quaternary AlInGaN multiple quantum wells. Jpn. J. Appl. Phys. 2001, 40, L1308–L1310. [CrossRef] 106. Lopatiuk-Tirpak, O.; Chernyak, L.; Borisov, B.A.; Kuryatkov, V.V.; Nikishin, S.A.; Gartsman, K. Electron irradiation-induced increase of minority carrier diffusion length, mobility, and lifetime in Mg-doped AlN/AlGaN short period superlattice. Appl. Phys. Lett. 2007, 91, 182103. [CrossRef] Appl. Sci. 2018, 8, 2362 16 of 17 107. Allerman, A.A.; Crawford, M.H.; Miller, M.A.; Lee, S.R. Growth and characterization of Mg-doped AlGaN-AlN short-period superlattices for deep-UV optoelectronic devices. J. Cryst. Growth 2010, 312, 756–761. [CrossRef] 108. Nikishin, S.; Borisov, B.; Mansurov, V.; Pandikunta, M.; Chary, I.; Rajanna, G.; Bernussi, A.; Kudryavtsev, Y.; Asomoza, R.; Bulashevich, K.A.; et al. Short period p-type AlN/AlGaN superlattices for deep UV light emitters. Mater. Res. Soc. Symp. Proc. 2010, 1202, 1202-I10-03. [CrossRef] 109. Shatalov, M.; Gaska, R.; Yang, J.; Shur, M. Method of Doping and Semiconductor Devices. U.S. Patent 8,426,225, 23 April 2013. 110. Martens, M.; Kuhn, C.; Ziffer, E.; Simoneit, T.; Kueller, V.; Knauer, A.; Rass, J.; Wernicke, T.; Einfeldt, S.; Weyers, M.; et al. Low absorption loss p-AlGaN superlattice cladding layer for current-injection deep ultraviolet laser diodes. Appl. Phys. Lett. 2016, 108, 151108. [CrossRef] 111. Suzuki, M.; Nishio, J.; Onomura, M.; Hongo, C. Doping characteristics and electrical properties of Mg-doped AlGaN grown by atmospheric-pressure MOCVD. J. Cryst. Growth 1998, 189/190, 511–515. [CrossRef] 112. Li, J.; Oder, T.N.; Nakarmi, M.L.; Lin, J.Y.; Jiang, H.X. Optical and electrical properties of Mg-doped p-type Al Ga N. Appl. Phys. Lett. 2002, 80, 1210–1212. [CrossRef] x 1x 113. Taniyasu, Y.; Kasu, M.; Makimoto, T. An aluminium nitride light-emitting diode with a wavelength of 210 nanometres. Nature 2006, 441, 325–328. [CrossRef] [PubMed] 114. Kinoshita, T.; Obata, T.; Yanagi, H.; Inoue, S.-I. High p-type conduction in high-Al content Mg-doped AlGaN. Appl. Phys. Lett. 2013, 102, 012105. [CrossRef] 115. Dussaigne, A.; Damilano, B.; Brault, J.; Massies, J.; Feltin, E.; Grandjean, N. High doping level in Mg-doped GaN layers grown at low temperature. J. Appl. Phys. 2008, 103, 013110. [CrossRef] 116. Chuang, S.L.; Chang, C.S. kp method for strained wurtzite semiconductors. Phys. Rev. B 1996, 54, 2491–2504. [CrossRef] 117. Blank, T.V.; Gol’dberg, Y.A. Mechanisms of current flow in metal-semiconductor ohmic contacts. Semiconductors 2007, 41, 1263–1292. [CrossRef] 118. Ho, J.-K.; Jong, C.-S.; Chiu, C.C.; Huang, C.-N.; Shih, K.-K.; Chen, L.-C.; Chen, F.-R.; Kai, J.-J. Low-resistance ohmic contacts to p-type GaN achieved by the oxidation of Ni/Au films. J. Appl. Phys. 1999, 86, 4491. [CrossRef] 119. Lim, S.-H.; Ra, T.-Y.; Kim, W.-Y. Interface observation in Au/Ni/p-GaN studies by HREM and energy-filtering TEM. J. Electron Microsc. 2003, 52, 459–464. [CrossRef] 120. Qiao, D.; Yu, L.S.; Lau, S.S.; Lin, J.Y.; Jiang, H.X.; Haynes, T.E. A study of the Au/Ni ohmic contact on p-GaN. J. Appl. Phys. 2000, 88, 4196–4200. [CrossRef] 121. Chary, I.; Chandolu, A.; Borisov, B.; Kuryatkov, V.; Nikishin, S.; Holtz, M. Influence of surface treatment and annealing temperature on the formation of low-resistance Au/Ni ohmic contacts to p-GaN. J. Electron. Mater. 2009, 38, 545–550. [CrossRef] 122. Chary, I.; Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Nikishin, S.; Holtz, M. Low resistance ohmic contacts to p-type GaN and AlGaN. Mater. Res. Soc. Symp. Proc. 2009, 1108, 1108-A09-30. 123. Nikishin, S.; Chary, I.; Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Karpov, S.Y.; Holtz, M. Mechanism of curriers injection in (Ni/Au)/p-Al Ga N:Mg (0  x < 0.1) ohmic contacts. Appl. Phys. Lett. 1x 2009, 95, 163502. [CrossRef] 124. Nikishin, S.; Borisov, B.; Kuryatkov, V.; Usikov, A.; Dmitriev, V.; Holtz, M. Deep UV AlGaN light emitting diodes grown by gas source molecular beam epitaxy on sapphire and AlGaN/sapphire substrates. Proc. SPIE 2006, 6121, 6121OT-1-16. [CrossRef] 125. Moustakas, T.D.; Liao, Y.; Kao, C.-K.; Thomidis, C.; Bhattacharyya, A.; Bhattarai, D.; Noldawer, A. Deep UV LEDs with high IQE based on AlGaN alloys with strong band structure potential fluctuations. Proc. SPIE 2012, 8278, 8278OL-1-11. [CrossRef] 126. Rass, J.; Kolbe, T.; Ploch, N.L.; Wernicke, T.; Mehnke, F.; Kuhn, C.; Enslin, J.; Guttmann, M.; Reich, C.; Mogilatenko, A.; et al. High power UV-B LEDs with long lifetime. Proc. SPIE 2015, 9363, 93631K-1-13. [CrossRef] 127. Mondal, R.K.; Chatterjee, V.; Singh, S.; Islam, S.M.; Pal, S. Optimization of structural parameters for highly efficient AlGaN based deep ultraviolet light emitting diodes. Superlattices Microstruct. 2017, 112, 339–352. [CrossRef] Appl. Sci. 2018, 8, 2362 17 of 17 128. Liu, M.; Zhou, S.; Liu, X.; Gao, Y.; Ding, X. Comparative experimental and simulation studies of high-power AlGaN-based 353 nm ultraviolet flip-chip and top-emitting LEDs. Jpn. J. Appl. Phys. 2018, 57, 031001. [CrossRef] 129. Bulashevich, K.A.; Ramm, M.S.; Karpov, S.Yu. Effects of electron and optical confinement on performance of UV laser diodes. Phys. Stat. Sol. 2009, 6, 603–606. [CrossRef] 130. Liang, Y.-H.; Towe, E. Heavy Mg-doping of (AlGa)N films for potential applications in deep ultraviolet light-emitting structures. J. Appl. Phys. 2018, 123, 095303. [CrossRef] 131. Park, J.H.; Kim, J.K.; Cho, J. Observation of space-charge-limited current in AlGaN/GaN ultraviolet light emitting diodes. Mater. Lett. 2018, 214, 217–219. [CrossRef] 132. Liu, D.; Cho, S.J.; Park, J.; Seo, J.-H.; Dalmau, R.; Zhao, D.; Kim, K.; Gong, J.; Kim, M.; Lee, I.-K.; et al. 229 nm UV LEDs on aluminum nitride single crystal substrates using p-type silicon for increased hole injection. Appl. Phys. Lett. 2018, 112, 081101. [CrossRef] © 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Sciences Multidisciplinary Digital Publishing Institute

III-Nitride Short Period Superlattices for Deep UV Light Emitters

Applied Sciences , Volume 8 (12) – Nov 23, 2018

Loading next page...
 
/lp/multidisciplinary-digital-publishing-institute/iii-nitride-short-period-superlattices-for-deep-uv-light-emitters-FzG4jtMVUh
Publisher
Multidisciplinary Digital Publishing Institute
Copyright
© 1996-2019 MDPI (Basel, Switzerland) unless otherwise stated
ISSN
2076-3417
DOI
10.3390/app8122362
Publisher site
See Article on Publisher Site

Abstract

applied sciences Review III-Nitride Short Period Superlattices for Deep UV Light Emitters Sergey A. Nikishin Nano Tech Center, Department Electrical and Computer Engineering, Texas Tech University, Lubbock, TX 79423, USA; sergey.a.nikishin@ttu.edu; Tel.: +1-806-834-8807 Received: 17 October 2018; Accepted: 20 November 2018; Published: 23 November 2018 Featured Application: Advanced infrared, visible, and ultraviolet light emitters. Abstract: III-Nitride short period superlattices (SPSLs), whose period does not exceed ~2 nm (~8 monolayers), have a few unique properties allowing engineering of light-emitting devices emitting in deep UV range of wavelengths with significant reduction of dislocation density in the active layer. Such SPSLs can be grown using both molecular beam epitaxy and metal organic chemical vapor deposition approaches. Of the two growth methods, the former is discussed in more detail in this review. The electrical and optical properties of such SPSLs, as well as the design and fabrication of deep UV light-emitting devices based on these materials, are described and discussed. Keywords: III-nitrides; short period superlattices; light emitters 1. Introduction The invention of semiconductor double heterostructure laser [1,2] and the concept of a semiconductor superlattice (SL) [3] can be considered as the foundation of modern semiconductor p–n junction-based light emitters, lasers, and light-emitting diodes. Double heterostructure (DHS) is a semiconductor “sandwich” where a layer of narrow band gap semiconductor (referred to as an active layer or a well, depending on its thickness) is placed between n-type and p-type layers (cladding layers) of the wide bandgap semiconductors. Under forward bias, the electrons and holes are injected from the cladding layers into an active layer, and well confined there. Confinement of these injected carriers leads to very effective radiative recombination in direct bandgap semiconductors. The SL periodic layered structure of different crystalline semiconductors, allows for engineering of a bandgap of active (referred as a well) and cladding (referred as a barrier) layers when their thickness is of about a few nanometers. Note, these structures must be grown into two-dimensional (2D) growth mode, in order to get a flat/abrupt interface between all layers. The SLs based on the direct bandgap semiconductors in different combinations with p–n DHS are used for design and fabrication of light emitters operating in very wide range of wavelengths, ensuring the needs of optical communication, medicine, security, lighting, and agriculture [4–11]. We will discuss only p–n junction-based structures in this review. III-Nitrides, AlN (E = 6.2 eV, [12,13]), GaN (E = 3.4 eV, [13]), InN (E = 0.7 eV, [14–16]), and their g g g AlGaInN, AlGaN, AlInN, InGaN alloys, are direct bandgap semiconductors which would significantly influence the development of new optoelectronic and light-emitting devices throughout the 21st century [17–19]. Most III-nitride (III-N) light-emitting and laser diodes contain different SLs with a period exceeding 4 nm [20–23]. These SLs, in addition to the bandgap engineering, allow for reducing the dislocation density propagating into an active layer of III-N light-emitting diodes (LEDs) grown on the heavily lattice-mismatched substrates, like silicon and sapphire [24–26]. Strain engineering in an active layer of LEDs allows for modification of an internal quantum efficiency of radiative carrier Appl. Sci. 2018, 8, 2362; doi:10.3390/app8122362 www.mdpi.com/journal/applsci Appl. Sci. 2018, 8, 2362 2 of 17 recombination [27,28]. The design and fabrication of such light emitters, based on “long period” SLs, are well described in many books and reviews [29–36]. III-N LEDs based on very short period superlattices (SPSLs, sometimes referenced as digital alloys, DA), periodic structures of GaN/AlN, AlGaN/AlN, AlGaInN/AlN, InGaN/GaN, InN/GaN, and InAlN/GaN having a few monolayer thick wells and barriers, and a period not exceeding 2 nm, are very attractive for the design and fabrication of a new generation of light emitters. The main important difference between SPSLs and SL with a long period, is that carriers tunneling through barriers between quantum wells (QWs) in SPSLs already affect energy levels, and even lead to the formation of minibands (at least in the conduction band). Bandgap behavior of the InGaN/GaN, InN/GaN, and InAlN/GaN SPSLs, and their applications in visible and infrared light emitters, are well summarized in recent publications [37–40]. The GaN/AlN, AlGaN/AlN, and AlGa(In)N/AlN (C < 0.02 mol fraction) SPSLs are very attractive for deep ultraviolet light emitters [41–48]. One of In the attractive features of AlGa(In)N/AlN SPSLs relates to the formation of very sharp heterointerfaces over the entire range of compositions, which makes it possible to obtain well/barrier thicknesses comparable to the interatomic distance, and to make tunneling the main carrier transport mechanism. For InGaN/GaN structures, for example, the roughness of heterointerfaces increases if the composition of In approaches ~20%, which is important for practical applications, since carrier tunneling is difficult. This review aims to summarize the most significant efforts demonstrated in this field since 2002, when the first LED based on AlGa(In)N/AlN SPSLs operating at 280 nm, was demonstrated [41,42]. 