• Home
  • Sitemap
  • Contact us
Article View

Research Paper

Applied Science and Convergence Technology 2016; 25(4): 81-84

Published online July 31, 2016

https://doi.org/10.5757/ASCT.2016.25.4.81

Copyright © The Korean Vacuum Society.

Photoluminescence Studies of InP/InGaP Quantum Structures Grown by a Migration Enhanced Molecular Beam Epitaxy

Il-Wook Choa, Mee-Yi Ryua,*, and Jin Dong Songb

aDepartment of Physics, Kangwon National University, Chuncheon 200-701, bCenter for Opto-Electronic Convergence Systems, Korea Institute of Science and Technology, Seoul 136-791

Correspondence to: E-mail: myryu@kangwon.ac.kr

Received: June 27, 2016; Revised: July 22, 2016; Accepted: July 23, 2016

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

InP/InGaP quantum structures (QSs) grown on GaAs substrates by a migration-enhanced molecular beam epitaxy method were studied as a function of growth temperature (T) using photoluminescence (PL) and emission-wavelength-dependent time-resolved PL (TRPL). The growth T were varied from 440°C to 520°C for the formation of InP/InGaP QSs. As growth T increases from 440°C to 520°C, the PL peak position is blue-shifted, the PL intensity increases except for the sample grown at 520°C, and the PL decay becomes fast at 10 K. Emission-wavelength-dependent TRPL results of all QS samples show that the decay times at 10 K are slightly changed, exhibiting the longest time around at the PL peak, while at high T, the decay times increase rapidly with increasing wavelength, indicating carrier relaxation from smaller QSs to larger QSs via wetting layer/barrier. InP/InGaP QS sample grown at 460°C shows the strongest PL intensity at 300 K and the longest decay time at 10 K, signifying the optimum growth T of 460°C.

Keywords: InP, Quantum structure, Photoluminescence, Time-resolved photoluminescence

Semiconductor quantum structures (QSs) such as quantum dots (QDs) and quantum dashes have been studied as a promising material systems for optoelectronic applications in light-emitting diodes, laser diodes, and optical amplifiers [13]. QD-based devices have attracted considerable interest due to their electronic and optical properties determined by their size, distribution, shape, and density, which are strongly dependent on the various growth parameters such as growth rates, growth temperature, and growth cycles etc. [48] Ugur et al. [4] presented the density and ordering of InP QDs controlled by growth temperature and growth rates. The QDs formed by a migration-enhanced epitaxy (MEE) method, which alternatively supplies sources followed with a growth interruption between source depositions have attracted attention due to the possibility of control in uniformity and alignment of QDs [79]. Cho et al. [9] demonstrated that the size, aspect ratio, and uniformity of InAs QDs could be enhanced by using the MEE method.

In this paper, we report the luminescence properties of InP/ InGaP quantum structures (QSs; quantum dots+ quantum dashes) grown on GaAs substrates by using a migration-enhanced molecular beam epitaxy (MBE) with varying growth temperature (T) from 440°C to 520°C. Photoluminescence (PL) and time-resolved PL (TRPL) have been performed to investigate the optical properties and carrier dynamics of InP/ InGaP QSs as a function of growth T.

Single-layer InP/InGaP QS samples were grown on semi-insulating GaAs substrates using a migration enhanced MBE. After GaAs buffer layer with thickness of 100 nm was grown on GaAs substrates at 580°C, a 50 nm-thick InGaP spacer was grown on top of GaAs buffer layer. Then InP/InGaP QSs was deposited on InGaP spacer layer at temperatures of 440°C–520°C. InP QSs were grown by using the MEE growth cycle with the repetition number of 4 which consists of 2-s In supply, 10-s growth interruption (GI), 2-s P supply, and 10-s GI. More information on the growth of InP/InGaP QSs can be found in Ref. 7.

In order to study the effect of growth temperature on the luminescence properties of InP/InGaP QSs, PL and TRPL measurements were performed at various temperatures. He-Cd laser (325 nm) as an excitation source and a CCD detector (ANDOR DV420-BU2) for collecting PL signals were used for PL measurement. The luminescence decays were measured using a time-correlated signal photon counting system. For TRPL measurement, a pulsed diode laser beam of 375 nm with pulse width of ∼90 ps was used for excitation and a micro-channel plate photomultiplier tube for signal detection.

