We investigate the impact of Mg-doping on the performance and degradation kinetics of AlGaN-based UV-C light-emitting diodes (LEDs). By comparing LEDs from three wafers with different nominal doping levels [Mg/(Al+Ga) ratio: 0.15%, 0.5%, and 1% in the gas phase during epitaxy] in the AlGaN:Mg electron-blocking layer (EBL), we demonstrate the following results: (i) A higher Mg-doping in the EBL results in a higher optical power at low current levels, which is ascribed to an increased hole injection efficiency. (ii) The reduction of the optical power follows a non-exponential trend, which can be reproduced by using the Hill's formula and is ascribed to the generation/activation of defects within the quantum wells. (iii) A higher Mg-doping in the EBL mitigates the degradation rate. An interpretation of the experimental data is proposed, assuming that hydrogen, which is present in and moving from the EBL, can reduce the rate of de-hydrogenation of point defects in the active region, which is responsible for degradation.

The technology of AlGaN-based ultraviolet (UV) light emitting diodes (LEDs) has shown impressive advancements over the past few years. In addition to the classic applications, such as UV curing,1–31. A. Endruweit, M. S. Johnson, and A. C. Long, “ Curing of composite components by ultraviolet radiation: A review,” Polym. Compos. 27(2), 119–128 (2006). https://doi.org/10.1002/pc.201662. C. Dreyer and F. Mildner, Springer Series in Materials Science ( Springer Verlag, 2016), pp. 415–434.3. P. E. Hockberger, “ A history of ultraviolet photobiology for humans, animals and microorganisms,” Photochem. Photobiol. 76(6), 561–579 (2007). https://doi.org/10.1562/0031-8655(2002)0760561AHOUPF2.0.CO2 disinfection and water purification,4–64. M. Mori, A. Hamamoto, A. Takahashi, M. Nakano, N. Wakikawa, S. Tachibana, T. Ikehara, Y. Nakaya, M. Akutagawa, and Y. Kinouchi, “ Development of a new water sterilization device with a 365 nm UV-LED,” Med. Bio. Eng. Comput. 45(12), 1237–1241 (2007). https://doi.org/10.1007/s11517-007-0263-15. M. A. Würtele, T. Kolbe, M. Lipsz, A. Külberg, M. Weyers, M. Kneissl, and M. Jekel, “ Application of GaN-based ultraviolet-C light emitting diodes—UV LEDs—for water disinfection,” Water Res. 45(3), 1481–1489 (2011). https://doi.org/10.1016/j.watres.2010.11.0156. W. Kowalski, Ultraviolet Germicidal Irradiation Handbook: UVGI for Air and Surface Disinfection ( Springer, Berlin, Heidelberg, 2009). plant growth and bioagent detection,7–107. A. Žukauskas, N. Kurilčik, P. Vitta, S. Juršėnas, E. Bakienė, and R. Gaska, “Optimization of a UV light-emitting diode based fluorescence-phase sensor,” Proc SPIE 6398, 63980Y (2006). https://doi.org/10.1117/12.6899078. M. Schreiner, J. Martínez-Abaigar, J. Glaab, and M. Jansen, “ UV-B induced secondary plant metabolites,” Opt. Photonik 9(2), 34–37 (2014). https://doi.org/10.1002/opph.2014000489. M. S. Shur and R. Gaska, “ Deep-ultraviolet light-emitting diodes,” IEEE Trans. Electron Devices 57(1), 12–25 (2010). https://doi.org/10.1109/TED.2009.203376810. P. J. Hargis, Jr., T. J. Sobering, G. C. Tisone, J. S. Wagner, S. A. Young, and R. J. Radloff, in Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, edited by J. Leonelli , D. K. Killinger , W. Vaughan , and M. G. Yost ( SPIE, 1995), p. 147. during the last few years, a big push came from the fight against COVID-19; in fact, specific antiviral treatments are being developed, for the sanitation of surfaces and objects.1111. W. J. Kowalski, T. J. Walsh, and V. Petraitis, COVID-19 Coronavirus Ultraviolet Susceptibility 2020 COVID-19 Coronavirus Ultraviolet Susceptibility ( PurpleSun, Inc., 2020), pp. 1–4. The classic advantages of UV LEDs compared to traditional UV lamps1212. K. C. Anyaogu, A. A. Ermoshkin, D. C. Neckers, A. Mejiritski, O. Grinevich, and A. V. Fedorov, “ Performance of the light emitting diodes versus conventional light sources in the UV light cured formulations,” J. Appl. Polym. Sci. 105(2), 803–808 (2007). https://doi.org/10.1002/app.26313 include the smaller size, lower drive voltage, and the higher spectral purity and the environmental friendliness. On the other hand, the market penetration is slowed by the low values of external quantum efficiency (EQE), the difficulty of obtaining high p-doping levels,1313. F. Piva, C. De Santi, M. Deki, M. Kushimoto, H. Amano, H. Tomozawa, N. Shibata, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Stability and degradation of AlGaN-based UV-B LEDs: Role of doping and semiconductor defects,” Microelectron. Rel. 100–101, 113418 (2019). https://doi.org/10.1016/j.microrel.2019.113418 and the relatively short lifetime that has been ascribed to the generation of non-radiative recombination centers during aging,1414. J. Glaab, J. Haefke, J. Ruschel, M. Brendel, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, M. Guttmann, C. Kuhn, J. Enslin, T. Wernicke, and M. Kneissl, “ Degradation effects of the active region in UV-C light-emitting diodes,” J. Appl. Phys. 123(10), 104502 (2018). https://doi.org/10.1063/1.5012608 which can involve the migration of hydrogen.1515. J. Glaab, J. Ruschel, T. Kolbe, A. Knauer, J. Rass, H. K. Cho, N. L. Ploch, S. Kreutzmann, S. Einfeldt, M. Weyers, and M. Kneissl, “ Degradation of (In)AlGaN-based UVB LEDs and migration of hydrogen,” IEEE Photonics Technol. Lett. 31(7), 529–532 (2019). https://doi.org/10.1109/LPT.2019.2900156The aim of this work is to contribute to the understanding of the degradation physics of UV-C LEDs, by investigating the impact of Mg-doping on the device lifetime. We considered a series of 265 nm UV-C LEDs with different nominal Mg-content in the electron blocking layer (EBL),9,169. M. S. Shur and R. Gaska, “ Deep-ultraviolet light-emitting diodes,” IEEE Trans. Electron Devices 57(1), 12–25 (2010). https://doi.org/10.1109/TED.2009.203376816. M. Shatalov, W. Sun, R. Jain, A. Lunev, X. Hu, A. Dobrinsky, Y. Bilenko, J. Yang, G. A. Garrett, L. E. Rodak, M. Wraback, M. Shur, and R. Gaska, “ High power AlGaN ultraviolet light emitters,” Semicond. Sci. Technol. 29(8), 084007 (2014). https://doi.org/10.1088/0268-1242/29/8/084007 which were subjected to electro-optical characterization and degradation testing. The experimental results indicate that a higher Mg-content in the EBL results in a higher efficiency at low current levels and in a longer lifetime. In this paper, we propose that hydrogen form in the EBL is the reason for mitigating the defect-generation process.The devices under test come from three wafers of AlGaN-based UV-C LEDs, with a peak emission wavelength of 264 ± 1 nm, have nominally identical epitaxial structures (see below), and they differ only in the Mg/(Al+Ga) ratio of the EBL. It is important to state that the Mg-concentration in the EBL was varied by adjusting the Mg/(Al+Ga) ratio of the corresponding partial pressures during MOVPE growth of the EBL to 0.15%, 0.5%, and 1.0%, respectively. The Mg/(Al+Ga) molar ratio in the solid differs from these values due to reactions in the gas phase and desorption and segregation of Mg.1717. C. Kuhn, T. Simoneit, M. Martens, T. Markurt, J. Enslin, F. Mehnke, K. Bellmann, T. Schulz, M. Albrecht, T. Wernicke, and M. Kneissl, “ MOVPE growth of smooth and homogeneous Al0.8Ga0.2N:Si superlattices as UVC laser cladding layers,” Phys. Status Solidi A 215(13), 1800005 (2018). https://doi.org/10.1002/pssa.201800005 