2. Growth and Structural Characterization Most of III-N SPSLs for deep UV light emitters were grown using both molecular beam epitaxy (MBE) [41–43] and metal organic chemical vapor deposition (MOCVD) [37,44] methods on (0001) sapphire, Si (111), and (0001) GaN/sapphire template substrates. The detailed analysis of deep UV LED efficiency (internal and external), grown on different substrates, can be found in ref. [44]. One well-known advantage of MBE over MOCVD is the in situ monitoring of the growth process using the reflection high energy electron diffraction (RHEED) [49]. Analysis of the RHEED patterns in real time allows for controlling, at the monolayer (ML) scale, the structural properties of any substrate at the onset of epitaxy, the nucleation process and growth mode of the epitaxial layer, the growth rate of III-N compounds, and the composition of their alloys, by monitoring the period of RHEED intensity oscillations during deposition [50–52]. This statement can be illustrated by a few RHEED patterns. The evolution of RHEED patterns illustrating the onset of gas source MBE (GSMBE) with ammonia on bulk (0001) AlN substrate is shown in Figure 1. As seen in Figure 1a, there are two types of reflections indicated separately by black solid and white dashed arrows. It was shown that this complex RHEED pattern can be attributable to the presence of Al O surface islands [53]. The well-defined (00), (01), 2 3 and (–01) reflections indicated by solid arrows can be attributed to the (1  1) AlN (0001) surface reconstruction at low temperatures. The weak additional reflections, indicated by dashed arrows, arise from formation of crystalline Al O islands on the AlN surface. These islands cannot be removed by 2 3 baking of the AlN substrate at high temperatures, up to ~1100 C [53]. However, nitridation of such a surface, by exposing it to the flux of ammonia for a few minutes at a substrate temperature of ~800 C, yields formation of a pure (1  1) surface structure on AlN (0001), as shown in Figure 1b. This (1  1) surface reconstruction was stable up to 900 C. At this temperature, AlN, Al Ga N, and SPSL of 0.6 0.4 AlN (3 ML)/Al Ga N (3 ML), with a total of 100 pairs, were successfully grown. The entire SPSL 0.08 0.92 was grown in the 2D mode, and formation of a (2  2) surface reconstruction is shown in Figure 1c. The surface was very flat, with the root mean square (rms) roughness of less than 1 nm, as measured by 1  1 m scans, using atomic force microscopy. Figure 2 shows the evolution of the RHEED patterns illustrating 2D!3D!2D growth mode transitions during ammonia GSMBE of Al Ga N (barrier)/Al Ga N (well) structure on 0.55 0.45 0.45 0.55 Al Ga N/AlN buffer grown on (0001) sapphire substrate [54,55]. 0.55 0.45 Appl. Sci. 2018, 8, x FOR PEER REVIEW 3 of 17 (a) (b) (c) Figure 1. Evolution of reflection high energy electron diffraction (RHEED) patterns for different stages at onset of gas source molecular beam epitaxy (GSMBE). (a) The surface of (0001) AlN substrate at low temperatures. The (00), (01), and (–01) reflections from the (1 × 1) AlN surface and reflections from crystalline Al2O3 islands are indicated by arrows; (b) (1 × 1) surface reconstruction of AlN exposed to ammonia; (c) (2 × 2) surface structure after deposition of about 20 pairs of short period superlattices (SPSLs). All the barrier layers were grown in 2D growth mode with 1 × 1 surface reconstruction, as shown Appl. Sci. 2018, 8, 2362 3 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 3 of 17 in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 °C. With the ammonia flux reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increa se was (a) (b) (c) attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest CL intensity and longest PL lifetime for these structures are due to formation of quantum well Figure Figure 1. 1. E Evolution volution of of refle reflection ction high energy electron high energy electron diffract diffraction ion (RHEED) (RHEED) pa patterns tterns for for diff differ erent stages ent stages (QW)/QD regions in AlxGa1−xN/AlyGa1−yN (0.3 < x < 0.45, 0.53 < y ≤ 1) QW structures [57,58]. The at at onset onset of g of gas as sou sour rce ce m molecular olecular beam beam epita epitaxy xy (G (GSMBE). SMBE). ( (a a) The surface ) The surface of of (0001) (0001) AlN AlN sub substrate at strate at approach described in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and low low temperatures. The (00 temperatures. The (00), ), (0(01), 1), and (–01) and (–01) refrleflection ections from s from the (1 the × 1) (1  AlN 1) AlN surface and reflections from surface and reflections successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm fr cry om stalline crystalline Al2O3 is Al laO ndsislands are indiar cae teindicated d by arrows by; ( arr b) (1 ows; × 1) (b surface reconst ) (1  1) surface ruction of reconstr AlN uction exposed of AlN to 2 3 from PAMBE-grown 2 ML thick GaN QDs was also observed [60]. exposed ammonia; ( to c ammonia; ) (2 × 2) surface (c) (2 structure after deposition  2) surface structure after ofdeposition about 20 pairs of of about short period su 20 pairs of short perlattice period s superlattices (SPSLs). (SPSLs). All the barrier layers were grown in 2D growth mode with 1 × 1 surface reconstruction, as shown in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 °C. With the ammonia flux reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth (a) (b) (c) conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm Ga)/NH Ga)/NH3 flux ratios: ( flux ratios: a (a ) 2D-grown barrier at 20 sccm of ammonia; ( ) 2D-grown barrier at 20 sccm of ammonia; (b b) 3D- ) 3D-gr grown well at own well at 5.5 sccm o 5.5 sccm off and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increase was All the barrier layers were grown in 2D growth mode with 1 1 surface reconstruction, as shown Analyzing the state-of-the art results mentioned above and discussed in literature within last attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest in Figure 2a, when an ammonia flux was sustained at 20 sccm. The wells grown under the same two to three years, we can conclude that one of the main current trends aimed at improving internal CL intensity and longest PL lifetime for these structures are due to formation of quantum well ammonia flux also demonstrated 2D growth mode. This mode was maintained at ammonia fluxes quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a low (QW)/QD regions in AlxGa1−xN/AlyGa1−yN (0.3 < x < 0.45, 0.53 < y ≤ 1) QW structures [57,58]. The greater than 7 sccm (N-rich conditions), at a substrate temperature of 795 C. With the ammonia flux defective and highly efficient active layer in such structures. Future research should focus on finding approach described in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and reduced to 5.5 sccm, the RHEED patterns become quite spotty, as shown in Figure 2b. This behavior the optimal ratio and distribution of QDs in the active layer, as well as on the development of the successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm of the RHEED pattern is typical when the growth mode changes from 2D to 3D [56]. These growth growth of LED structures on relatively inexpensive templates or bulk AlN substrates. Of course, it is from PAMBE-grown 2 ML thick GaN QDs was also observed [60]. conditions can be attributed to the “metal (Ga, Al)-rich” conditions, although additional experiments much more effective to design and develop such an active layer using MBE, which provides in situ are required. It was shown that the barrier layer recovers when ammonia flux increased to 20 sccm and the RHEED pattern shows 2D growth mode, as shown in Figure 2c, by the time the next well is grown. Note that significant increase in the deep UV cathodoluminescence (CL) and photoluminescence (PL) emission from such grown structures was observed [57]. The increase was attributed to the formation of quantum dots (QDs) within the wells. It was concluded that the greatest CL intensity and longest PL lifetime for these structures are due to formation of quantum well (QW)/QD regions in Al Ga N/Al Ga N (0.3 < x < 0.45, 0.53 < y  1) QW structur es [57,58]. The approach described x 1x y 1y (a) (b) (c) in [54,55,57,58] was recently adjusted for a plasma-assisted MBE (PAMBE), and successfully used in an active layer of deep UV LEDs emitting at 232 nm [59]. The emission at 219 nm from PAMBE-grown Figure 2. Evolution of RHEED patterns illustrating 2D and 3D growth modes at different (Al + 2 ML thick GaN QDs was also observed [60]. Ga)/NH3 flux ratios: (a) 2D-grown barrier at 20 sccm of ammonia; (b) 3D-grown well at 5.5 sccm of Analyzing the state-of-the art results mentioned above and discussed in literature within last ammonia; (c) next 2D-grown barrier at 20 sccm of ammonia on a 3D-grown well. two to three years, we can conclude that one of the main current trends aimed at improving internal quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a Analyzing the state-of-the art results mentioned above and discussed in literature within last low two to three defective yea and rs, we ca highly n concl efficient ude that one of active layer the ma in suchin str current trends ai uctures. Futuremed resear at i ch mshould proving interna focus on l finding the optimal ratio and distribution of QDs in the active layer, as well as on the development of quantum efficiency (IQE) and external quantum efficiency (EQE) of UV LEDs is the creation of a low the defect growth ive and of high LEDly stef ructur ficien es t act onir ve elatively layer in inexpensive such structures. templates Future or rese bulk arch AlN sho substrates. uld focus on Of fcourse, inding it is much more effective to design and develop such an active layer using MBE, which provides in the optimal ratio and distribution of QDs in the active layer, as well as on the development of the growth of LED structures on relatively inexpensive templates or bulk AlN substrates. Of course, it is much more effective to design and develop such an active layer using MBE, which provides in situ Appl. Sci. 2018, 8, 2362 4 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 4 of 17 situ monitoring of the growth process. Despite the fact that the MOCVD process dominates in the monitoring of the growth process. Despite the fact that the MOCVD process dominates in the industrial growth of such structures, the results obtained using the MPE should make it possible to industrial growth of such structures, the results obtained using the MPE should make it possible to identify the most important structural and morphological factors influencing the radiative efficiency of identify the most important structural and morphological factors influencing the radiative efficiency the active layer of the LED. These new concepts can then be transferred to the MOCVD processes. of the active layer of the LED. These new concepts can then be transferred to the MOCVD processes. High resolution X-ray diffraction (HR-XRD) of AlN/AlGaN, AlN/GaN, and AlGaN/InGaN were High resolution X-ray diffraction (HR-XRD) of AlN/AlGaN, AlN/GaN, and AlGaN/InGaN were carried out by many researchers [61–65] in order to estimate strain and dislocation density. carried out by many researchers [61–65] in order to estimate strain and dislocation density. Note, HR-XRD measurements, in conjunction with Raman measurements, allow for estimation Note, HR-XRD measurements, in conjunction with Raman measurements, allow for estimation of the residual strain in SPSLs more precisely [66–69]. Usually, HR-XRD studies are carried out of the residual strain in SPSLs more precisely [66–69]. Usually, HR-XRD studies are carried out using using a high-resolution diffractometer in double- and triple-axis alignment. A long range 2-! scan a high-resolution diffractometer in double- and triple-axis alignment. A long range 2θ-ω scan of the of the (0002) reflection is shown in Figure 3 for a typical AlN/Al Ga N SPSL grown on (0001) (0002) reflection is shown in Figure 3 for a typical AlN/AlxGa x 1-x 1N x SPSL grown on (0001) Al Ga N/AlN/sapphire template. Individual peaks corresponding to the AlN and Al Ga N Al0.4 0.4Ga0.6 0.6 N/AlN/sapphire template. Individual peaks corresponding to the AlN and Al0.4Ga0.6 0.4 N buffe 0.6 r buffer layers, and the 0th, 1, and 2 satellites of the SPSL are well defined in Figure 3. The average layers, and the 0th, ±1, and ±2 satellites of the SPSL are well defined in Figure 3. The average SPSL SPSL composition, 0.68, was determined from the 2 position of the 0th peak [61]. From the position of composition, 0.68, was determined from the 2θ position of the 0th peak [61]. From the position of the the 0th and 1 satellite peaks, the average period of the SPSL was determined to be 2.236 nm. Using 0th and ±1 satellite peaks, the average period of the SPSL was determined to be 2.236 nm. Using the the experimentally determined period, and assuming well composition of Al Ga N and pure AlN experimentally determined period, and assuming well composition of Al0.08 0.08Ga0. 