Temperature dependence of excitonic transitions in AlxGa1−xAs/GaAs quantum wells

Photoluminescence properties of a Si doped InGaAs/InGaAlAs superlattice

“…As frequently observed for several semiconductor materials and semiconductor heterostructures [19,20,22,[29][30][31], also for the SL sample studied here, it can be noticed, by analysing the 'inset' for the low temperature region, that the Varshni model overestimates the experimental curve, the Viña model underestimates it, while the Pässler models gives the best adjustments.…”

“…Using the data obtained by Grilli et al [28] for the GaAs bulk crystal, Pässler [22] observed that the Varshni and Viña curves are almost indistinguishable for temperatures higher than 100 K. However, for T < 80 K there is a clear difference between these curves; in this range, especially for T < 40 K, the model proposed by Varshni shows a much faster shrinkage (and stronger curvature) than the experimental curve, overestimating the energy gap for T = 0 K. On the other hand, the curve obtained through the expression proposed by Viña shows, in this range, a much slower shrinkage than the curve given by the experimental points, disappearing completely at T < 20 K, representing a plateau behaviour and underestimating the energy gap for T = 0 K. These results, as well as the better efficiency of the model proposed by Pässler (p-type model, equation (3)), have been frequently observed for many semiconductor materials and structures of semiconductor materials [19,20,22,[29][30][31].…”

Dispersion-related description of temperature dependencies of band gaps in semiconductors

“…͑1͒ to numerical fittings of temperature dependencies of exciton line positions measured in ternary compounds 17,18 and quantum well structures. [19][20][21][22][23] However, a detailed assessment of the parameter sets associated with a variety of least-mean-square fittings 5,[17][18][19][20][21][22][23] using Eq. ͑1͒ shows that, due to the relative simple analytical structure ͑the approximate nature͒ of this model function, the fitted parameter values ⌰ p and p often do not yield good values for the two relevant lowest-order moments, 6 ͗ប͘ ϵk B ⌰ and ͗(ប) 2 ͘ϵ(1ϩ⌬ 2 )(k B ⌰) 2 .…”

The effect of confinement on the temperature dependence of the excitonic transition energy in GaAs/AlxGa1−xAs quantum wells

“…The values of the fitting parameters associated to different models describing the behavior of E g (T ) decrease with the introduction of the barriers but increase as the Al concentration increases [18]. This decrease of the values of the fitting parameters with increasing T was attributed to the transition from the three-dimensional (3D) to the two-dimensional (2D) confinement condition [16,17]. The same trend is also observed in the behavior of the parameter related to the contribution of exciton-acoustic phonon interaction used to fit experimentally measured line widths by resonant Raman scattering, absorption, and photoluminescence (PL) spectroscopy [19][20][21][22][23][24][25].…”

“…In Al x Ga 1−x As bulk material, the adjustable parameters of different models describing E g (T )-especially the set of parameters (α i , i ) where α i ≡ −(dE/dT ) T →∞ ≡ S(∞) is the high-temperature limiting value of the forbidden gap entropy and i is the phonon energy in units of the absolute temperature-are dependent on the aluminum concentration (x) showing an increasing behavior with increase of Al concentration in the composition range 0 < x < 0.45 [14,15]. As far as GaAs/Al x Ga 1−x As quantum well (QW) heterostructures are concerned, Lourenc ¸o et al [16,17] verified that the aluminum concentration of the barrier material has influence on the temperature dependence of the excitonic transition energy. The values of the fitting parameters associated to different models describing the behavior of E g (T ) decrease with the introduction of the barriers but increase as the Al concentration increases [18].…”