The structure consists of an epitaxially laterally overgrown AlN/sapphire substrate with a threading dislocation density (TDD) of 1.5 × 109 cm−2,1818. N. Susilo, E. Ziffer, S. Hagedorn, L. Cancellara, C. Netzel, N. L. Ploch, S. Wu, J. Rass, S. Walde, L. Sulmoni, M. Guttmann, T. Wernicke, M. Albrecht, M. Weyers, and M. Kneissl, “ Improved performance of UVC-LEDs by combination of high-temperature annealing and epitaxially laterally overgrown AlN/sapphire,” Photonics Res. 8(4), 589–594 (2020). https://doi.org/10.1364/PRJ.385275 followed by AlN and AlGaN transition layers, a thick n-Al0.76Ga0.24N layer,1919. F. Mehnke, T. Wernicke, H. Pingel, C. Kuhn, C. Reich, V. Kueller, A. Knauer, M. Lapeyrade, M. Weyers, and M. Kneissl, “ Highly conductive n-AlxGa1−xN layers with aluminum mole fractions above 80%,” Appl. Phys. Lett. 103(21), 212109 (2013). https://doi.org/10.1063/1.4833247 the active region composed of three Al0.48Ga0.52N quantum wells (QWs) alternated with Al0.63Ga0.37N quantum barriers, an EBL composed of a first nominally undoped 5 nm Al0.85Ga0.15N interlayer (IL) followed by a 25 nm thick p-Al0.80Ga0.20N/Al0.70Ga0.30N multibarrier EBL,2020. T. Kolbe, A. Knauer, J. Rass, H. K. Cho, S. Hagedorn, S. Einfeldt, M. Kneissl, and M. Weyers, “ Effect of electron blocking layer doping and composition on the performance of 310 nm light emitting diodes,” Materials 10(12), 1396 (2017). https://doi.org/10.3390/ma10121396 a 183 nm p-Al0.27Ga0.73N/Al0.17Ga0.83N superlattice (Mg/(Al+Ga) ratio 1.5%), and a 40 nm thick p-GaN contact layer (Mg/Ga ratio 2%).As a first step, a characterization of the electro-optical properties was carried out before a long-term operation. The results demonstrate that the optical power emitted by the devices at high measuring current levels is independent of the Mg-doping in the EBL [Fig. 1(a)]. On the other hand, at low measuring current levels, we found that a high Mg-doping can result in a higher efficiency, compared to devices with low Mg-doping. This result can be explained by considering that a higher Mg-doping in the EBL results in a higher hole injection efficiency,2020. T. Kolbe, A. Knauer, J. Rass, H. K. Cho, S. Hagedorn, S. Einfeldt, M. Kneissl, and M. Weyers, “ Effect of electron blocking layer doping and composition on the performance of 310 nm light emitting diodes,” Materials 10(12), 1396 (2017). https://doi.org/10.3390/ma10121396 which can improve the optical performance of the LEDs, especially at low measuring current levels. Unfortunately, Mg concentration and acceptor concentration are not equal and too high doping levels lead to a decreased acceptor density2121. U. Kaufmann, P. Schlotter, H. Obloh, K. Köhler, and M. Maier, “ Hole conductivity and compensation in epitaxial GaN:Mg layers,” Phys. Rev. B 62(16), 10867 (2000). https://doi.org/10.1103/PhysRevB.62.10867 and lower injection efficiency.2020. T. Kolbe, A. Knauer, J. Rass, H. K. Cho, S. Hagedorn, S. Einfeldt, M. Kneissl, and M. Weyers, “ Effect of electron blocking layer doping and composition on the performance of 310 nm light emitting diodes,” Materials 10(12), 1396 (2017). https://doi.org/10.3390/ma10121396 Another possible explanation considers recent reports, suggesting that gallium vacancies may behave as non-radiative recombination centers,2222. F. Piva, C. de Santi, M. Deki, M. Kushimoto, H. Amano, H. Tomozawa, N. Shibata, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Modeling the degradation mechanisms of AlGaN-based UV-C LEDs: From injection efficiency to mid-gap state generation,” Photonics Res. 8(11), 1786 (2020). https://doi.org/10.1364/PRJ.401785 when they are not passivated by hydrogen atoms in the form