0.92 92N and pure AlN barriers, the well and the barrier thicknesses are found to be 0.808 nm and 1.428 nm, respectively [61]. barriers, the well and the barrier thicknesses are found to be 0.808 nm and 1.428 nm, respectively [61]. Simulations based on the experimentally determined SPSL parameters yield an excellent fit to the Simulations based on the experimentally determined SPSL parameters yield an excellent fit to the experimental data, as shown in Figure 3. The deviation of the well and barrier thicknesses from their experimental data, as shown in Figure 3. The deviation of the well and barrier thicknesses from their integer lattice parameters (integer ML multiples) can be attributed to many factors, including residual integer lattice parameters (integer ML multiples) can be attributed to many factors, including residual strain in the SPSL, formation of interfacial layers, interface roughness, composition fluctuations in the strain in the SPSL, formation of interfacial layers, interface roughness, composition fluctuations in well and barrier, stacking faults (SFs), and inversion domain boundaries (IDBs). The detailed analysis the well and barrier, stacking faults (SFs), and inversion domain boundaries (IDBs). The detailed of significance of all these factors was conducted in reference [61]. analysis of significance of all these factors was conducted in reference [61]. Figure 3. A long range 2-! scan of (0002) reflection for AlN/Al Ga N (0.07 < x < 0.09) obtained 1x Figure 3. A long range 2θ-ω scan of (0002) reflection for AlN/AlxGa1−xN (0.07 < x < 0.09) obtained using using a hybrid X-ray mirror. Black line—data [61], red line—simulations (courtesy of Dr. A. Chandolu). a hybrid X-ray mirror. Black line—data [61], red line—simulations (courtesy of Dr. A. Chandolu). Crystalline microstructure of any semiconductor is a very important factor influencing the Crystalline microstructure of any semiconductor is a very important factor influencing the performance of all light emitters. Cross-sectional structure of SPSLs should be investigated by performance of all light emitters. Cross-sectional structure of SPSLs should be investigated by transmission electron microscopy (TEM), in order to get a nanoscale resolution. Two TEM cross-sections transmission electron microscopy (TEM), in order to get a nanoscale resolution. Two TEM cross- of AlN/AlGaN SPSL, grown by GSMBE, are shown in Figure 4. Although the growth conditions sections of AlN/AlGaN SPSL, grown by GSMBE, are shown in Figure 4. Although the growth (substrate temperature, flux ratio, growth time) were the same, the crystalline quality of these SPSLs conditions (substrate temperature, flux ratio, growth time) were the same, the crystalline quality of were very different. It is clearly seen that SPSL grown directly on sapphire substrate contains a very these SPSLs were very different. It is clearly seen that SPSL grown directly on sapphire substrate high density of inversion domain boundaries (IDBs). These domains start to grow from substrate/layer contains a very high density of inversion domain boundaries (IDBs). These domains start to grow interface, mostly due to incomplete nitridation of sapphire at the onset of epitaxial growth [45,70]. from substrate/layer interface, mostly due to incomplete nitridation of sapphire at the onset of It was shown [71] that IDBs dominate the light emission process in GaN containing these defects. epitaxial growth [45,70]. It was shown [71] that IDBs dominate the light emission process in GaN A similar result was reported for MBE grown AlGaN/GaN SLs [72]. However, SPSLs with high density containing these defects. A similar result was reported for MBE grown AlGaN/GaN SLs [72]. of IDBs have inferior electrical properties [45] and cannot be used in the preparation of light-emitting However, SPSLs with high density of IDBs have inferior electrical properties [45] and cannot be used devices. The most detailed impact of sapphire nitridation on the formation of inverse domains in AlN in the preparation of light-emitting devices. The most detailed impact of sapphire nitridation on the layers grown by MOCVD was discussed in a recent paper [70]. formation of inverse domains in AlN layers grown by MOCVD was discussed in a recent paper [70]. Appl. Sci. 2018, 8, 2362 5 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 17 (a) (b) Figure Figure 4. 4. TEM TEM cross- cross-s sect ection ion of of AlN/A AlN/AlGaN lGaN S SPSLs PSLs (cou (courtesy rtesy of Dr. S. of Dr. S. N. N. G G. . C Chu) hu) ((a a) grow ) grown n direct directly ly on on bar bare (0001) sapphire; ( e (0001) sapphire; (bb ) ) grown on ~50 nm thick AlN buf grown on ~50 nm thick AlN bufferflayer er layer. The white scale bars are 20 nm . The white scale bars are 20 nm long. The long. The [0001 [0001] direction ] direis ction shown is sh by own by black a black arrows.rrows. 3. Bandgap of AlN/AlGa(In)N SPSLs 3. Bandgap of AlN/AlGa(In)N SPSLs The bandgap structure of III-Ns SPSLs convenient for the deep UV light emitters can be simulated The bandgap structure of III-Ns SPSLs convenient for the deep UV light emitters can be using different software [73–75]. The BESST (Bandgap Engineering Superlattice Simulation Tool) simulated using different software [73–75]. The BESST (Bandgap Engineering Superlattice Simulation commercially available package from STR Inc. [76] was used to analyze the results of different Tool) commercially available package from STR Inc. [76] was used to analyze the results of different teams [77,78], as well as most of our experimental results. This software is suitable for modelling teams [77,78], as well as most of our experimental results. This software is suitable for modelling optoelectronic devices utilizing SPSLs as essential units of their heterostructure designs. The BESST optoelectronic devices utilizing SPSLs as essential units of their heterostructure designs. The BESST calculates the SPSL electron and hole minibands using a tight-binding approach and numerical solution calculates the SPSL electron and hole minibands using a tight-binding approach and numerical of the Schrödinger equation with account of complex valence band structure of III-nitride compounds. solution of the Schrödinger equation with account of complex valence band structure of III-nitride Coupled solution of the Poisson equation for electric potential, accounting for polarization charges compounds. Coupled solution of the Poisson equation for electric potential, accounting for at the heterostructure interfaces, and discrete drift-diffusion transport equations, allows building up polarization charges at the heterostructure interfaces, and discrete drift-diffusion transport the band diagram of a device at an arbitrary bias and calculating the corresponding electron and equations, allows building up the band diagram of a device at an arbitrary bias and calculating the hole currents, as well as the radiative recombination rate and emission spectrum. Field-dependent corresponding electron and hole currents, as well as the radiative recombination rate and emission mobilities of electrons and holes used in the transport equations are found, self-consistently, with the spectrum. Field-dependent mobilities of electrons and holes used in the transport equations are simulated minibands of SPSLs. found, self-consistently, with the simulated minibands of SPSLs. Fourier-transform infrared optical reflectance (FTIR) [79–82], photoluminescence (PL) [83–85], Fourier-transform infrared optical reflectance (FTIR) [79–82], photoluminescence (PL) [83–85], and cathodoluminescence (CL) [41,42,86] are widely used to estimate the effective bandgaps of and cathodoluminescence (CL) [41,42,86] are widely used to estimate the effective bandgaps of AlN/Ga(Al,In)N SPSLs. At room temperature, these methods are mostly qualitative, although still AlN/Ga(Al,In)N SPSLs. At room temperature, these methods are mostly qualitative, although still very useful for express control and adjustment of light-emitter properties during fabrication. The room very useful for express control and adjustment of light-emitter properties during fabrication. The temperature FTIR and CL were successfully used to facilitate fabrication of the first deep UV LEDs room temperature FTIR and CL were successfully used to facilitate fabrication of the first deep UV operating at 280 nm using undoped and n- and p-type AlN/Al Ga N SPSLs [41,42]. x 1x LEDs operating at 280 nm using undoped and n- and p-type AlN/AlxGa1−xN SPSLs [41,42]. As an example, the experimental FTIR and CL effective bandgaps of AlN/Al Ga N SPSLs 0.08 0.92 As an example, the experimental FTIR and CL effective bandgaps of AlN/Al0.08Ga0.92N SPSLs and and simulations of these structures obtained using two different approaches [46,76] are shown in simulations of these structures obtained using two different approaches [46,76] are shown in Figure Figure 5. One can see that both simulations provide the slope of the optical energy gap dependence 5. One can see that both simulations provide the slope of the optical energy gap dependence on the on the SPSL period (actually, on the AlN barrier “effective” width), similar to that obtained by CL, SPSL period (actually, on the AlN barrier “effective” width), similar to that obtained by CL, whereas whereas FTIR data demonstrate a different slope. This may have originated from the fact that FTIR FTIR data demonstrate a different slope. This may have originated from the fact that FTIR measures measures the spectral dependence of light absorption, which may involve higher electron minibands the spectral dependence of light absorption, which may involve higher electron minibands and lower and lower hole minibands or levels, whereas luminescence occurs mainly from the ground state hole minibands or levels, whereas luminescence occurs mainly from the ground state minibands, due minibands, due to their dominant occupation. Simulations by BESST show that SPSLs, regarded here, to their dominant occupation. Simulations by BESST show that SPSLs, regarded here, possess two possess two different electron minibands and up to three heavy- and light-hole minibands, which can different electron minibands and up to three heavy- and light-hole minibands, which can be be considered as single energy levels at large SPSL periods because of miniband narrowing. Hence, considered as single energy levels at large SPSL periods because of miniband narrowing. Hence, contribution of the extra minibands to the light absorption may explain the difference in the optical contribution of the extra minibands to the light absorption may explain the difference in the optical energy gap determination by FTIR and CL. energy gap determination by FTIR and CL. Of course, the Stokes shift, which is related to the excited state configuration in the well material, Of course, the Stokes shift, which is related to the excited state configuration in the well material, is a factor in all radiative recombination processes and, therefore, should also yield to red shift of is a factor in all radiative recombination processes and, therefore, should also yield to red shift of the the CL’s estimated bandgap. Note that scatter in the experimental data shown in Figure 5 can be CL’s estimated bandgap. Note that scatter in the experimental data shown in Figure 5 can be attributed to monolayer level uncertainty in the well and barrier thickness across the wafer, as well to attributed to monolayer level uncertainty in the well and barrier thickness across the wafer, as well local composition fluctuations in the well alloy [61,65]. to local composition fluctuations in the well alloy [61,65]. Appl. Sci. 2018, 8, 2362 6 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 17 Figure 5. Experimental optical reflectance and cathodoluminescence (CL)-obtained effective bandgaps Figure 5. Experimental optical reflectance and cathodoluminescence (CL)-obtained effective of AlN/Al Ga N SPSLs vs its period, grown with two nominal well thicknesses (2 and 3 0.08 0.98 bandgaps of AlN/Al0.08Ga0.98N SPSLs vs its period, grown with two nominal well thicknesses (2 and 3 monolayers (MLs), blue and green symbols, respectively) [45]. Theoretical simulations based on monolayers (MLs), blue and green symbols, respectively) [45]. Theoretical simulations based on two two approaches [46,76] are shown by continuous dashed and solid curves, respectively. approaches [46,76] are shown by continuous dashed and solid curves, respectively. 