Temperature Dependent Photoluminescence. (PL) properties of InAs/InP quantum dashes

Transcription

1 Copyright 2011 American Scientific Publishers All rights reserved Printed in the United States of America Journal of Nanoscience and Nanotechnology Vol. 11, 1 5, 2011 Temperature Dependent Photoluminescence Properties of InAs/InP Quantum Dashes Subjected to Low Energy Phosphorous Ion Implantation and Subsequent Annealing M. H. Hadj Alouane 1, B. Ilahi 1, H. Maaref 1, B. Salem 2, V. Aimez 3, D. Morris 3, and M. Gendry 4 1 Laboratoire de Micro-Optoélectronique et Nanostructures, Université de Monastir, Faculté des Sciences, Université de Monastir, 19, Tunisia 2 Laboratoire des Technologies de la Microélectronique, LTM-CNRS/CEA-Grenoble, 17, rue des martyrs, F Grenoble cedex 9, France 3 Centre de Recherche en Nanofabrication et Nanocaractérisation (CRN2), Université de Sherbrooke, (Québec) Canada J1K 2R1 4 Université de Lyon, Institut des Nanotechnologies de Lyon (INL), UMR CNRS 5270, Ecole Centrale de Lyon, 36 Avenue Guy de Collongue, F Ecully, France We report on the impact of phosphorous ion-implantation-induced band gap tuning on the temperature dependent photoluminescence (PL) properties of InAs/InP quantum dashes (QDas). The high temperature range carriers activation energy, extracted from Arrhenius plots, is found to decrease from 238 to 42 mev when the ion implantation dose increases from cm 2 to cm 2, which is consistent with the observed emission energy blueshift increase with increasing the ion implantation doses. This effect is attributed to the As/P exchange which reduces the carrier confining potential depth. For intermediate ion implantation doses the reduced carrier confining potential barrier combined with the non-uniform intermixing process, that causes an increased QDas size dispersion, result in anomalous temperature-dependent PL properties. Indeed, the temperature induced PL emission energy redshift measured between 10 K and 300 K is found to be strongly affected by the carrier redistribution within the broadened localized QDas states. Keywords: InAs/InP, Quantum Dashes, Selective Intermixing, Ion Implantation, Photoluminescence, Thermal Activation. 1. INTRODUCTION Self assembled quantum dots (QDs) have attracted a great deal of interest from their unique physical properties and potential device applications. 1 QDs based laser diodes are expected to have less temperature sensitivity and a lower threshold current density than those with quantum well active layer. 2 3 Furthermore, selective post growth tuning of QD emission energy is highly required for monolithic integration of QD-based optoelectronic devices. 4 It can be ensured by spatial selective intermixing across the same sample surface. Among the existing intermixing techniques, low energy-ion-implantation and subsequent Author to whom correspondence should be addressed. annealing is the most suitable process allowing: reproducibility, area selectivity and precise control of the defects depth and concentration. Extensive reports already exist on the effect of intermixing on the low temperature luminescence peak position, linewidth and integrated intensity. 5 9 However, there are still some unknowns concerning the evolution of the temperature dependent photoluminescence properties as a function of the intermixing degree. To fabricate efficient optoelectronic device operating at room temperature, a clear understanding of the effect of temperature on the optical properties is crucial. In this work, we study the temperature dependent photoluminescence (PL) properties of InAs/InP quantum dashes (QDas) selectively intermixed by phosphorus ion implantation and subsequent rapid thermal annealing (RTA). J. Nanosci. Nanotechnol. 2011, Vol. 11, No. xx /2011/11/001/005 doi: /jnn

