Mass of GaAsN in Pulsed Magnetic Fields up to 60 T with Free-Electron Laser IR Radiation

We use the unique combination of the widely tunable (4 μm – 250 μm) Free-Electron laser (FEL) FELBE and pulsed magnetic fields up to 60T of the High Magnetic Field Laboratory HLD to perform spectroscopic investigations on the dilute nitride system GaAsN. We carry out systematic cyclotron resonance (CR) spectroscopy and analyze the dependence of the electron effective mass on the nitrogen content. The red triangles in Figure 1 illustrate our findings for the illumination wavelength 46 μm at 100 K. We observe a slight increase of the effective mass with nitrogen content, which is in very good agreement with the Band Anti-Crossing (BAC) model [1], the empirical Tight Binding (TB) calculations [2] and the Two band BAC model [3], which are represented in Figure 1 by dashed, dotted and dash-dotted black lines, respectively. We compare our results with magneto-photoluminescence (PL) investigations performed by Alberi et al. [4] and Masia et al. [5], which are presented with blue circles and stars respectively. Magneto-PL investigations reveal a very fast increase of the effective mass with nitrogen content, well above the mentioned models [1-3], but consistent with the modified k·p calculations by Lindsay and O’Reilly [6]. Our magneto-PL study (not shown) exhibits a very similar behavior as shown by Alberi et al. and Masia et al., which allows us to exclude the different samples as a source for the deviation.
It is well known that nitrogen tends to form pairs and clusters during the growth, which is only considered in the modified k·p calculations [4]. Magneto-PL is a method which is very sensitive to localization of the neighboring atoms and thus to clusters. For this reason the magneto-PL results are consistent with [4], but cannot be described by [1-3], which do not take clusters into account. On the other hand, CR spectroscopy is only sensitive to delocalized states and this is why our results are in such good agreement with [1-3].

[1] J. Wu et al. Phys. Rev. B 64, 085320 (2000).
[2] N. Shtinkov et al. Phys. Rev. B 67, 081202(R) (2003).
[3] Tomic et al. Phys. Rev. B 69, 245305 (2004).
[4] Alberi et al. Phys. Rev. Lett. 110, 156405 (2013).
[5] Masia et al. Phys. Rev. B 73, 073201, (2006).
[6] A. Lindsay and E. P. O’Reilly Phys. Rev. Lett. 93, 196402 (2004).