4. AlN/AlGa(In)N SPSL Doping and Ohmic Contacts 4. AlN/AlGa(In)N SPSL Doping and Ohmic Contacts The efficiency of deep UV LEDs is very sensitive to the doping level of p- and n-type emitters. The efficiency of deep UV LEDs is very sensitive to the doping level of p- and n-type emitters. There are no significant issues with n-type doping of III-Ns compounds and their alloys, since Si, There are no significant issues with n-type doping of III-Ns compounds and their alloys, since Si, a a common n-type dopant, behaves as a shallow donor, even in wide bandgap AlGaN alloys [87]. common n-type dopant, behaves as a shallow donor, even in wide bandgap AlGaN alloys [87]. Although the activation energy of Si significantly increases in AlN [88–90], the resistivity of Si-doped Although the activation energy of Si significantly increases in AlN [88–90], the resistivity of Si-doped AlN/Al Ga (In)N (0.05 < x < 0.1) SPSLs is very low, 0.015–0.040 Wcm, and the electron concentration x 1x 19 3 AlN/AlxGa1−x(In)N (0.05 < x < 0.1) SPSLs is very low, 0.015–0.040 Ω·cm, and the electron concentration exceeds 10 cm [42,91]. Such n-type SPSLs can also be used as contact layers. Using a Ti/Al/Ti/Au 19 −3  5 2 exceeds 10 cm [42,91]. Such n-type SPSLs can also be used as contact layers. Using a Ti/Al/Ti/Au stack annealed at 700 C, specific contact resistance of the order of 10 Wcm was obtained for –5 2 stack annealed at 700 °C, specific contact resistance of the order of 10 Ω·cm was obtained for AlN/AlGa(In)N SPSL with ~5.1 eV bandgap [92]. AlN/AlGa(In)N SPSL with ~5.1 eV bandgap [92]. Unfortunately, for III-Ns compounds and their alloys, there is only one convenient p-type dopant, Unfortunately, for III-Ns compounds and their alloys, there is only one convenient p-type Mg, the experimentally determined activation energy of which varies from ~120 to ~220 meV in dopant, Mg, the experimentally determined activation energy of which varies from ~120 to ~220 meV GaN [93,94], and reaches more than 500 meV in AlN [94,95]. A detailed analysis of the Mg activation in GaN [93,94], and reaches more than 500 meV in AlN [94,95]. A detailed analysis of the Mg energy in p-AlGaN epitaxial layers over the entire composition range was recently published [96]. activation energy in p-AlGaN epitaxial layers over the entire composition range was recently It should be noted that if Mg-doped AlGa(In)N is grown using MOCVD, then Mg activation is required. published [96]. It should be noted that if Mg-doped AlGa(In)N is grown using MOCVD, then Mg This can be done by both rapid thermal annealing at elevated temperatures [97,98] and holding the activation is required. This can be done by both rapid thermal annealing at elevated temperatures sample under an electron beam irradiation [98–101]. It was also shown that Mg–O co-doping reduces [97,98] and holding the sample under an electron beam irradiation [98–101]. It was also shown that acceptor activation energy in GaN [102,103], as in AlN [104]. However, this method of doping is not Mg–O co-doping reduces acceptor activation energy in GaN [102,103], as in AlN [104]. However, this widely used since oxygen can react with aluminum and gallium, forming undesirable oxides of these method of doping is not widely used since oxygen can react with aluminum and gallium, forming metals, especially when used in molecular beam epitaxy. undesirable oxides of these metals, especially when used in molecular beam epitaxy. The hole density at room temperature in Al Ga N:Mg alloys with Mg concentration at x 1x 19 20 3 19 The hole density at room temperature in AlxGa1−xN:Mg alloys with Mg concentration at ~10 – ~10 –10 cm is shown in Figure 6a. Note that all Mg concentrations were obtained using secondary 20 –3 10 cm is shown in Figure 6a. Note that all Mg concentrations were obtained using secondary ion ion mass spectrometry (SIMS) [25,103]. mass spectrometry (SIMS) [25,103]. A significant decrease in the hole concentration with an increase in the Al content in A significant decrease in the hole concentration with an increase in the Al content in AlGaN is AlGaN is consistent with an increase in the activation energy of Mg in the wide bandgap layers. consistent with an increase in the activation energy of Mg in the wide bandgap layers. For For AlN/Al Ga (In)N (0.03 < x < 0.08) SPSLs with 2–3 ML thick wells, and periods of the 1x AlN/AlxGa1−x(In)N (0.03 < x < 0.08) SPSLs with 2–3 ML thick wells, and periods of the order of 6–8 order of 6–8 MLs, the average AlN concentration in SPSLs can be changed in the range of y ave 19 3 MLs, the average AlN concentration in SPSLs can be changed in the range of yave = 0.5–0.8. The = 0.5–0.8. The average Mg concentration in these SPSLs is usually at the level of 10 cm . 19 −3 18 3 average Mg concentration in these SPSLs is usually at the level of 10 cm . The concentration of holes The concentration of holes can be at the level of 10 cm , even in SPSLs with high average AlN 18 −3 can be at the level of 10 cm , even in SPSLs with high average AlN content, as is seen in Figure 6a. content, as is seen in Figure 6a. Such structures were obtained using both MBE and MOCVD Such structures were obtained using both MBE and MOCVD methods [25,41,42,45,91,105–110]. methods [25,41,42,45,91,105–110]. Figure 6b shows the results of temperature-dependent Hall characterization of three SPSLs and Figure 6b shows the results of temperature-dependent Hall characterization of three SPSLs and 19 −3 19 −3 19 3 19 3 one AlGaN layer. The Mg concentration was ~10 cm in SPSLs, and ~3 × 10 cm in AlGaN. Note one AlGaN layer. The Mg concentration was ~10 cm in SPSLs, and ~3  10 cm in AlGaN. that the composition of the well was in the range 0.03 < x < 0.08 which is very similar to the Al concentration in the AlGaN thick layer. Average AlN concentrations of these SPSLs were yave = 0.65, Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 17 0.55, and 0.45. The average SPSL compositions, yave, were obtained using X-ray diffraction [61]. The hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in Appl. Sci. 2018, 8, 2362 7 of 17 SPSLs, and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 17 Mg-doped Al0.05Ga0.95N layer. Similar experimental results were reported by many different teams Note that the composition of the well was in the range 0.03 < x < 0.08 which is very similar to the 0.55, and 0.45. The average SPSL compositions, yave, were obtained using X-ray diffraction [61]. The [45,105–115]. Al concentration in the AlGaN thick layer. Average AlN concentrations of these SPSLs were y = ave hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. 0.65, 0.55, and 0.45. The average SPSL compositions, y , were obtained using X-ray diffraction [61]. ave Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to The hole density, here, corresponds to the hole concentration averaged over the full SPSL thickness. be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in Fitting lines 1, 2, and 3 were calculated using BESST simulator, assuming AlN barriers in the SPSL to SPSLs, and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly be relaxed. It is clearly seen that the hole concentration varies very little with the temperature in SPSLs, Mg-doped Al0.05Ga0.95N layer. Similar experimental results were reported by many different teams and is highly dependent on the temperature (almost 1.5 orders of magnitude) in a uniformly Mg-doped [45,105–115]. Al Ga N layer. Similar experimental results were reported by many different teams [45,105–115]. 0.05 0.95 (a) (b) Figure 6. Hole concentration in AlxGa1-xN alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on average Al concentration in epitaxial layers (black symbols) and in SPSLs (red symbols and green box); (b) The temperature dependence of the hole concentration in Al0.05Ga0.95N epitaxial layer (~0.5 μm thick, filled circles) and in AlN/AlxGa1−xN SPSLs (open symbols). Lines 1, 2, and 3 are the results of the applied BESST simulator [76]. (a) (b) The valence band edge profile in AlN(~0.90 nm)/Al0.03Ga0.97N(~0.52 nm) SPSL, computed by the Figure 6. Hole concentration in AlxGa1-xN alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on Figure 6. Hole concentration in Al Ga N alloys and AlN/AlGa(In)N SPSLs. (a) The dependence on 1x BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in averag average e Al Al con concentration centration in in epitaxial epitaxial lay layers ers (black (blacsymbols) k symbols) and and in SPSLs in SPSLs (red s (red symbols ymbol and s gr and een gree box); n this SPSL corresponds to the zero-energy horizontal line shown in Figure 7a. The acceptor levels in box); ( (b) The b) The t temperatur emperature dependence of th e dependence of the holee hole con concentration centration in Al in Al Ga 0.05Ga N0.95 epitaxial N epitax layer ial lay (~0.5 er (~0.5 m 0.05 0.95 Al0.03Ga0.97N-well and AlN-barrier layers are indicated by the dashed lines. μ thick, m thick filled , filcir led cles) circand les) and in AlN/Al in AlN/A Ga lxGa1 N−xN SPSLs SPSL(open s (open symbol symbols).sLines ). Lines 1, 1, 2, 2, and and 3 are the res 3 are the results ulof ts x 1x It is clearly seen that the acceptor levels in the quantum wells are above the Fermi level. Thus, the applied BESST simulator [76]. of the applied BESST simulator [76]. these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor The valence band edge profile in AlN(~0.90 nm)/Al Ga N(~0.52 nm) SPSL, computed by the 0.03 0.97 energy leve The valenc ls, yi e b eld aind ed ng a part ge pro ial act file in iv AlN(~0.90 nm)/A ation of acceptors in these l0.03Ga0.97N(~0 lay .5e2 rs. The d nm) SPSL, computed by the egree of activation BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in BESST simulator [76] at room temperature, are shown in Figure 7. The position of the Fermi level in depends on temperature, creating hole tunneling from barriers to wells, which results in the this SPSL corresponds to the zero-energy horizontal line shown in Figure 7a. The acceptor levels in exponential dependences o this SPSL corresponds to tf h hole concent e zero-energy horizont ration, as sho al lw ine shown n in Figure in Fig 6. ure 7a. The acceptor levels in Al Ga N-well and AlN-barrier layers are indicated by the dashed lines. Al0.03 0.03 Ga0.97 0.97 N-well and AlN-barrier layers are indicated by the dashed lines. It is clearly seen that the acceptor levels in the quantum wells are above the Fermi level. Thus, these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor energy levels, yielding a partial activation of acceptors in these layers. The degree of activation depends on temperature, creating hole tunneling from barriers to wells, which results in the exponential dependences of hole concentration, as shown in Figure 6. (a) (b) Figure 7. Valence band layout in AlN/Al0.03Ga0.97N SPSL, computed for room temperature (a) and Figure 7. Valence band layout in AlN/Al Ga N SPSL, computed for room temperature (a) and 0.03 0.97 band d band diagram iagram and heavy and heavy (H (HH), H), lig light ht (LH) (LH), , and sp and sp lit-off lit-of ( f S (SH) H) hole hole concen concentrations trations in thi in this s sam sample ple (b (). b ). The acceptor level posit The acceptor level positions ions were wereest estimated imated by negle by neglecting cting the effe the effect ct of ofthe hole the hole confin confinement