2 Temperature Dependent Photoluminescence Properties of InAs/InP Alouane et al. 2. EXPERIMENTAL DETAILS The InAs QDas were grown at 520 C with a 4 monolayers (ML) nominal thickness of InAs on sulfur-doped InP substrate using solid-source molecular beam epitaxy. 10 The QDas were then covered with a 70 nm-thick InP layer followed by a nm-thick In 53 Ga 47 As layer used as a sacrificial layer. Structural studies indicate that the QDas are elongated along the 1 10 direction with a surface density around cm 2. These QDas exhibit a structured pyramidal shape with typical base widths of 22 ± 1.2 nm and very homogenous high of 2.4 nm. There lengths are in the 100 nm range. The samples were then subjected to a phosphorus ion implantation, at 18 kev, at 200 C with doses ranging from to cm 2. 7 The energy of phosphorus ions was chosen to confine the damage peak into the InGaAs sacrificial layer that can be selectively removed after the rapid thermal annealing (RTA) process. After implantation, the samples were annealed at 6 C for 120 s to set off the atomic intermixing. For comparison, a reference sample (only annealed) was also prepared. For the PL measurements, the samples were excited with the nm line of an Ar + laser and the spectra were collected using a thermoelectrically cooled InGaAs photodetector using a conventional lock-in technique. 3. RESULTS AND DISCUSSION The 10 K PL spectra of the as-grown sample, the reference sample and the samples implanted at various doses and subsequently annealed are depicted in the Figure 1. The presence of sharp water-vapor absorption bands between 0.85 and 0.91 ev are artifacts of our detection system and unfortunately tends to deform the PL emission band in that spectral range. The manifestation of multiple peaks arises from the state filling due to the three dimensional carrier confinement. 11 A blueshift of the emission energies and fluctuation of the PL linewidth have been found to occur when the ion implantation doses are increased from to ions cm 2. Higher implantation doses result in a saturation of the blueshift and a decrease of the integrated PL intensity. 7 Furthermore, the QDas are also found to be inhomogeneously intermixed for the intermediate ion implantation doses resulting in an increase of the PL linewidth. 7 The increase of the QDas intermixing degree and consequent lowering of the carrier s confining potential depth together with the broadened localized states distribution has recently been found to induce atypical evolution of the QDas temperature dependent PL linewidth and emission energy. 12 The observed behavior has been correlated with a temperature dependent competing effect between electron phonon scattering and thermally activated carriers transfer between inhomogeneously intermixed QDas. The integrated PL intensity variation as a function of temperature for the As-grown and the implanted QDas Normalized PL intensity (arb. units) Energy (ev) cm cm 2 Reference As-grown Fig K-PL spectra of InAs/InP QDas samples as a function of the ion implantation dose. The spectra are normalized and shifted for clarity. samples is shown by the Figure 2(a). Indeed, the temperature dependence of the PL intensity changes significantly after the intermixing process. A fit of the experimental data can be performed in the high temperature range using the following Arrhenius equation: I I = 0 (1) 1 + C exp E a / k b T Where k b is the Boltzmann constant, E a is the thermal activation energy, I 0 is the PL intensity at 0KandC is a constant. The observed thermal quenching of the PL at that temperature range is mainly related to the escape process of the dissociated excitons (electron-hole pairs) into the corresponding barriers In this situation, the activation energy E a, extracted from Arrhenius plots, approximates the differences between the QDas emission energy and that of the wetting layer or barrier material. 16 For the As-grown sample, the derived activation energy, around 238 mev, is close to the energy difference between the ground state emission energy of the QDas and that of the wetting layer. 17 The observed PL emission energy blueshift with increasing the intermixing degree is commonly interpreted in terms of QDas compositional change and consequent effective 2 J. Nanosci. Nanotechnol. 11, 1 5, 2011