VGa−Hn.1414. J. Glaab, J. Haefke, J. Ruschel, M. Brendel, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, M. Guttmann, C. Kuhn, J. Enslin, T. Wernicke, and M. Kneissl, “ Degradation effects of the active region in UV-C light-emitting diodes,” J. Appl. Phys. 123(10), 104502 (2018). https://doi.org/10.1063/1.5012608 Previous reports23–2523. T. Stephan, K. Koehler, M. Maier, M. Kunzer, P. Schlotter, and J. Wagner, “ Influence of Mg doping profile on the electroluminescence properties of GaInN multiple-quantum-well light-emitting diodes,” in Light-Emitting Diodes: Research, Manufacturing, and Applications VIII ( SPIE, 2004), Vol. 5366, p. 118.24. N. Zhang, X.-C. Wei, K.-Y. Lu, L.-S. Feng, J. Yang, B. Xue, Z. Liu, J.-M. Li, and J.-X. Wang, “ Effect of back diffusion of Mg dopants on optoelectronic properties of InGaN-based green light-emitting diodes,” Chin. Phys. Lett. 33(11), 117302 (2016). https://doi.org/10.1088/0256-307X/33/11/11730225. N. E. Posthuma, S. You, H. Liang, N. Ronchi, X. Kang, D. Wellekens, Y. N. Saripalli, and S. Decoutere, “ Impact of Mg out-diffusion and activation on the p-GaN gate HEMT device performance,” in Proceedings of the International Symposium on Power Semiconductor Devices and ICs ( IEEE, 2016), pp. 95–98. indicated that a higher Mg-doping may result in a higher H-content in the p-type layers and neighboring regions. We suggest that the use of a higher doping in the EBL may result in a higher H incorporation, which could cause a higher passivation of gallium vacancies by hydrogen, and, thus, a higher radiative recombination efficiency.To study the impact of the Mg-doping in the EBL on the reliability, a single representative device from each of the three wafers was submitted to constant current stress. We defined the stress conditions considering that (a) recent reports2626. J. Ruschel, J. Glaab, B. Beidoun, N. L. Ploch, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, and M. Kneissl, “ Current-induced degradation and lifetime prediction of 310 nm ultraviolet light-emitting diodes,” Photonics Res. 7(7), B36–B40 (2019). https://doi.org/10.1364/PRJ.7.000B36 suggested that the reduction in the optical power is dependent on carrier density in the quantum wells2727. T. Kolbe, A. Knauer, J. Ruschel, J. Rass, H. K. Cho, S. Hagedorn, J. Glaab, N. L. Ploch, S. Einfeldt, and M. Weyers, “ Comparison of ultraviolet B light-emitting diodes with single or triple quantum wells,” Phys. Status Solidi A 218, 2100100 (2021). https://doi.org/10.1002/pssa.202100100 and could be promoted by Auger recombination events2828. N. Renso, C. de Santi, A. Caria, F. D. Torre, L. Zecchin, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Degradation of InGaN-based LEDs: Demonstration of a recombination-dependent defect-generation process,” J. Appl. Phys. 127(18), 185701 (2020). https://doi.org/10.1063/1.5135633 (b) the degradation rate may be temperature dependent.To allow a fair comparison between the wafers with different Mg-doping in the EBL, all samples were aged with the same junction temperature and the same (initial) carrier density in the wells. For this purpose, first the junction temperature was evaluated by the method proposed by Xi and Schubert in Ref. 2929. Y. Xi and E. F. Schubert, “ Junction-temperature measurement in GaN ultraviolet light-emitting diodes using diode forward voltage method,” Appl. Phys. Lett. 85(12), 2163–2165 (2004). https://doi.org/10.1063/1.1795351 and adjusted by changing the baseplate temperature for each stressed sample. Second, for each sample, the stress current level was chosen to yield the same initial optical power. The optical power is proportional to VBnp(hν) (where V is the recombination volume, B is the bimolecular recombination coefficient, and hν is the energy of the emitted photon). Stressing the devices at the same output power then corresponds to fixing the carrier density (at the beginning of the stress) at the same value for all samples, under the assumption that the B coefficient does not depend on the Mg-doping in the EBL. This is a realistic assumption, given that at high current densities, all samples have comparable optical powers (compare Fig. 1). During the stress experiment, the optical properties of the devices were repeatedly monitored.The results on the electro-optical degradation of the devices are summarized in Fig. 2(a) that reports the optical power vs current (L–I) plots taken before/after stress. Stress was found to induce a decrease in the optical power for all analyzed samples. This drop was found to be stronger in the low-measuring current region (I = 1 mA and J = 0.31 A cm−2), where the slope of the double logarithmic L-I plots is larger than 2 [Fig. 2(b), filled symbols]. Values in this range are usually ascribed to the interplay of Shockley–Read–Hall (SRH) recombination and tunneling/surface leakage currents,30,3130. K. S. Kim, D. P. Han, H. S. Kim, and J. I. Shim, “ Analysis of dominant carrier recombination mechanisms depending on injection current in InGaN green light emitting diodes,” Appl. Phys. Lett. 104(9), 091110 (2014). https://doi.org/10.1063/1.486764731. M. Buffolo, A. Magri, C. de Santi, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Gradual degradation of InGaAs LEDs: Impact on non-radiative lifetime and extraction of defect characteristics,” Materials 14(5), 1114 (2021). https://doi.org/10.3390/ma14051114 indicating the presence of defect-mediated recombination and conduction processes. A drop in the efficiency in the low measuring current regime is indicative of the increase in the density of defects in the active region of the devices, consistently with previous reports on the topic.2222. F. Piva, C. de Santi, M. Deki, M. Kushimoto, H. Amano, H. Tomozawa, N. Shibata, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Modeling the degradation mechanisms of AlGaN-based UV-C LEDs: From injection efficiency to mid-gap state generation,” Photonics Res. 8(11), 1786 (2020). https://doi.org/10.1364/PRJ.401785On the other hand, at high measuring current levels (I = 100 mA and J = 31 A cm−2), a strong degradation was detected only for the samples with low Mg-doping in the EBL, while devices with high Mg-doping showed only a moderate decrease in the optical power. At high measuring current levels, the initial slope of the double-logarithmic L-I plots are close to 1 [Fig. 2(b), open symbols]. However, it increases for longer operation times. This is indicative of the fact that stress is inducing an increase in the non-radiative recombination and/or an increase in carrier leakage from the quantum wells, due to the generation of defects.3232. D. Monti, C. de Santi, S. da Ruos, F. Piva, J. Glaab, J. Rass, S. Einfeldt, F. Mehnke, J. Enslin, T. Wernicke, M. Kneissl, G. Meneghesso, E. Zanoni, and M. Meneghini, “ High-current stress of UV-B (In)AlGaN-based LEDs: Defect-generation and diffusion processes,” IEEE Trans. Electron Devices 66(8), 3387–3392 (2019). https://doi.org/10.1109/TED.2019.2920521We now discuss the degradation trends of the optical power. Figure 3(a) shows the optical power over time measured at the stress current level on the samples with different Mg-doping in the EBL. Based on previous reports, defect generation is supposed to be promoted by the energy released by non-radiative Auger–Meitner recombination events.2626. J. Ruschel, J. Glaab, B. Beidoun, N. L. Ploch, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, and M. Kneissl, “ Current-induced degradation and lifetime prediction of 310 nm ultraviolet light-emitting diodes,” Photonics Res. 7(7), B36–B40 (2019). https://doi.org/10.1364/PRJ.7.000B36 Consistently with earlier papers, the energy of ∼4.7 eV (bandgap energy of such an event33,3433. R. Armitage, W. Hong, Q. Yang, H. Feick, J. Gebauer, E. R. Weber, S. Hautakangas, and K. Saarinen, “ Contributions from gallium vacancies and carbon-related defects to the ‘yellow luminescence’ in GaN,” Appl. Phys. Lett. 82(20), 3457–3459 (2003). https://doi.org/10.1063/1.157816934. J. Neugebauer and C. G. van de Walle, “ Gallium vacancies and the yellow luminescence in GaN,” Appl. Phys. Lett. 69(4), 503–505 (1996). https://doi.org/10.1063/1.117767) would be sufficient for the activation of non-radiative centers. The proposed mechanism is the de-hydrogenation of gallium vacancies (from VGaHn to VGaHn−1) that requires a moderate energy (Ea ∼ 2.03 eV1515. J. Glaab, J. Ruschel, T. Kolbe, A. Knauer, J. Rass, H. K. Cho, N. L. Ploch, S. Kreutzmann, S. Einfeldt, M. Weyers, and M. Kneissl, “ Degradation of (In)AlGaN-based UVB LEDs and migration of hydrogen,” IEEE Photonics Technol. Lett. 31(7), 529–532 (2019). https://doi.org/10.1109/LPT.2019.2900156) prior to stress that the VGaHn complexes are neutral; if one or more hydrogen atoms are released, the VGaHn−1 can act as centers of non-radiative recombination,1414. J. Glaab, J. Haefke, J. Ruschel, M. Brendel, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, M. Guttmann, C. Kuhn, J. Enslin, T. Wernicke, and M. Kneissl, “ Degradation effects of the active region in UV-C light-emitting diodes,” J. Appl. Phys. 123(10), 104502 (2018). https://doi.org/10.1063/1.5012608 thus lowering the efficiency of the devices.The experimental results indicated that the degradation data could not be fitted by an exponential function, as conventionally done for visible LEDs (see, for instance, the IES TM-21 standard). Remarkably, the results could be fitted by the Hill's formula, an equation typically used to describe chemical reactions (in our case, this would be the de-hydrogenation of gallium vacancies), having the form below: Here, a, b, c, and d are fitting parameters. Equation (1) was used to fit the experimental data of all devices. a=0.1, c=0.01, and d=0.45 were fixed for all samples, whereas b was fitted independently. The results of the fitting are shown in Fig. 3(a) and are well aligned with the experimental data.Based on the results above, we explain the degradation process as follows: during stress, the samples with different Mg doping are operated at the same temperature and, at the beginning, at the same carrier density in the active region. Temperature and carrier density are supposed to be the main driving forces of degradation, assuming that defect generation results from the energy released by Auger–Meitner recombination events,2828. N. Renso, C. de Santi, A. Caria, F. D. Torre, L. Zecchin, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Degradation of InGaN-based LEDs: Demonstration of a recombination-dependent defect-generation process,” J. Appl. Phys. 127(18), 185701 (2020). https://doi.org/10.1063/1.5135633 whose rate depends on carrier density and temperature. Given the similar properties of the active region, it is reasonable to assume that the defect-generation rate is the same for the samples with different Mg-doping in the EBL. Defect generation is proposed to proceed through the de-hydrogenation of defects, as discussed above. In the samples with higher Mg-doping, a higher density of hydrogen is present in the EBL. Hydrogen could diffuse