ement in the in the qu quantum antum wells wells. . It is clearly seen that the (a) ( acceptor levels in the quantum wells arebabove ) the Fermi level. Thus, these acceptors should be not activated. In the AlN barriers, the Fermi level crosses the acceptor energy Figure 7. Valence band layout in AlN/Al0.03Ga0.97N SPSL, computed for room temperature (a) and levels, yielding a partial activation of acceptors in these layers. The degree of activation depends band diagram and heavy (HH), light (LH), and split-off (SH) hole concentrations in this sample (b). The acceptor level positions were estimated by neglecting the effect of the hole confinement in the quantum wells. Appl. Sci. 2018, 8, x FOR PEER REVIEW 8 of 17 Figure 7b shows the distributions of the heavy, light, and split-off holes in the same SPSL obtained by self-consistent solution of the Poisson and Schrödinger equations [76], accounting for the complex valence band structure of III-nitrides [116]. It is clearly seen that the total hole density in such an SPSL mostly depends on the heavy hole concentration. The contribution of the split-off holes can be neglected, due to the fact that this subband in the SPSL is far below the heavy and light subbands. Note, the holes in the SPSL are accumulated in the wells, partially penetrating into the AlN Appl. Sci. 2018, 8, 2362 8 of 17 barriers. One could conclude that the efficiency of Mg activation is a very weak function of temperature in wide bandgap AlN/AlxGa1−x(In)N SPSL (Eg from ~4.2 to ~5.3 eV) [107,108]. Thus, these SPSLs are on temperature, creating hole tunneling from barriers to wells, which results in the exponential very attractive for fabrication of transparent low resistive ohmic contacts for deep UV light emitters. dependences of hole concentration, as shown in Figure 6. Various contact metallization schemes have been applied for Mg-doped p-GaN, including Au, Figure 7b shows the distributions of the heavy, light, and split-off holes in the same SPSL obtained Ni, Ti, Pd, Pt, Au/Ni, Au/Pt, Au/Cr, Au/Pd, Au/Mg/Au, Au/Pt/Pd, Au/Cr/Ni, Au/Pt/Ni, Au/Ni/Pt, by self-consistent solution of the Poisson and Schrödinger equations [76], accounting for the complex Au-Zn/Ni, Si/Ni/Mg/Ni, etc. [117]. Oxidized Au/Ni is the most common ohmic contact to p-GaN, due valence band structure of III-nitrides [116]. It is clearly seen that the total hole density in such an to the relatively low specific contact resistance (ρc) and simplicity of fabrication [118]. It was shown SPSL mostly depends on the heavy hole concentration. The contribution of the split-off holes can be that formation of crystalline NiO and Ni3Ga4 play an essential role in the reduction of the specific neglected, due to the fact that this subband in the SPSL is far below the heavy and light subbands. contact resistance of p-GaN [118–121]. While the formation of NiO yields a reduction of the specific Note, the holes in the SPSL are accumulated in the wells, partially penetrating into the AlN barriers. contact resistance, the formation of Ni3Ga4 results in the eventual increase in the contact resistance One could conclude that the efficiency of Mg activation is a very weak function of temperature in −6 −6 2 [119–121]. The lowest specific contact resistance ρc ~9.2 × 10 and 2 × 10 Ω cm at 150 °C were wide bandgap AlN/Al Ga (In)N SPSL (E from ~4.2 to ~5.3 eV) [107,108]. Thus, these SPSLs are x 1x g achieved for p-type Mg-doped GaN and AlxGa1−xN layers [121,122]. very attractive for fabrication of transparent low resistive ohmic contacts for deep UV light emitters. Similar approach was successfully used for fabrication of ohmic contacts on p-type Various contact metallization schemes have been applied for Mg-doped p-GaN, including Au, Ni, AlN/AlxGa1−xN SPSLs [108]. Figure 8 shows the temperature dependence of the specific contact Ti, Pd, Pt, Au/Ni, Au/Pt, Au/Cr, Au/Pd, Au/Mg/Au, Au/Pt/Pd, Au/Cr/Ni, Au/Pt/Ni, Au/Ni/Pt, resistances, ρc, for three SPSLs with average AlN content of yave ~0.7 and for the ~300 nm thick Mg- Au-Zn/Ni, Si/Ni/Mg/Ni, etc. [117]. Oxidized Au/Ni is the most common ohmic contact to p-GaN, 19 −3 doped p-Al0.03Ga0.97N epitaxial layer. The Mg concentration in all samples was ~10 cm and room due to the relatively low specific contact resistance (r ) and simplicity of fabrication [118]. It was −5 2 temperature ρc was ~4 × 10 Ω·cm . shown that formation of crystalline NiO and Ni Ga play an essential role in the reduction of the 3 4 It was shown that the temperature dependence of the specific contact resistance is primarily specific contact resistance of p-GaN [118–121]. While the formation of NiO yields a reduction of the controlled by the activation energy of Mg acceptors in AlxGa1−xN [123]. The ρc of metal/SPSL ohmic specific contact resistance, the formation of Ni Ga results in the eventual increase in the contact 3 4 6 6 2 contact, as seen in Figure 8, also follows the temperature dependence of hole concentration in these resistance [119–121]. The lowest specific contact resistance r ~9.2  10 and 2  10 W cm at structure  s, which are very similar to that shown in Figure 6b, since the density of acceptors activated 150 C were achieved for p-type Mg-doped GaN and Al Ga N layers [121,122]. x 1x in AlN barriers is the primary factor influencing the hole concentration in the well, producing a Similar approach was successfully used for fabrication of ohmic contacts on p-type weaker temperature dependence compared to bulk materials. Thus, a weak dependence of ρc on AlN/Al Ga N SPSLs [108]. Figure 8 shows the temperature dependence of the specific contact x 1x temperature is expected, even when thermionic mechanism of current injection dominates in the resistances, r , for three SPSLs with average AlN content of y ~0.7 and for the ~300 nm thick c ave metal/SPSL ohmic contact. This expectation agrees with the results shown in Figure 8. 19 3 Mg-doped p-Al Ga N epitaxial layer. The Mg concentration in all samples was ~10 cm and 0.03 0.97 5 2 room temperature r was ~4  10 Wcm . Figure 8. Specific contact resistances of Mg-doped AlN/Al Ga N SPSLs (y ~0.7) and Al Ga N x 1x ave 0.03 0.97 Figure 8. Specific contact resistances of Mg-doped AlN/AlxGa1−xN SPSLs (yave ~0.7) and Al0.03Ga0.97N epitaxial layer at different temperatures. epitaxial layer at different temperatures. It was shown that the temperature dependence of the specific contact resistance is primarily 5. Deep UV Light Emitters Based on AlN/AlGa(In)N SPSLs controlled by the activation energy of Mg acceptors in Al Ga N [123]. The r of metal/SPSL ohmic x 1x c contact, as seen in Figure 8, also follows the temperature dependence of hole concentration in these The ability to change the bandgap in SPSL by several hundred meV, by changing the thickness structures, which are very similar to that shown in Figure 6b, since the density of acceptors activated in of the well by 1 ML, opened up a very simple path to the implementation of the DHS LEDs. Since AlN barriers is the primary factor influencing the hole concentration in the well, producing a weaker temperature dependence compared to bulk materials. Thus, a weak dependence of r on temperature is expected, even when thermionic mechanism of current injection dominates in the metal/SPSL ohmic contact. This expectation agrees with the results shown in Figure 8. Appl. Sci. 2018, 8, 2362 9 of 17 5. Deep UV Light Emitters Based on AlN/AlGa(In)N SPSLs Appl. Sci. 2018, 8, x FOR PEER REVIEW 9 of 17 The ability to change the bandgap in SPSL by several hundred meV, by changing the thickness of the well by 1 ML, opened up a very simple path to the implementation of the DHS LEDs. Since SPSL with particular barrier/well thicknesses has its effective bandgap, we can consider them to form SPSL with particular barrier/well thicknesses has its effective bandgap, we can consider them to form a heterostructure in the sense that an SPSL-based active layer with a smaller effective bandgap is a heterostructure in the sense that an SPSL-based active layer with a smaller effective bandgap is sandwiched between n- and p-type SPSL- based cladding layers with a larger bandgap. sandwiched between n- and p-type SPSL- based cladding layers with a larger bandgap. A schematic cross-section of a typical DHS for a sub-300 nm LED is shown in Figure 9. The A schematic cross-section of a typical DHS for a sub-300 nm LED is shown in Figure 9. The structure structure consists of six main parts: (1) thin, <100 nm, AlN buffer; (2) thick, ~200 nm, Al0.7–0.8Ga0.3−0.2N consists of six main parts: (1) thin, <100 nm, AlN buffer; (2) thick, ~200 nm, Al Ga N buffer 0.7–0.8 0.3–0.2 buffer for reduction of dislocation density in a device structure (this layer can be replaced by for reduction of dislocation density in a device structure (this layer can be replaced by AlN/GaN AlN/GaN SPSL—“dislocation absorber”); (3) n-type AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(2 ML) SPSL SPSL—“dislocation absorber”); (3) n-type AlN(5 ML)/Al Ga (In)N(2 ML) SPSL cladding 0.92–0.97 0.08–0.03 cladding layer of ~400 nm; (4) undoped AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(3 ML) SPSL active region layer of ~400 nm; (4) undoped AlN(5 ML)/Al Ga (In)N(3 ML) SPSL active region (5 pairs); 0.92–0.97 0.08–0.03 (5 pairs); (5) p-type AlN(5 ML)/Al0.92–0.97Ga0.08−0.03(In)N(2 ML) SPSL cladding layer of ~200 nm; (6) thin, (5) p-type AlN(5 ML)/Al Ga (In)N(2 ML) SPSL cladding layer of ~200 nm; (6) thin, ~2–5 nm, 0.92–0.97 0.08–0.03 ~2–5 nm, p-type contact layer with the composition of SPSL’s well material or pure GaN:Mg. As can p-type contact layer with the composition of SPSL’s well material or pure GaN:Mg. As can be seen be seen from the TEM cross-section of this device, shown in the same Figure 9, the barrier/well from the TEM cross-section of this device, shown in the same Figure 9, the barrier/well interfaces are interfaces are quite sharp, and the thickness of the well in the active region is greater than in the quite sharp, and the thickness of the well in the active region is greater than in the adjacent n- and adjacent n- and p-type barriers. p-type barriers. Figure 9. A not-in-scale schematic cross-section (on the left) of a double heterostructure (DHS) Figure 9. A not-in-scale schematic cross-section (on the left) of a double heterostructure (DHS) light- light-emitting diode (LED) and TEM cross-section (on the right) of the active region, and adjacent n- emitting diode (LED) and TEM cross-section (on the right) of the active region, and adjacent n- and and p-type emitters of this diode. p-type emitters of this diode. The current–voltage (I–V) characteristic of a typical 260 nm DHS mesa-LED is shown in Figure 10a. The current–voltage (I–V) characteristic of a typical 260 nm DHS mesa-LED is shown in Figure The LED turns on at ~6.0 V, and has a relatively high differential series resistance R of 115 5 W under ser 10a. The LED turns on at ~6.0 V, and has a relatively high differential series resistance Rser of 115 ± 5 forward bias from 8 V to 12 V. The R of mesa diodes is the sum of the contact (R ), spreading (R ), ser cont sp Ω under forward bias from 8 V to 12 V. The Rser of mesa diodes is the sum of the contact (Rcont), and vertical (R ) resistances. The contact resistance of a 160 m diameter diode, obtained from ver spreading (Rsp), and vertical (Rver) resistances. The contact resistance of a 160 μm diameter diode, specific contact resistance, is R ~90 W. The estimated spreading resistance of this LED is R ~20 W, cont sp obtained from specific contact resistance, is Rcont ~90 Ω. The estimated spreading resistance of this taking into account the resistivity of n-type AlGaN buffer layer (not shown in the inset of Figure 10a). LED is Rsp ~20 Ω, taking into account the resistivity of n-type AlGaN buffer layer (not shown in the The resistance of the etched part of the mesa, corresponding to transport across the SPSL layers, is inset of Figure 10a). The resistance of the etched part of the mesa, corresponding to transport across R  r  (h/A) where r is the perpendicular (along the growth direction) resistivity of the SPSL, ver 1 1 the SPSL layers, is Rver ≈ ρ1 × (h/A) where ρ1 is the perpendicular (along the growth direction) resistivity h is the height of the mesa (~300 nm, with the thickness of p-SL ~200 nm), and A is the contact area. of the SPSL, h is the height of the mesa (~300 nm, with the thickness of p-SL ~200 nm), and A is the Finally, we obtain R ~5 W, resulting in r ~50 W cm for a p-type SPSL. Comparing this to the in-plane ver 1 contact area. Finally, we obtain Rver ~5 Ω, resulting in ρ1 ~50 Ω cm for a p-type SPSL. Comparing this conductivity (r ) of the p-type SPSL obtained from Hall measurements, r ~4 W cm, we obtain the 2 2 to the in-plane conductivity (ρ2) of the p-type SPSL obtained from Hall measurements, ρ2 ~4 Ω cm, conductivity anisotropy r /r ~13. These simple considerations indicate relatively low r , considering 1 2 1 we obtain the conductivity anisotropy ρ1/ρ2 ~13. These simple considerations indicate relatively low the high AlN fraction in these SPSLs, and underscore the importance of reducing the contact resistance ρ1, considering the high AlN fraction in these SPSLs, and underscore the importance of reducing the to p-type materials. However, it is also important to lower R by optimization of the buffer layer sp contact resistance to p-type materials. However, it is also important to lower Rsp by optimization of thickness and its resistivity. Similar I–V characteristics were reported by different teams for LEDs the buffer layer thickness and its resistivity. Similar I–V characteristics were reported by different emitting near the same deep UV range of the wavelengths [42–46,91–128]. teams for LEDs emitting near the same deep UV range of the wavelengths [42–46,91–128]. The typical electroluminescence spectra of DHS LEDs based on AlN/AlGa(In)N SPSLs are shown The typical electroluminescence spectra of DHS LEDs based on AlN/AlGa(In)N SPSLs are in Figure 10b. All these spectra were obtained on LEDs with a similar geometry, and with the same shown in Figure 10b. All these spectra were obtained on LEDs with a similar geometry, and with the currents. A significant decrease in EQE is evident for devices emitting in ever deeper UV. The external same currents. A significant decrease in EQE is evident for devices emitting in ever deeper UV. The external quantum efficiency (EQE) of such simple LEDs does not exceed 0.1%. The absence of the electron blocking layer is the main drawback [129] of the structure shown in Figure 9. In order to increase the efficiency of such LED structures, it was proposed to insert a few monolayers-thick AlN electron blocking layer between the active layer and the p-emitter [35,36]. Appl. Sci. 2018, 8, 2362 10 of 17 quantum efficiency (EQE) of such simple LEDs does not exceed 0.1%. The absence of the electron blocking layer is the main drawback [129] of the structure shown in Figure 9. In order to increase Appl. Sci. 2018, 8, x FOR PEER REVIEW 10 of 17 the efficiency of such LED structures, it was proposed to insert a few monolayers-thick AlN electron blocking layer between the active layer and the p-emitter [35,36]. Summarizing, it should be noted that several new approaches to the use of SPSLs in combination Summarizing, it should be noted that several new approaches to the use of SPSLs in combination with QDs in the active layer of the LED have been proposed [60,130–132]. The most promising, from with QDs in the active layer of the LED have been proposed [60,130–132]. The most promising, my point of view, is the idea set forth in the ref. [60], where, experimentally, deep UV emission at 219 from my point of view, is the idea set forth in the ref. [60], where, experimentally, deep UV emission at nm from ultrathin MBE-grown GaN/AlN quantum structure was demonstrated. I believe in the 219 nm from ultrathin MBE-grown GaN/AlN quantum structure was demonstrated. I believe in the success of this approach, since we observed a similar effect almost 10 years ago [54,55,57,58]. success of this approach, since we observed a similar effect almost 10 years ago [54,55,57,58]. (a) (b) Figure 10. Figure 10. Electrical and opt Electrical and optical ical char characteristics acteristics of A of lN/A AlN/AlGa(In)N lGa(In)N SPSL-b SPSL-based ased DHS me DHS smesa-LEDs. a-LEDs. (a) Current–voltage characterist (a) Current–voltage characteristic ic of a typical LED of a typical LED operating at 262 nm wa operating at 262 nm wavelength. velength. Note Note that light that light em emission ission wa wass observed observed w with ith forward forward dc curr dc current above 2 mA; ( ent above 2 mA; (b) Electr b) oluminescence Electroluminescence spectra of spectra of similar similar AlN/AlGa(In)N AlN/AlGa(In)N SPSL- SPSL-based DHS based D mesa-LEDs HS mesa-LED operating s operat at 255, ing at 255, 280, 280, and 290 and 290 nm. nm. 6. Conclusions 6. Conclusions It was demonstrated that AlN/Al Ga N p- and n-type SPSLs with average AlN content up to x 1x It was demonstrated that AlN/AlxGa1−xN p- and n-type SPSLs with average AlN content up to 18 3 y ~0.7 and bandgap over 5.1 eV could have hole and electron concentrations exceeding 10 cm . ave 18 −3 yave ~0.7 and bandgap over 5.1 eV could have hole and electron concentrations exceeding 10 cm . 5 2 Low-resistance ohmic contacts with specific contact resistance below ~4  10 Wcm can be formed −5 2 Low-resistance ohmic contacts with specific contact resistance below ~4 × 10 Ω·cm can be formed on these SPSLs. The LEDs based on SPSLs with emission wavelengths from 290 to 232 nm were on these SPSLs. The LEDs based on SPSLs with emission wavelengths from 290 to 232 nm were demonstrated by different teams. demonstrated by different teams. Funding: This paper received no external funding. Funding: This paper received no external funding. Acknowledgments: I would like to acknowledge all my colleagues involved in this research. I thank S. Yu. Acknowledgments: I would like to acknowledge all my colleagues involved in this research. I thank S. Yu. Karpov for very helpful discussions. Karpov for very helpful discussions. Conflicts of Interest: The author declares no conflict of interest. Conflicts of Interest: The author declares no conflict of interest. References References 1. Alferov, Z.I. Nobel Lecture: The double heterostructure concept and its applications in physics, electronics, 1. Alferov, Z.I. Nobel Lecture: The double heterostructure concept and its applications in physics, electronics, and technology. Rev. Mod. Phys. 2001, 73, 767–782. [CrossRef] and technology. Rev. Mod. Phys. 2001, 73, 767–782, doi:10.1103/RevModPhys.73.767. 2. Kroemer, H. Nobel Lecture: Quasielectric fields and band offsets: Teaching electrons new tricks. 2. Kroemer, H. Nobel Lecture: Quasielectric fields and band offsets: Teaching electrons new tricks. Rev. Mod. Rev. Mod. Phys. 2001, 73, 783–793. [CrossRef] Phys. 2001, 73, 783–793, doi:10.1103/RevModPhys.73.883. 3. Esaki, L.; Tsu, R. Superlattice and negative differential conductivity in semiconductors. IBM J. Res. Dev. 1970, 3. Esaki, L.; Tsu, R. Superlattice and negative differential conductivity in semiconductors. IBM J. Res. Dev. 14, 61–65. [CrossRef] 1970, 14, 61–65, doi:10.1147/rd.141.0061. 4. Casey, H.C., Jr.; Panish, M.B. Heterostructure Lasers; Academic Press: New York, NY, USA, 1978; pp. 156–300. 4. Casey, H.C., Jr.; Panish, M.B. Heterostructure Lasers; Academic Press: New York, NY, USA, 1978; pp. 156– ISBN 0-12-163102-8. 300, ISBN 0-12-163102-8. 5. Ivanov, S.V.; Kopev, P.S.; Ledentson, N.N.; Meltser, B.Y.; Nadtochy, M.Y.; Ustinov, V.M.; Vasilev, A.M. 5. Ivanov, S.V.; Kopev, P.S.; Ledentson, N.N.; Meltser, B.Y.; Nadtochy, M.Y.; Ustinov, V.M.; Vasilev, A.M. Ultrathin layers and short-period superlattices in semiconductor structures. Phys. Stat. Sol. 1990, 118, Ultrathin layers and short-period superlattices in semiconductor structures. Phys. Stat. Sol. 1990, 118, 169– 169–177. [CrossRef] 177, doi:10.1002/pssa.2211180119. 6. Chang, M.-H.; Das, D.; Varde, P.V.; Pecht, M. Light emitting diodes reliability review. Microelectron. Reliab. 2012, 52, 762–782, doi:10.1016/j.microrel.2011.07.063. 7. Opel, D.R.; Hagstrom, E.; Pace, A.K.; Sisto, K.; Hirano-Ali, S.A.; Desai, S.; Swan, J. Light emitting diodes, a brief review and clinical experience. J. Clin. Aesthet. Dermatol. 2015, 8, 36–44, PMC4479368. Appl. Sci. 2018, 8, 2362 11 of 17 6. Chang, M.-H.; Das, D.; Varde, P.V.; Pecht, M. Light emitting diodes reliability review. Microelectron. Reliab. 2012, 52, 762–782. [CrossRef] 7. Opel, D.R.; Hagstrom, E.; Pace, A.K.; Sisto, K.; Hirano-Ali, S.A.; Desai, S.; Swan, J. Light emitting diodes, a brief review and clinical experience. J. Clin. Aesthet. Dermatol. 2015, 8, 36–44, PMCID:PMC4479368. [PubMed] 8. Yadav, A.; Titkov, I.E.; Sokolovskii, G.S.; Karpov, S.Y.; Dudelev, V.V.; Soboleva, K.K.; Strassburg, M.; Pietzonka, I.; Lugauer, H.-J.; Rafailov, E.U. AlGaInP red-emitting light emitting diode under extremely high pulsed pumping. Proc. SPIE 2016, 9768. [CrossRef] 9. Karpov, S.Y. Light-emitting diodes for solid-state lighting: Searching room for improvements. Proc. SPIE 2016, 9768. [CrossRef] 10. Gupta, S.D. (Ed.) Light Emitting Diodes for Agriculture; Springer: Singapore, 2017. [CrossRef] 11. Schubert, E.F. Light-Emitting Diodes, 3rd ed.; Rensselaer Polytechnic Institute Troy: New York, NY, USA, 2018; pp. 17-1–17-15, 19-6–19-16, 21-2–21-8. ISBN 978-0-9863826-6-6. 12. Pantha, B.N.; Dahal, R.; Nakarmi, M.L.; Nepal, N.; Li, J.; Lin, J.Y.; Jiang, H.X.; Paduano, Q.S.; Weyburne, D. Correlation between optoelectronic and structural properties and epilayer thickness of AlN. Appl. Phys. Lett. 2007, 90. [CrossRef] 13. Zhou, C.; Ghods, A.; Saravade, V.G.; Patel, P.V.; Yunghans, K.L.; Ferguson, C.; Feng, Y.; Kucukgok, B.; Lu, N.; Ferguson, I.T. Review—The current and emerging applications of the III-Nitrides. ECS J. Sol. Stat. Sci. Technol. 2017, 6, Q-149–Q-156. [CrossRef] 14. Davydov, V.Y.; Klochikhin, A.A.; Emtsev, V.V.; Kurdyukov, D.A.; Ivanov, S.V.; Vekshin, V.V.; Bechstedt, F.; Furthmüller, J.; Aderhold, J.; Graul, J.; et al. Band gap of hexagonal InN and InGaN alloys. Phys. Stat. Sol. 2002, 234, 787–795. [CrossRef] 15. Wu, J.; Walukiewicz, W.; Shan, W.; Yu, K.M.; Ager, J.W., III; Li, S.X.; Haller, E.E.; Lu, H.; Schaff, W.J. Temperature dependence of the fundamental band gap of InN. J. Appl. Phys. Lett. 2003, 94, 4457–4460. [CrossRef] 16. Andreev, B.A.; Kudryavtsev, K.E.; Yablonskiy, A.N.; Lobanov, D.N.; Bu shuykin, P.A.; Krasilnikova, L.V.; Skorokhodov, E.V.; Yunin, P.A.; Novikov, A.V.; Davydov, V.Y.; Krasilnik, Z.F. Towards the indium nitride laser: Obtaining infrared stimulated emission from planar monocrystalline InN structures. Sci. Rep. 2018, 8. [CrossRef] [PubMed] 17. Akasaki, I. Nobel Lecture: Fascinated journeys into blue light. Rev. Mod. Phys. 2015, 87, 1119–1131. [CrossRef] 18. Amano, H. Nobel Lecture: Growth of GaN on sapphire via low-temperature deposited buffer layer and realization of p-type GaN by Mg doping followed by low-energy electron beam irradiation. Rev. Mod. Phys. 2015, 87, 1133–1138. [CrossRef] 19. Nakamura, S. Nobel Lecture: Background story of the invention of efficient blue InGaN light emitting diodes. Rev. Mod. Phys. 2015, 87, 1139–1151. [CrossRef] 20. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Holtz, M.; Nikishin, S.; Temkin, H. AlGaInN light-emitting diodes grown on Si (111). Appl. Phys. Lett. 2002, 80, 3682–3684. [CrossRef] 21. Sizov, V.S.; Tsatsulnikov, A.F.; Sakharov, A.V.; Lundin, W.V.; Zavarin, E.E.; Cherkashin, N.A.; Hÿtch, M.J.; Nikolaev, A.E.; Mintairov, A.M.; He, Y.; et al. The use of short_period InGaN/GaN superlattices in blue-region light-emitting diodes. Semiconductors 2010, 44, 924–930. [CrossRef] 22. Lähnemann, J.; Ajay, A.; Den Hertog, M.I.; Monroy, E. Near-infrared intersubband photodetection in GaN/AlN nanowires. Nano Lett. 2017, 17, 6954–6960. [CrossRef] [PubMed] 23. Mehta, K.; Liu, Y.-S.; Wang, J.; Jeong, H.; Detchprohm, T.; Park, Y.J.; Alugubelli, S.R.; Wang, S.; Ponce, F.A.; Shen, S.-C.; et al. Lateral current spreading in III-N ultraviolet vertical-cavity surface-emitting lasers using modulation-doped short period superlattices. IEEE J. Quant. Electron. 