3 Alouane et al. Temperature Dependent Photoluminescence Properties of InAs/InP (a) Integrated PL intensity log scale (arb. units) Reference cm 2 E a = 238 mev E a = 225 mev 5 E a = 205 mev E a = 165 mev cm 2 E a = 210 mev E a = 140 mev 5 5 E 2 = 100 mev E 2 = 85 mev E 2 = 42 mev /KT (mev 1 ) (b) Emission energy (ev) E11 1E12 1E13 1E14 1E15 Lon dose (cm 2 ) Activation energy (mev) Fig. 2. (a) The Arrhenius plot of the integrated luminescence intensity of all the investigated samples, showing the activation energy dependence on the intermixing degree. Symbols are experimental data and the lines are the corresponding fit. (b) Activation energy and 10 K emission energy blueshift as a function of the ion implantation dose. band gap enhancement. 5 9 Consequently, the carrier s effective barrier height that can be qualitatively linked to the extracted activation energy should decrease with increasing the intermixing degree. Accordingly, increasing the intermixing degree, results in a continual decrease of the activation energy from 238 to 42 mev. For more quantitative analyses, the energy of the emission band maximum blueshift and activation energy are depicted in the Figure 2(b) as a function of the ion implantation dose. Increasing the ion doses leads to a simultaneous decrease of the activation energy (barrier height) and increase of the emission energy blueshift. Surprisingly, the emission energy blueshift saturate for higher doses ( ions cm 2 ) while the activation energy decreases down to 42 mev for implantation dose of ions cm 2. This activation energy can rather be correlated to the thermally activated capture of excitons by ion implantation induced nonradiative defects. This is in agreement with previously reported drastic decrease of the 10 K PL intensity for higher ion doses. 7 Additionally, the dissolution of the intermixed QDas to a rough 2D layer at those doses becomes also very probable. J. Nanosci. Nanotechnol. 11, 1 5,

4 Temperature Dependent Photoluminescence Properties of InAs/InP Alouane et al. For implantation doses below ions cm 2, the inhomogeneous intermixing enhanced QDas size distribution, together with the reduced carriers confining potential barrier are found to be responsible for atypical temperature dependence PL linewidth and emission energy. 12 As shown by the Figure 3(a), where the energy of the emission band maximum are plotted versus temperature, for each sample. Except for samples showing a very large PL linewidth, the data points are well reproduced using (a) Emission energy (ev) (b) ΔE (mev) cm 2 Only-annealed As-grown cm Temperature (K) E11 1E12 1E13 1E14 1E15 Lon dose (cm 2 ) Fig. 3. (a) Temperature dependence of the PL peak energy of the investigated samples. The symbols are experimental data and the solid lines are calculated according to the Varshni law. (b) Energy redshift in the K temperature range and low temperature FWHM of the PL spectrum of the InAs/InP QDas implanted with various phosphorus ion doses K PL linewidth (mev) empirical Varshni relation: 18 E g T = E g 0 T 2 + T With and parameters approaching those found for the InAs material. 19 Furthermore, for samples implanted at the highest doses, a discrepancy between experimental data and empirical curves has been found to occur at low temperature range (up to 160 K). This large energy difference might be related to an extra confinement of the excitons. In our case, the high implantation doses and consequent high defects density are expected to create a high density of point defects that may agglomerate into complex defects and stable clusters becoming hard to restore after annealing. This would further confirm the attribution of the lowest activation energy to the thermally activated capture of excitons by nonradiative defects. However, for samples implanted at intermediate doses, it is impossible to reproduce their data points over all the temperature range using a simple empirical Varshni relation. The position of the PL peak decreases rapidly with increasing temperature, an inflexion point is observed in the K temperature range, and the curve tends to recover a normal Varshni-like behavior at higher temperatures. The fast red-shift of the PL emission energy is typical of thermally activated carrier redistribution from smaller QDas having small confining potential to larger QDas with deeper confining potential. Accordingly, as shown by the Figure 3(b) where, we have plotted the energy of the