Exciton Luminescence of Single-Crystal GaN
- PDF / 350,826 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 25 Downloads / 217 Views
at Pex > 20W/cm2. INTRODUCTION GaN has become subject of intensive research, since it is not only a promising candidate for optical devices in the near-UV spectral range, but also for high-power and high-frequency electronic applications. Triggered by a realization of a blue LED in 1994, many groups have started working on group-ITT nitrides grown on sapphire with the objective to develop a blue laser diode. However, fundamental problems regarding, e.g., crystal quality, cleavability and p-type doping are still to be solved. Even though a sound understanding of the dominant recombination channels at room temperature in general and the intrinsic luminescence in particular are of crucial importance for the design of such a laser diode , little is known about these processes. In this work, we fill this gap for both phases not only by investigations on exciton recombination, band-band recombination and carrier trapping, but also by determination of the figures of merit for device operation, which are the quantum efficiency and the (non)radiative lifetimes. We mainly focus on the two radiative transitions at FXc=3.27 eV and FXH=3.47 eV (T=4 K). Cathodoluminescence (CL) studies were used to assign these to cubic and hexagonal GaN [1], respectively. By a careful analysis of the temperature dependence of the photoluminescence, we show that FXc and FXH follow the band-gap of cubic [2] and hexagonal GaN [3], respectively, being shifted to lower energies by the estimated free-exciton binding energy Rx - 28 meV [3]. Having identified FXC and FXH to arise from free-exciton recombination, we determine their quantum efficiency and investigate the room-temperature behavior of the band-gap luminescence (BGL), which consists of the band-band and free-exciton transitions. We model the dependence of the BGL intensity on Pt× and extract the (non)radiative lifetimes. The values obtained are not only consistent with time-resolved PL studies [4], but also with the phenomenon of defect saturation, which we observe to occur for the sample investigated. 607
Mat. Res. Soc. Symp. Proc. Vol. 395 e 19 9 6 Materials Research Society
EXPERIMENT: Sample and Photoluminescence Apparatus All the measurements were performed on one sample grown in a custom-made MBE system equipped with a DC glow-discharge N 2 plasma source. A 1.5 pm thick single-phase cubic GaN film was grown on a (100) GaAs substrate. On top of this film, cubic and hexagonal microcrystals with diameters of - 3 Pm were grown by a vapor-liquid-solid like epitaxial process under Ga-rich conditions [5]. The crystals, being extremely pure and hence defining an optical standard, dominate the sample's luminescence. Therefore, this sample structure was chosen for our investigations. The structural and morphological properties of the sample were verified by means of in situ RHEED, XRD, TEM and SEM, while CL was used to selectively analyze the crystals and the film [1]. The PL measurements were carried out in a standard apparatus using the 325 nm line
of a HeCd laser for carrier injection. N
Data Loading...