2018, 54. [CrossRef] 24. Nikishin, S.A.; Faleev, N.N.; Antipov, V.G.; Francoeur, S.; Grave de Peralta, L.; Seryogin, G.A.; Temkin, H.; Prokofyeva, T.I.; Holtz, H.; Chu, S.N.G. High quality GaN grown on Si(111) by gas source molecular beam epitaxy with ammonia. Appl. Phys. Lett. 1999, 75, 2073–2075. [CrossRef] 25. Nikishin, S.; Kipshidze, G.; Kuryatkov, V.; Zubrilov, A.; Choi, K.; Gherasoiu, Iu.; Grave de Peralta, L.; Prokofyeva, T.; Holtz, M.; Asomoza, R.; et al. Gas source molecular beam epitaxy of high quality AlGaN on Si and sapphire. Mat. Res. Soc. Symp. 2001, 639. [CrossRef] Appl. Sci. 2018, 8, 2362 12 of 17 26. Xiao, M.; Zhang, J.; Hao, Y. High-quality Ga-rich AlGaN grown on trapezoidal patterned GaN template using super-short period AlN/GaN superlattices for rapid coalescence. J. Cryst. Growth 2017, 464, 153–158. [CrossRef] 27. Aumer, M.E.; LeBoeuf, S.F.; Bedairb, S.M.; Smith, M.; Lin, J.Y.; Jiang, H.X. Effects of tensile and compressive strain on the luminescence properties of AlInGaN/InGaN quantum well structures. Appl. Phys. Lett. 2000, 77, 821–823. [CrossRef] 28. Son, J.H.; Lee, J.-L. Strain engineering for the solution of efficiency droop in InGaN/GaN light-emitting diodes. Opt. Express 2010, 18, 5466–5471. [CrossRef] [PubMed] 29. Nakamura, S.; Fasol, G. The Blue Laser Diode; Springer: Berglin/Heidelberg, Germany; New York, NY, USA, 1997; pp. 147–300. ISBN 978-3-662-03464-4 or 978-3-662-03462-0. 30. Morkoç, H. Handbook of Nitride Semiconductors and Devices. V.3: GaN-Based Optical and Electronic Devices; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2009; pp. 4–334. ISBN 978-3-527-40839-9. 31. Kneissl, M.; Rass, J. III-Nitride Ultraviolet Emitters Technology and Applications; Springer Series in Materials Science; Springer: Berglin/Heidelberg, Germany, 2016; Volume 227, pp. 3–12, 90–104, ISBN 978-3-319-24098-5. 32. Amano, H. Progress and prospect of growth of wide-band-gap group III Nitrides. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 1–9. ISBN 978-981-10-3754-2. 33. Egawa, T.; Oda, O. LEDs based on hetero epitaxial GaN on Si substrates. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 1–9. ISBN 978-981-10-3754-2. 34. Tadatomo, K.; Okagawa, H.; Ohuchi, Y.; Tsunekawa, T.; Kudo, H.; Sudo, Y.; Kato, M.; Taguchi, T. High output power near-ultraviolet and violet light-emitting diodes fabricated on patterned sapphire substrates using metalorganic vapor phase epitaxy. J. Light Vis. Env. 2003, 27, 140–145. [CrossRef] 35. Kneissl, M.; Rass, J.; Schade, L.; Schwarz, U.T. Growth and Optical Properties of GaN-Based Non- and Semipolar LEDs. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 93–128. ISBN 978-981-10-3754-2. 36. Hirayama, H.; Kamata, N.; Tsubaki, K. AlGaN-Based Deep-Ultraviolet Light-Emitting Diodes. In III-Nitride Based Light Emitting Diodes and Applications, 2nd ed.; Seong, T.-Y., Han, J., Amano, H., Morkoç, H., Eds.; Springer: Singapore, 2017; pp. 93–128. ISBN 978-981-10-3754-2. 37. Staszczak, G.; Gorczyca, I.; Grzank, E.; Smalc-Koziorowska, J.; Targowski, G.; Czernecki, R.; Siekacz, M.; Grzanka, S.; Skierbiszewski, C.; Schulz, T.; et al. Bandgap behavior of InGaN/GaN short period superlattices grown by metal-organic vapor phase epitaxy. Phys. Stat. Sol. 2017, 254, 1600710. [CrossRef] 38. Dimitrakopulos, G.P.; Vasileiadis, I.G.; Bazioti, C.; Smalc-Koziorowska, J.; Kret, S.; Dimakis, E.; Florini, N.; Kehagias, T.; Suski, T.; Karakostas, T.; et al. Compositional and strain analysis of In(Ga)N/GaN short period superlattices. J. Appl. Phys. 2018, 123, 024304. [CrossRef] 39. Gorczyca, I.; Suski, T.; Christensen, N.E.; Svane, A. Theoretical study of nitride short period superlattices. J. Phys. Condens. Matter 2018, 30, 063001. [CrossRef] [PubMed] 40. Lundin, W.V.; Nikolaev, A.E.; Sakharov, A.V.; Zavarin, E.E.; Valkovskiy, G.A.; Yagovkina, M.A.; Usov, S.O.; Kryzhanovskaya, N.V.; Sizov, V.S.; Brunkov, P.N.; et al. Single quantum well deep-green LEDs with buried InGaN/GaN short-period superlattice. J. Cryst. Growth 2011, 315, 267–271. [CrossRef] 41. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Nikishin, S.; Holtz, M.; Chu, S.N.G.; Temkin, H. Deep Ultraviolet AlGaInN-Based Light-Emitting Diodes on Si(111) and Sapphire. Phys. Stat. Sol. 2002, 192, 286–291. [CrossRef] 42. Kipshidze, G.; Kuryatkov, V.; Zhu, K.; Borisov, B.; Holtz, M.; Nikishin, S.; Temkin, H. AlN/AlGaInN superlattice light emitting diodes at 280 nm. J. Appl. Phys. 2003, 93, 1363–1366. [CrossRef] 43. Jmerik, V.N.; Lutsenko, E.V.; Ivanov, S.V. Plasma-assisted molecular beam epitaxy of AlGaN heterostructures for deep-ultraviolet optically pumped lasers. Phys. Stat. Sol. 2013, 210, 439–450. [CrossRef] 44. Shatalov, M.; Sun, W.; Jain, R.; Lunev, A.; Hu, X.; Dobrinsky, A.; Bilenko, Yu.; Yang, J.; Garrett, G.A.; Rodak, L.E.; et al. High power AlGaN ultraviolet light emitters. Semicond. Sci. Technol. 2014, 29, 084007. [CrossRef] 45. Nikishin, S.A.; Holtz, M.; Temkin, H. Digital alloys AlN/AlGaN for deep UV light emitting diodes. Jpn. J. Appl. Phys. 2005, 44, 7221–7226. [CrossRef] Appl. Sci. 2018, 8, 2362 13 of 17 46. Sun, W.; Tan, C.-K.; Tansu, N. AlN/GaN Digital alloy for mid- and deep-ultraviolet optoelectronics. Sci. Rep. 2017, 7, 11826–11833. [CrossRef] [PubMed] 47. Gorczyca, I.; Suski, T.; Strak, P.; Staszczak, G.; Christensen, N.E. Band gap engineering of In(Ga)N/GaN short period superlattices. Sci. Rep. 2017, 7, 16055. [CrossRef] [PubMed] 48. Kaneko, M.; Ueta, S.; Horita, M.; Kimoto, T.; Suda, J. Deep-ultraviolet light emission from 4H-AlN/4H-GaN short-period superlattice grown on 4H-SiC (11–20). Appl. Phys. Lett. 2018, 112, 012106. [CrossRef] 49. Neave, J.H.; Joyce, B.A.; Dobson, P.J.; Norton, N. Dynamics of films growth of GaAs by MBE from RHEED observations. Appl. Phys. A 1983, 31, 1–8. [CrossRef] 50. Yang, Z.; Li, L.K.; Wang, W.I. High-quality GaN and AlN grown by gas-source molecular beam epitaxy using ammonia as the nitrogen source. J. Vac. Sci. Technol. B 1996, 14, 2354–2356. [CrossRef] 51. Grandjean, N.; Massies, J. GaN and Al Ga N molecular beam epitaxy monitored by reflection high-energy x 1x electron diffraction. Appl. Phys. Lett. 1997, 71, 1816–1818. [CrossRef] 52. Daudin, B.; Widmann, F. Layer-by-layer grown of AlN and GaN by molecular beam epitaxy. J. Cryst. Growth 1997, 182, 1–5. [CrossRef] 53. Nikishin, S.A.; Borisov, B.A.; Chandolu, A.; Kuryatkov, V.V.; Temkin, H.; Holtz, M.; Mokhov, E.N.; Makarov, Y.; Helava, H. Short period superlattices of AlN/Al Ga N grown on AlN substrates. Appl. Phys. Lett. 2004, 0.08 0.92 85, 4355–4357. [CrossRef] 54. Borisov, B.; Nikishin, S.; Kuryatkov, V.; Temkin, H. Enhanced deep ultraviolet luminescence from AlGaN quantum wells grown in the three dimensional mode. Appl. Phys. Lett. 2005, 87, 191902. [CrossRef] 55. Borisov, B.A.; Nikishin, S.A.; Kuryatkov, V.V.; Kuchinski, V.I.; Holtz, M.; Temkin, H. Enhanced radiative recombination in AlGaN quantum wells grown by molecular-beam epitaxy. Semiconductors 2006, 40, 454–458. [CrossRef] 56. Panish, M.B.; Temkin, H. Gas Source Molecular Beam Epitaxy. Growth and Properties of Phosphorus Containing III–V Heterostructures; Springer Series in Materials Science; Springer: Berlin/Heidelberg, Germany, 1993; Volume 26, pp. 115–230. ISBN 3-540-56540-X (Berlin) or 0-387-56540-X (New York). 57. Nikishin, S.A.; Borisov, B.A.; Garrett, G.A.; Sarney, W.L.; Sampath, A.V.; Shen, H.; Wraback, M.; Holtz, M. Enhanced luminescence from Al Ga N/Al Ga N quantum wells grown by gas source molecular beam x 1x y 1y epitaxy with ammonia. Proc. SPIE 2007, 6473, 647306. 58. Nikishin, S.A.; Borisov, B.A.; Kuryatkov, V.V.; Holtz, M.; Garrett, G.A.; Sarney, W.L.; Sampath, A.V.; Shen, H.; Wraback, M. Correlations between the growth modes and luminescence properties of AlGaN quantum structures. Jpn. J. Appl. Phys. 2008, 47, 1556–1558. [CrossRef] 59. Islam, S.M.; Lee, K.; Verma, J.; Protasenko, V.; Rouvimov, S.; Bharadwaj, S.; Xing, H.G.; Jena, D. MBE-grown 232–270 nm deep-UV LEDs using monolayer thin binary GaN/AlN quantum heterostructures. Appl. Phys. Lett. 2017, 110, 041108. [CrossRef] 60. Islam, S.M.; Protasenko, V.; Lee, K.; Rouvimov, S.; Verma, J.; Xing, H.G.; Jena, D. Deep-UV emission at 219 nm from ultrathin MBE GaN/AlN quantum heterostructures. Appl. Phys. Lett. 2017, 111, 091104. [CrossRef] 61. Chandolu, A.; Nikishin, S.; Holtz, M.; Temkin, H. X-ray diffraction study of AlN/AlGaN short-period superlattices. J. Appl. Phys. 2007, 102, 114909. [CrossRef] 62. Kandaswamy, P.K.; Guillot, F.; Bellet-Amalric, E.; Monroy, E.; Nevou, L.; Tchernycheva, M.; Michon, A.; Julien, F.H.; Baumann, E.; Giorgetta, F.R.; et al. GaN/AlN short-period superlattices for intersubband optoelectronics: A systematic study of their epitaxial growth, design, and performance. J. Appl. Phys. 2008, 104, 093501. [CrossRef] 63. Kladko, V.; Kuchuk, A.; Lytvyn, P.; Yefanov, O.; Safriuk, N.; Belyaev, A.; Mazur, Y.I.; DeCuir, E.A., Jr.; Ware, M.E.; Salamo, G.J. Substrate effects on the strain relaxation in GaN/AlN short-period superlattices. Nanoscale Res. Lett. 2012, 7, 289. [CrossRef] [PubMed] 64. Noh, Y.-K.; Seo, J.-H.; Choi, H.-S.; Kim, M.-D.; Oh, J.-E. Thickness dependence of temperature-induced emission mechanism in InGaN/AlGaN short-period superlattices. J. Appl. Phys. 2012, 112, 043102. [CrossRef] 65. Kutt, R.N.; Shcheglov, M.P.; Ratnikov, V.V.; Yagovkina, M.A.; Davydov, V.Y.; Smirnov, A.N.; Rozhavskaya, M.M.; Zavarin, E.E.; Lundin, V.V. X-ray diffraction study of short-period AlN/GaN superlattices. Crystallogr. Rep. 2013, 58, 953–958. [CrossRef] 66. Pinquier, C.; Frandon, J.; Demangeot, F.; Smirnov, M.; Sarua, A.; Kuball, M.; Monroy, E.; Daudin, B. Behavior of phonons in short-period GaN-AlN superlattices. Phys. Stat. Sol. 2004, 1, 2706–2710. [CrossRef] Appl. Sci. 2018, 8, 2362 14 of 17 67. Davydov, V.Yu.; Roginskii, E.M.; Smirnov, A.N.; Kitaev, Y.E.; Yagovkina, M.A.; Kyutt, R.N.; Rozhavskaya, M.M.; Zavarin, E.E.; Lundin, W.V.; Smirnov, M.B. Lattice dynamics of short-period AlN/GaN short period superlattices: Theory and experiments. Phys. Stat. Sol. 2013, 210, 484–487. [CrossRef] 68. Pankin, D.V.; Smirnov, M.B.; Davydov, V.Y.; Smirnov, A.N.; Zavarin, E.E.; Lundin, W.V. Elastic Strains and Delocalized Optical Phonons in AlN/GaN Superlattices. Semiconductors 2016, 50, 1043–1048. [CrossRef] 69. Sarkar, K.; Datta, D.; Gosztola, D.J.; Shi, F.; Nicholls, A.; Stroscio, M.A.; Dutta, M. Raman analysis of phonon modes in a short period AlN/GaN superlattice. Superlattices Microstruct. 2018, 115, 116–122. [CrossRef] 70. Stolyarchuk, N.; Markurt, T.; Courville, A.; March, K.; Tottereau, O.; Vennéguès, P.; Albrecht, M. Impact of sapphire nitridation on formation of Al-polar inversion domains in N-polar AlN epitaxial layers. J. Appl. Phys. 2017, 122, 155303. [CrossRef] 71. Schuck, P.J.; Mason, M.D.; Grober, R.D.; Ambacher, O.; Lima, A.P.; Miskys, C.; Dimitrov, R.; Stutzmann, M. Spatially resolved photoluminescence of inversion domain boundaries in GaN-based lateral polarity heterostructures. Appl. Phys. Lett. 2001, 79, 952–954. [CrossRef] 72. Shubina, T.V.; Jmerik, V.N.; Tkachman, M.G.; Vekshin, V.A.; Ratnikov, V.V.; Toropov, A.A.; Sitnikova, A.A.; Ivanov, S.V.; Bergman, J.P.; Karlsson, F.; et al. Nanometric-scale fluctuations of intrinsic electric fields in GaN/AlGaN quantum wells with inversion domains. Phys. Stat. Sol. 2002, 234, 919–923. [CrossRef] 73. SiLENSe—Software Tool for Light Emitting Diode (LED) Bandgap Engineering. Available online: https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.str-soft. com%2Fproducts%2FSiLENSe%2Findex.htm&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=YpequeQnoIjnZ5cNklA7U5Zts7E%2FGL0Jz%2F1oAfZEdX4%3D& reserved=0 (accessed on 22 November 2018). 74. Crosslight Software Inc. Available online: https://na01.safelinks.protection.outlook.com/ ?url=https%3A%2F%2Fcrosslight.com%2F&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=vCC9BOBAXHLNXsh6KhW1CvooXSMJOC2KdKHyv77BAjI%3D& reserved=0 (accessed on 22 November 2018). 75. Synopsys, Inc. Available online: https://na01.safelinks.protection.outlook.com/?url= https%3A%2F%2Fwww.synopsys.com%2F&data=02%7C01%7CSergey.A.Nikishin%40ttu.edu% 7C1b70d2ef115047f2580008d622bf3be0%7C178a51bf8b2049ffb65556245d5c173c%7C0%7C0% 7C636734602099986989&sdata=33YOp0jlN06RqaUg2AIjE6oOB2OH%2Bi%2BNnzafC3Y2lgM%3D& reserved=0 (accessed on 22 November 2018). 