Reservoir model for twodimensional electron gases in quantizing magnetic fields: a review
Download 1.56 Mb. Pdf ko'rish
|
71 adabiyot zawadzki2013
physica
s s p status solidi b the complete probability of the two CR transitions is 2eB hc n þ 1 ð Þ 1 a ð Þ þ n þ 2 ð Þa ½ ¼ 2eB hc n þ 1 þ a ð Þ ¼ N: ð19Þ Thus, also in case of the partial occupation of the level, when two CR transitions are possible, the complete probability may be expressed by N in the classical description. This reasoning is valid for a 2D system with its ‘universal’ degeneracy of LLs, but not for a 3D system. The CR measurements and the quantum transport measurements were performed on the same GaAs/Ga 0.67 A- l 0.33 As heterostuctures in the QHE regime. The best fit to the CR data was used to determine two parameters: t and N, see Eq. (17). The relation r xy ¼ B/Nec was used to determine the electron density N from the Hall effect at all magnetic fields. As mentioned above, in this interpretation in order to have a plateau of r xy the density N must increase linearly with B. Figures 16 –18 summarize the main findings. It is seen in Figs. 16 and 17 that both the CR and r xy measured in the Hall effect indicate oscillatory character of the electron density N with varying magnetic field for two different samples. In Fig. 18 the electron density determined from the CR data is compared with calculations assuming that an outside reservoir pins the Fermi level. It is seen that the reservoir hypothesis works quite well. The oscillations of N(B) were observed on two samples investigated by Manasreh et al. [12] and on all three samples studied by Raymond et al. [13]. Two important conclusions of the above analysis are: (i) The electron density N in GaAs QW oscillates as a function of a magnetic field; (ii) the ‘classical’ formula r xy ¼ B/Nec seems to work at all fields, also in the quantum Hall regime. 5 Magneto-photoluminescence (MPL) From the early days of optical experiments with semiconductor heterostructures it was observed that the energies of interband MPL transitions exhibit striking nonlinear behav- iour as functions of an external magnetic field. Such nonlinearities are characteristic of 2D systems and are not seen on bulk materials. The investigated systems can contain only one or more populated electric subbands. In the second case, it was shown that the nonlinearities were related to an electron transfer between the subbands. The situation is different if only one subband is populated. An important example of such a situation is a rather narrow and not strongly doped GaAs/Ga 0.67 Al 0.33 /As QW. In this case one cannot explain the nonlinearities by the above mechanism. A typical example of nonlinearities in MPL data is presented in Figure 15 CR transitions in a 2D system: (a) for the Fermi energy located between LLs and (b) for E F located within an LL (schematically). The filling factor of the n þ 1 level is a. After Ref. [13]. Figure 16 The electron density for GaAs/GaAlAs (sample 2) versus magnetic field, as determined by a fit to the CR data and from the Hall resistance with the use of the formula r xy ¼ B/Nec. After Ref. [13]. Figure 17 The same as in Fig. 16 but for sample 3. After Ref. [13]. Figure 18 The electron density for GaAs/GaAlAs (sample 1) versus magnetic field, as calculated using the reservoir model (solid line). The assumed zero- field value is measured by the Hall effect. Arrows indicate the filling factors. Full circles are the experimental CR values. After Ref. [13]. Phys. Status Solidi B 251, No. 2 (2014) 257 www.pss-b.com ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Download 1.56 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling