IntroductionSince the first demonstration of high-brightness blue light-emitting diodes (LEDs) in the early 90s [1,2], group-III nitride semiconductors have gone a long way from being just promising materials in the industrial world to the basis for modern visible and ultra-violet (UV) optoelectronics. This has happened, on the one hand, due to the unique physical properties of these semiconductors, namely, a direct bandgap that could be varied from 0.65 to 6.2 eV, depending on the III-nitride alloy composition [3][4][5]; and, on the other, to reliable and effective p-doping demonstrated in earlier studies [6,7]. Very quickly, blue III-nitride LEDs were capable of providing an emission efficiency suitable for application, in spite of an extremely high threading dislocation density (TDD), ∼10 8 -10 10 cm −2 , inherent in epitaxial materials on mismatched sapphire or SiC substrates. This breakthrough stimulated an enormous amount of activity in III-nitride semiconductor technology, and soon the practical achievements outpaced the understanding of physical mechanisms underlying the LED operation. In part, this was caused by some novel physical factors, such as electrical spontaneous polarization and a strong piezoeffect [8,9], the significance of which for LED operation was recognized only after a remarkable delay.The research and development stage has been completed with a world-wide mass production of III-nitride LEDs, establishing new targets aimed at improving the device performance; primarily brightness and efficiency. These targets require a deeper insight into the physics of LED operation, as further effort can no longer rely on the intuitive design of LED heterostructures and analogy with conventional III-V devices. As the potential of many experimental approaches to LED optimization is likely to be almost exhausted, the role of theoretical work in the further improvement of device performance has become much more important.