emission band maximum redshift ( E) between 10 K and 300 K as well as the evolution of the low temperature PL linewidth (qualitatively traducing the QDas size dispersion) as a function of the phosphorus ion implantation dose. E is found to increase with increasing the ion implantation doses up to cm 2. The sample implanted at ion cm 2 has the largest value of temperature enhanced redshift with largest value of the 10 K PL linewidth. However, for higher implantation doses, E decreases with the reduction of the PL linewidth traducing an improved QDas size distribution. At those doses, the intermixing process tends to improve the uniformity in size of the QDas resulting in the minimization of carriers redistribution and therefore to less temperature induced energy redshift. For too high implantation doses, E remains nearly constant which is consistent with emission energy blueshift saturation. 4. SUMMARY In summary, we have investigated the effects of the low energy phosphorus ion implantation and subsequent rapid thermal annealing on the temperature dependence of the PL properties. The study of the thermal activation energy reveals that the carriers confining potential depth is reduced when the implantation doses increases. 4 J. Nanosci. Nanotechnol. 11, 1 5, 2011

5 Alouane et al. For intermediate ion implantation doses, the temperature induced PL emission energy redshift measured between 10 K and 300 K is found to be strongly affected by the carrier redistribution within the broadened localized QDas states. Indeed, for intermediate ion implantation doses the reduced carrier confining potential barrier combined with the non-uniform intermixing process, that causes an increased QDas size dispersion, result in anomalous temperature-dependent PL properties. References and Notes 1. D. Bimberg, J. Phys. D 38, 2055 (2005). 2. Y. Arakawa and H. Sakaki, Appl. Phys. Lett. 40, 939 (1982). 3. G. T. Liu, A. Stinz, H. Li, K. J. Malloy, and L. F. Lester, Electron. Lett. 35, 1163 (1999). 4. H. S. Djie, Y. Wang, Y. H. Ding, D. N. Wang, J. C. M. Hwan, X. M. Fang, Y. Wu, J. M. Fastenau, A. W. K. Liu, G. T. Dang, W. H. Chang, and B. S. Ooi, IEEE Journal of Selected Topics in Quantum Electronics 14, 4 (2008). 5. P. Lever, H. H. Tan, C. Jagadish, P. Reece, and M. Gal, Appl. Phys. Lett. 82, 2053 (2003). 6. Y. Ji, W. Lu, G. Chen, X. Chen, and Q. Wang, J. Appl. Phys. 93, 2 (2003). 7. B. Salem, V. Aimez, D. Morris, A. Turala, P. Regreny, and M. Gendry, Appl. Phys. Lett. 87, (2005). 8. H. S. Djie, B. S. Ooi, and V. Aimez, Appl. Phys. Lett. 87, (2005). Temperature Dependent Photoluminescence Properties of InAs/InP 9. B. Ilahi, B. Salem, V. Aimez, L. Sfaxi, H. Maaref, and D. Morris, Nanotechnology 17, 3707 (2006). 10. M. Gendry, C. Monat, J. Brault, P. Regreny, G. Hollinger, B. Salem, G. Guillot, T. Benyattou, C. Bru-Chevallier, G. Bremond, and O. Marty, J. Appl. Phys. 95, 4761 (2004). 11. B. Salem, T. Benyattou, G. Guillot, C. Bru-Chevallier, G. Bremond, C. Monat, G. Hollinger, and M. Gendry, Phys. Rev. B 66, (2002). 12. M. H. Hadj Alouane, B. Ilahi, H. Maaref, B. Salem, V. Aimez, D. Morris, A. Turala, P. Regreny, and M. Gendry, J. Appl. Phys. 108, (2010). 13. Z. Y. Xu, Z. D. LU, X. P. Yang, Z. L. Yuan, B. Z. Zheng, J. Z. Xu, W. K. Ge, Y. Wan, and L. L. Chang, Phys. Rev. B 54, (1996). 14. R. Chen, H. Y. Liu, and H. D. Sun, J. Appl. Phys. 107, (2010). 15. B. Ilahi, L. Sfaxi, and H. Maaref, J. Luminescence 27, 741 (2007). 16. C. Lobo, R. Leon, S. Marcinkevicius, W. Yang, P. C. Sercel, X. Z. Liao, J. Zou, and D. J. H. Cockayne, Phys. Rev. B 60, (1999). 17. H. Chouiab, N. Chauvin, C. Bru-Chevallier, C. Monat, P. Regreny, and M. Gendry, Appl. Surface Sci. 253, 90 (2006). 18. Y. P. Varshni, Physica (Amsterdam) 34, 149 (1967). 19. S. Paul, J. B. Roy, and P. K. Basu, J. Appl. Phys. 69, 827 (1991). 20. V. G. Dorogan, Yu. I. Mazur, J. H. Lee, Zh. M. Wang, M. E. Ware, and G. J. Salamo, J. Appl. Phys. 104, (2008). 21. T. V. Torchynska, J. L. Casas Espinola, L. V. Borkovska, S. Ostapenko, M. Dybiec, O. Polupan, N. O. Korsunska, A. Stintz, P. G. Eliseev, and K. J. Malloy, J. Appl. Phys. 101, (2007). Received: 28 June Accepted: 22 December J. Nanosci. Nanotechnol. 11, 1 5,