toward the active region, following the concentration gradient3535. C. H. Seager, S. M. Myers, A. F. Wright, D. D. Koleske, and A. A. Allerman, “ Drift, diffusion, and trapping of hydrogen in p-type GaN,” J. Appl. Phys. 92(12), 7246–7252 (2002). https://doi.org/10.1063/1.1520719 or it could drift following the electric field.1515. J. Glaab, J. Ruschel, T. Kolbe, A. Knauer, J. Rass, H. K. Cho, N. L. Ploch, S. Kreutzmann, S. Einfeldt, M. Weyers, and M. Kneissl, “ Degradation of (In)AlGaN-based UVB LEDs and migration of hydrogen,” IEEE Photonics Technol. Lett. 31(7), 529–532 (2019). https://doi.org/10.1109/LPT.2019.2900156 H reaching the active region could mitigate the vacancy de-hydrogenation process (i.e., reduce the de-hydrogenation rate), thus resulting in a slower degradation for the samples with higher Mg in the EBL. It is worth noticing that in principle, the release of hydrogen from the EBL could also change the band diagram of the EBL, thus modifying the injection efficiency. This is qualitatively expected to result in an improvement in hole injection (due to a higher Mg activation) and an increasing electron-blocking capability by the EBL, which could also contribute to the lower degradation observed when high Mg concentrations are used in the EBL.

In conclusion, we investigated the impact of the Mg-doping level in the EBL on the performance and degradation kinetics of UV-C LEDs emitting at a wavelength of 265 nm. First, we demonstrated that a higher Mg-doping in the EBL results in a better optical performance, which has been ascribed to a higher hole injection efficiency and to a lower non-radiative recombination efficiency in the active region. Second, by stressing devices with different Mg-doping at the same junction temperature and carrier density in the active region, we demonstrated that a higher Mg-doping results in a lower degradation rate. Third, the temporal behavior of the optical power has been analytically investigated: degradation has been explained by assuming that stress proceeds through the de-hydrogenation of point defects (behaving as non-radiative recombination centers), which is promoted by the energy released by non-radiative recombination events.

This research was partly supported by the German Federal Ministry of Education and Research (BMBF) through the consortium “Advanced UV for Life” under the Project Contract Nos. 03ZZ0130A, 03ZZ0130B, 03ZZ134B, and 03ZZ134C.

Conflict of Interest

The authors have no conflicts to disclose.

Author Contributions

Francesco Piva: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Jens Rass: Conceptualization (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Sven Einfeldt: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Norman Susilo: Conceptualization (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Tim Wernicke: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Michael Kneissl: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Gaudenzio Meneghesso: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Enrico Zanoni: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Matteo Meneghini: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Massimo Grigoletto: Data curation (equal); Investigation (equal); Methodology (equal). Riccardo Brescancin: Data curation (equal); Investigation (equal); Methodology (equal). Carlo De Santi: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Matteo Buffolo: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Jan Ruschel: Conceptualization (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Johannes Glaab: Conceptualization (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Daniel Hauer Vidal: Conceptualization (equal); Investigation (equal); Methodology (equal); Visualization (equal); Writing – review & editing (equal). Martin Guttmann: Conceptualization (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