76. Bandgap Engineering Superlattice Simulation Tool (BESST)—Package for Simulation of Optoelectronic Devices Based on Group-III Nitride superlattices. Available online: http://www.str-soft.com/products/ BESST/index.htm (accessed on 22 November 2018). 77. Kozodoy, P.; Hansen, M.; DenBaars, S.P.; Mishra, U.K. Enhanced Mg doping efficiency in Al Ga N/GaN 0.2 0.8 superlattices. Appl. Phys. Lett. 1999, 74, 3681–3683. [CrossRef] 78. Bulashevich, K.A.; Karpov, S.Y. Heterojunctions between group-III nitride short period superlattices. Phys. Stat. Sol. 2005, 2, 2394–2398. [CrossRef] 79. Holtz, M.; Prokofyeva, T.; Seon, M.; Copeland, K.; Vanbuskirk, J.; Williams, S.; Nikishin, S.; Tretyakov, V.; Temkin, H. Composition dependence of the optical phonon energies in hexagonal Al Ga N. J. Appl. Phys. 1x 2001, 89, 7977–7982. [CrossRef] 80. Kuryatkov, V.; Chandolu, A.; Borisov, B.; Kipshidze, G.; Zhu, K.; Nikishin, S.; Temkin, H.; Holtz, M. Solar-blind ultraviolet photodetectors based on superlattices of AlN/AlGa(In)N. Appl. Phys. Lett. 2003, 82, 1323–1325. [CrossRef] 81. Herzog, J.B.; Mintairov, A.M.; Sun, K.; Cao, Y.; Jena, D.; Merz, J.L. Infrared reflectivity spectroscopy of optical phonons in short-period AlGaN/GaN superlattices. Proc. SPIE 2007, 6648, 664803. 82. Marx, G.; Engelbrecht, J.A.A.; Lee, M.E.; Wagener, M.C.; Henry, A. Comments on a peak of Al Ga N x 1x observed by infrared reflectance. Infrared Phys. Technol. 2016, 76, 493–499. [CrossRef] 83. Pendlebury, S.T.; Lynam, P.; Mowbray, D.J.; Parbrook, P.J.; Wood, D.A.; Lada, M.; O’Neill, J.P.; Cullis, A.G.; Skolnick, M.S. Optical characterization of AlGaN epitaxial layers and GaN/AlGaN quantum wells. Phys. Stat. Sol. 2001, 188, 871–875. [CrossRef] Appl. Sci. 2018, 8, 2362 15 of 17 84. Adelmann, C.; Sarigiannidou, E.; Jalabert, D.; Hori, Y.; Rouvière, J.-L.; Daudin, B.; Fanget, S.; Bru-Chevallier, C.; Shibata, T.; Tanaka, M. Growth and optical properties of GaN/AlN quantum wells. Appl. Phys. Lett. 2003, 82, 4154–4156. [CrossRef] 85. Paskov, P.; Monemar, B.; Kamyiama, S.; Amano, H.; Akasaki, I. Optical Properties of undoped, n-doped and p-doped GaN/AlN superlattices. Mater. Res. Soc. Symp. Proc. 2007, 955, 0955-I12-03. [CrossRef] 86. Da-Bing, L.; Wei-Guo, H.; Hideto, M.; Kazumasa, H.; Hang, S. Enhanced deep ultraviolet emission from Si-doped Al Ga N/AlN MQWs. Chin. Phys. B 2010, 19, 127891. [CrossRef] 1x 87. Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Nikishin, S.; Holtz, M.; Temkin, H. Si-doped Al Ga N (0.56  x  1) layers grown by molecular beam epitaxy with ammonia. Appl. Phys. Lett. 2005, x 1x 87, 132106–132108. [CrossRef] 88. Taniyasu, Y.; Kasu, M.; Toshiki Makimoto, T. Electrical conduction properties of n-type Si-doped AlN with 2 1 1 high electron mobility (>100 cm V s ). Appl. Phys. Lett. 2004, 85, 4672–4674. [CrossRef] 89. Contreras, S.; Konczewicz, L.; Messaoud, J.B.; Peyre, H.; Khalfioui, M.A.; Matta, S.; Leroux, M.; Damilano, B.; Brault, J. High temperature electrical transport study of Si-doped AlN. Superlattices Microstruct. 2016, 98, 253–258. [CrossRef] 90. Pampili, P.; Parbrook, P.J. Doping on III-nitride materials. Mat. Sci. Semicond. Proc. 2017, 62, 180–191. [CrossRef] 91. Nikishin, S.A.; Kuryatkov, V.V.; Chandolu, A.; Borisov, B.A.; Kipshidze, G.D.; Ahmad, I.; Holtz, M.; Temkin, H. Deep ultraviolet light emitting diodes based on short period superlattices of AlN/AlGa(In)N. Jpn. J. Appl. Phys. 2003, 42, L1362–L1365. [CrossRef] 92. Yun, J.; Choi, K.; Mathur, K.; Kuryatkov, V.; Borisov, B.; Kipshidze, G.; Nikishin, S.; Temkin, H. Low resistance ohmic contacts to digital alloys of n-AlGaN/AlN. IEEE Electron Device Lett. 2006, 27, 22–24. [CrossRef] 93. Kozodoy, P.; Xing, H.; DenBaars, S.P.; Mishra, U.K.; Saxler, A.; Perrin, R.; Elhamri, S.; Mitchel, W.C. Heavy doping effects in Mg-doped GaN. J. Appl. Phys. 2000, 87, 1832–1835. [CrossRef] 94. Jiang, X.-H.; Shi, J.-J.; Zhang, M.; Zhong, H.-X.; Huang, P.; Ding, Y.-M.; He, Y.-P.; Cao, X. Reduction of the Mg acceptor activation energy in GaN, AlN, Al Ga N and Mg d-doping (AlN) /(GaN) : The strain effect. 0.83 0.17 Ga 5 1 J. Phys. D Appl. Phys. 2015, 48, 475104. [CrossRef] 95. Nam, K.B.; Nakarmi, M.L.; Li, J.; Lin, J.Y.; Jiang, H.X. Mg acceptor level in AlN probed by deep ultraviolet photoluminescence. Appl. Phys. Lett. 2003, 83, 878–880. [CrossRef] 96. Liang, Y.-H.; Towe, E. Progress in efficient doping of high aluminum-containing group III-nitrides. Appl. Phys. Rev. 2018, 5, 011107. [CrossRef] 97. Ilegems, M.; Dingle, R. Luminescence of Mg- and Be- doped GaN. J. Appl. Phys. 1973, 44, 4234–4235. [CrossRef] 98. Gelhausen, O.; Phillips, M.R.; Goldys, E.M.; Paskova, T.; Monemar, B.; Strassburg, M.; Hoffmann, A. Dissociation of H-related defect complexes in Mg-doped GaN. Phys. Rev. B 2004, 69, 125210. [CrossRef] 99. Amano, H.; Kito, M.; Hiramatsu, K.; Akasaki, I. P-type conduction in Mg-doped GaN treated with low-energy electron beam irradiation (LEEBI). Jpn. J. Appl. Phys. 1989, 28, L2112–L2114. [CrossRef] 100. Chernyak, L.; Osinsky, A.; Fuflyigin, V.; Schubert, E.F. Electron beam-induced increase of electron diffusion length in -type GaN and AlGaN/GaN superlattices. Appl. Phys. Lett. 2000, 77, 875–877. [CrossRef] 101. Phillips, M.R.; Merklein, M.; Hardy, T.; Manning, T.J.; Wintrebert-Fouquet, M. A mechanism for Mg acceptor activation in GaN by low energy electron beam irradiation. Microsc. Microanal. 2010, 16, 816–817. [CrossRef] 102. Korotkov, R.Y.; Gregie, J.M.; Wessels, B.W. Electrical properties of p-type GaN:Mg codoped with oxygen. Appl. Phys. Lett. 2001, 78, 222–224. [CrossRef] 103. Kipshidze, G.; Kuryatkov, V.; Borisov, B.; Kudryavtsev, Yu.; Asomoza, R.; Nikishin, S.; Temkin, H. Mg and O codoping in p-type GaN and Al Ga N (0 < x < 0.08). Appl. Phys. Lett. 2002, 80, 2910–2912. [CrossRef] x 1x 104. Wu, R.Q.; Shen, L.; Yang, M.; Sha, Z.D.; Cai, Y.Q.; Feng, Y.P.; Huang, Z.G.; Wu, Q.Y. Enhancing hole concentration in AlN by Mg:O codoping: Ab initio study. Phys. Rev. B 2008, 77, 073203. [CrossRef] 105. Khan, M.A.; Adviarahan, V.; Zhang, J.P.; Chen, C.; Kuokstis, E.; Chitnis, A.; Shatalov, M.; Yang, J.W.; Simin, G. Stripe geometry ultraviolet light emitting diodes at 305 nanometers using quaternary AlInGaN multiple quantum wells. Jpn. J. Appl. Phys. 2001, 40, L1308–L1310. [CrossRef] 106. Lopatiuk-Tirpak, O.; Chernyak, L.; Borisov, B.A.; Kuryatkov, V.V.; Nikishin, S.A.; Gartsman, K. Electron irradiation-induced increase of minority carrier diffusion length, mobility, and lifetime in Mg-doped AlN/AlGaN short period superlattice. Appl. Phys. Lett. 2007, 91, 182103. [CrossRef] Appl. Sci. 2018, 8, 2362 16 of 17 107. Allerman, A.A.; Crawford, M.H.; Miller, M.A.; Lee, S.R. Growth and characterization of Mg-doped AlGaN-AlN short-period superlattices for deep-UV optoelectronic devices. J. Cryst. Growth 2010, 312, 756–761. [CrossRef] 108. Nikishin, S.; Borisov, B.; Mansurov, V.; Pandikunta, M.; Chary, I.; Rajanna, G.; Bernussi, A.; Kudryavtsev, Y.; Asomoza, R.; Bulashevich, K.A.; et al. Short period p-type AlN/AlGaN superlattices for deep UV light emitters. Mater. Res. Soc. Symp. Proc. 2010, 1202, 1202-I10-03. [CrossRef] 109. Shatalov, M.; Gaska, R.; Yang, J.; Shur, M. Method of Doping and Semiconductor Devices. U.S. Patent 8,426,225, 23 April 2013. 110. Martens, M.; Kuhn, C.; Ziffer, E.; Simoneit, T.; Kueller, V.; Knauer, A.; Rass, J.; Wernicke, T.; Einfeldt, S.; Weyers, M.; et al. Low absorption loss p-AlGaN superlattice cladding layer for current-injection deep ultraviolet laser diodes. Appl. Phys. Lett. 2016, 108, 151108. [CrossRef] 111. Suzuki, M.; Nishio, J.; Onomura, M.; Hongo, C. Doping characteristics and electrical properties of Mg-doped AlGaN grown by atmospheric-pressure MOCVD. J. Cryst. Growth 1998, 189/190, 511–515. [CrossRef] 112. Li, J.; Oder, T.N.; Nakarmi, M.L.; Lin, J.Y.; Jiang, H.X. Optical and electrical properties of Mg-doped p-type Al Ga N. Appl. Phys. Lett. 2002, 80, 1210–1212. [CrossRef] x 1x 113. Taniyasu, Y.; Kasu, M.; Makimoto, T. An aluminium nitride light-emitting diode with a wavelength of 210 nanometres. Nature 2006, 441, 325–328. [CrossRef] [PubMed] 114. Kinoshita, T.; Obata, T.; Yanagi, H.; Inoue, S.-I. High p-type conduction in high-Al content Mg-doped AlGaN. Appl. Phys. Lett. 2013, 102, 012105. [CrossRef] 115. Dussaigne, A.; Damilano, B.; Brault, J.; Massies, J.; Feltin, E.; Grandjean, N. High doping level in Mg-doped GaN layers grown at low temperature. J. Appl. Phys. 2008, 103, 013110. [CrossRef] 116. Chuang, S.L.; Chang, C.S. kp method for strained wurtzite semiconductors. Phys. Rev. B 1996, 54, 2491–2504. [CrossRef] 117. Blank, T.V.; Gol’dberg, Y.A. Mechanisms of current flow in metal-semiconductor ohmic contacts. Semiconductors 2007, 41, 1263–1292. [CrossRef] 118. Ho, J.-K.; Jong, C.-S.; Chiu, C.C.; Huang, C.-N.; Shih, K.-K.; Chen, L.-C.; Chen, F.-R.; Kai, J.-J. Low-resistance ohmic contacts to p-type GaN achieved by the oxidation of Ni/Au films. J. Appl. Phys. 1999, 86, 4491. [CrossRef] 119. Lim, S.-H.; Ra, T.-Y.; Kim, W.-Y. Interface observation in Au/Ni/p-GaN studies by HREM and energy-filtering TEM. J. Electron Microsc. 2003, 52, 459–464. [CrossRef] 120. Qiao, D.; Yu, L.S.; Lau, S.S.; Lin, J.Y.; Jiang, H.X.; Haynes, T.E. A study of the Au/Ni ohmic contact on p-GaN. J. Appl. Phys. 2000, 88, 4196–4200. [CrossRef] 121. Chary, I.; Chandolu, A.; Borisov, B.; Kuryatkov, V.; Nikishin, S.; Holtz, M. Influence of surface treatment and annealing temperature on the formation of low-resistance Au/Ni ohmic contacts to p-GaN. J. Electron. Mater. 2009, 38, 545–550. [CrossRef] 122. Chary, I.; Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Nikishin, S.; Holtz, M. Low resistance ohmic contacts to p-type GaN and AlGaN. Mater. Res. Soc. Symp. Proc. 2009, 1108, 1108-A09-30. 123. Nikishin, S.; Chary, I.; Borisov, B.; Kuryatkov, V.; Kudryavtsev, Y.; Asomoza, R.; Karpov, S.Y.; Holtz, M. Mechanism of curriers injection in (Ni/Au)/p-Al Ga N:Mg (0  x < 0.1) ohmic contacts. Appl. Phys. Lett. 1x 2009, 95, 163502. [CrossRef] 124. Nikishin, S.; Borisov, B.; Kuryatkov, V.; Usikov, A.; Dmitriev, V.; Holtz, M. Deep UV AlGaN light emitting diodes grown by gas source molecular beam epitaxy on sapphire and AlGaN/sapphire substrates. Proc. SPIE 2006, 6121, 6121OT-1-16. [CrossRef] 125. Moustakas, T.D.; Liao, Y.; Kao, C.-K.; Thomidis, C.; Bhattacharyya, A.; Bhattarai, D.; Noldawer, A. Deep UV LEDs with high IQE based on AlGaN alloys with strong band structure potential fluctuations. Proc. SPIE 2012, 8278, 8278OL-1-11. [CrossRef] 126. Rass, J.; Kolbe, T.; Ploch, N.L.; Wernicke, T.; Mehnke, F.; Kuhn, C.; Enslin, J.; Guttmann, M.; Reich, C.; Mogilatenko, A.; et al. High power UV-B LEDs with long lifetime. Proc. SPIE 2015, 9363, 93631K-1-13. [CrossRef] 127. Mondal, R.K.; Chatterjee, V.; Singh, S.; Islam, S.M.; Pal, S. Optimization of structural parameters for highly efficient AlGaN based deep ultraviolet light emitting diodes. Superlattices Microstruct. 2017, 112, 339–352. [CrossRef] Appl. Sci. 2018, 8, 2362 17 of 17 128. Liu, M.; Zhou, S.; Liu, X.; Gao, Y.; Ding, X. Comparative experimental and simulation studies of high-power AlGaN-based 353 nm ultraviolet flip-chip and top-emitting LEDs. Jpn. J. Appl. Phys. 2018, 57, 031001. [CrossRef] 129. Bulashevich, K.A.; Ramm, M.S.; Karpov, S.Yu. Effects of electron and optical confinement on performance of UV laser diodes. Phys. Stat. Sol. 2009, 6, 603–606. [CrossRef] 130. Liang, Y.-H.; Towe, E. Heavy Mg-doping of (AlGa)N films for potential applications in deep ultraviolet light-emitting structures. J. Appl. Phys. 2018, 123, 095303. [CrossRef] 131. Park, J.H.; Kim, J.K.; Cho, J. Observation of space-charge-limited current in AlGaN/GaN ultraviolet light emitting diodes. Mater. Lett. 2018, 214, 217–219. [CrossRef] 132. Liu, D.; Cho, S.J.; Park, J.; Seo, J.-H.; Dalmau, R.; Zhao, D.; Kim, K.; Gong, J.; Kim, M.; Lee, I.-K.; et al. 229 nm UV LEDs on aluminum nitride single crystal substrates using p-type silicon for increased hole injection. Appl. Phys. Lett. 2018, 112, 081101. [CrossRef] © 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Journal

Applied SciencesMultidisciplinary Digital Publishing Institute

Published: Nov 23, 2018

There are no references for this article.