1. A. Endruweit, M. S. Johnson, and A. C. Long, “ Curing of composite components by ultraviolet radiation: A review,” Polym. Compos. 27(2), 119–128 (2006). https://doi.org/10.1002/pc.20166, Google ScholarCrossref2. C. Dreyer and F. Mildner, Springer Series in Materials Science ( Springer Verlag, 2016), pp. 415–434. Google Scholar3. P. E. Hockberger, “ A history of ultraviolet photobiology for humans, animals and microorganisms,” Photochem. Photobiol. 76(6), 561–579 (2007). https://doi.org/10.1562/0031-8655(2002)0760561AHOUPF2.0.CO2, Google ScholarCrossref4. M. Mori, A. Hamamoto, A. Takahashi, M. Nakano, N. Wakikawa, S. Tachibana, T. Ikehara, Y. Nakaya, M. Akutagawa, and Y. Kinouchi, “ Development of a new water sterilization device with a 365 nm UV-LED,” Med. Bio. Eng. Comput. 45(12), 1237–1241 (2007). https://doi.org/10.1007/s11517-007-0263-1, Google ScholarCrossref5. M. A. Würtele, T. Kolbe, M. Lipsz, A. Külberg, M. Weyers, M. Kneissl, and M. Jekel, “ Application of GaN-based ultraviolet-C light emitting diodes—UV LEDs—for water disinfection,” Water Res. 45(3), 1481–1489 (2011). https://doi.org/10.1016/j.watres.2010.11.015, Google ScholarCrossref6. W. Kowalski, Ultraviolet Germicidal Irradiation Handbook: UVGI for Air and Surface Disinfection ( Springer, Berlin, Heidelberg, 2009). Google ScholarCrossref7. A. Žukauskas, N. Kurilčik, P. Vitta, S. Juršėnas, E. Bakienė, and R. Gaska, “Optimization of a UV light-emitting diode based fluorescence-phase sensor,” Proc SPIE 6398, 63980Y (2006). https://doi.org/10.1117/12.689907, Google ScholarCrossref8. M. Schreiner, J. Martínez-Abaigar, J. Glaab, and M. Jansen, “ UV-B induced secondary plant metabolites,” Opt. Photonik 9(2), 34–37 (2014). https://doi.org/10.1002/opph.201400048, Google ScholarCrossref9. M. S. Shur and R. Gaska, “ Deep-ultraviolet light-emitting diodes,” IEEE Trans. Electron Devices 57(1), 12–25 (2010). https://doi.org/10.1109/TED.2009.2033768, Google ScholarCrossref10. P. J. Hargis, Jr., T. J. Sobering, G. C. Tisone, J. S. Wagner, S. A. Young, and R. J. Radloff, in Optical Instrumentation for Gas Emissions Monitoring and Atmospheric Measurements, edited by J. Leonelli , D. K. Killinger , W. Vaughan , and M. G. Yost ( SPIE, 1995), p. 147. Google ScholarCrossref11. W. J. Kowalski, T. J. Walsh, and V. Petraitis, COVID-19 Coronavirus Ultraviolet Susceptibility 2020 COVID-19 Coronavirus Ultraviolet Susceptibility ( PurpleSun, Inc., 2020), pp. 1–4. Google Scholar12. K. C. Anyaogu, A. A. Ermoshkin, D. C. Neckers, A. Mejiritski, O. Grinevich, and A. V. Fedorov, “ Performance of the light emitting diodes versus conventional light sources in the UV light cured formulations,” J. Appl. Polym. Sci. 105(2), 803–808 (2007). https://doi.org/10.1002/app.26313, Google ScholarCrossref13. F. Piva, C. De Santi, M. Deki, M. Kushimoto, H. Amano, H. Tomozawa, N. Shibata, G. Meneghesso, E. Zanoni, and M. Meneghini, “ Stability and degradation of AlGaN-based UV-B LEDs: Role of doping and semiconductor defects,” Microelectron. Rel. 100–101, 113418 (2019). https://doi.org/10.1016/j.microrel.2019.113418, Google ScholarCrossref14. J. Glaab, J. Haefke, J. Ruschel, M. Brendel, J. Rass, T. Kolbe, A. Knauer, M. Weyers, S. Einfeldt, M. Guttmann, C. Kuhn, J. Enslin, T. Wernicke, and M. Kneissl, “ Degradation effects of the active region in UV-C light-emitting diodes,” J. Appl. Phys. 123(10), 104502 (2018). https://doi.org/10.1063/1.5012608, Google ScholarScitation, ISI15. J. Glaab, J. Ruschel, T. Kolbe, &#