Figure 1 shows the PL spectra of InP/InGaP QS samples taken at (a) 10 K and (b) 300 K. T440, T460, T480, and T520 used for the samples represent the MEE growth T of 440°C, 460°C, 480°C, and 520°C, respectively. As growth T increases from 440°C to 520°C, the PL peak is steadily blue-shifted from 734 to 688 nm, respectively, at 10 K as shown in Fig. 1(a). The PL intensity of T440 sample is much weak compared to that of other samples and T520 exhibits the narrowest full width half maximum. On the other hand, at 300 K, T460 shows the strongest PL intensity while the PL intensity of T520 is weaker than that of T440 as shown in Fig. 1(b). The PL peak position at 300 K is also blueshifted from 776 to 732 nm with increasing growth T from 440°C to 520°C, respectively. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) images [7] revealed that the shape of QS changes quantum dot (QD) into quantum dash (QDash) with increasing growth T, and the structure transformation from QD to QDash occurs when growth T changes from 460°C to 480°C. As a result of change of QSs, the redshift of the PL peak energy is expected with increasing growth T. However, figure 1 shows the blueshift of the PL peak with increasing growth T, which can be interpreted in the formation of InGaP QS due to intermixing of Ga in InGaP wetting layer (WL) with InP QS [1011]. The lower energy shoulder peak of T460 shown in Fig. 1(b) could be attributed to large size of QDs due to linked and/or agglomerated QDs [7].

The PL decay curves of InP/InGaP QS samples taken at the PL peak position of each sample are shown in Fig. 2 measured at 10 K. The PL decay becomes fast with increasing growth T. The decay curves are composed of two time constants, fast and slow decay components. These time constants can be obtained by fitting the decay curves with a double exponential function; I(t)=A1exp(−t/τ1)+A2exp (−t/τ2), where τ1 and τ2 are the fast and the slow decay times, respectively, and A1 and A2 are the pre-exponential constants. The fast decay time τ1 is dominant for all QS samples and attributed to the PL decay time of InP/InGaP QSs. As growth T increases from 440°C to 520°C, τ1 decreases from 0.90 to 0.74 ns while τ2 is almost constant (∼3.4 ns) as shown in the inset of Fig. 2. The slow decay time τ2 may be related to the recapture process of carriers to larger QS and/or WL [1214]. As seen in Figs. 1 and 2, T480 and T520 exhibit similar PL properties such as rapid decrease of PL intensity with increasing T and almost same decay curves, but different with those of T440 and T460. These results are consistent with AFM and SEM images.

Figure 3 shows the emission-wavelength-dependent PL decay curves of InP/InGaP QS samples measured at 10 K. The decays of both T460 and T520 become slightly fast with increasing emission wavelength as shown in Figs. 2(a) and 2(b). All QS samples show similar trend with emission wavelength. The estimated PL decay times (τ1) of all QS samples are shown in Fig. 4. The PL spectra of InP/ InGaP QS samples measured at 10 K are also presented in Fig. 4. All samples show the longest decay time around at the PL peak position and similar dependence on wavelength. The decay times increase with increasing wavelength up to around PL peak and then decrease slightly with increasing wavelength further. It is noted that the decay times of T520 are shorter than those of other three samples.

Figure 5 shows the PL decay curves of (a) T460 and (b) T520 as a function of emission wavelength measured at 140 and 80 K, respectively. Temperatures of 140 and 80 K for T460 and T520, respectively, were chosen because T460 and T520 show clear wavelength-dependent decay curves at these Ts. The PL decays of both samples become slow with increasing emission wavelength. It is noted that the PL decays at these high Ts are much more dependent on emission wavelength compared with those at 10 K shown in Fig. 3. The PL decay time τ1 (τ2) of T460 increases from 0.54 to 1.97 ns (from 4.00 to 6.56 ns) as emission wavelength increases from 736 to 772 nm, respectively. For T520 sample, the PL curves exhibit a single exponential decay at longer wavelengths (>692 nm). The decay time t1 of T520 increases from 0.64 to 2.04 with increasing wavelength from 680 to 714 nm, respectively, while t2 increases from 1.86 to 5.38 ns as wavelength increases from 680 to 694 nm. Both samples represent that the fast decay component becomes more dominant with increasing wavelength. The increase of PL decay time and enhancement of fast component with wavelength are ascribed to the redistribution of carriers which escape thermally from smaller QSs (short wavelength) and recapture to larger QSs (long wavelength) via WL/barrier. Byun et al. [14] presented the increase of the PL decay time for InP/InGaP QSs with increasing emission wavelength at 20 K.

It is important to note that the PL intensities of T440, T460, and T480 are almost constant up to 100 K and then decrease with further increasing T while that of T520 starts to decrease rapidly at higher than 60 K (not shown here). As seen in Fig. 1, the PL intensity of T460 decreases slowly compared with other samples, showing the strongest PL at 300 K. T460 sample shows the strong PL intensity at 300 K and slow decay time at 10 K, indicating the optimum temperature of 460°C for the growth of InP/InGaP QSs using a migration enhanced MBE.

The luminescence properties of InP/InGaP QSs grown at different growth T using a migration-enhanced MBE were studied by using PL and emission-wavelength-dependent TRPL. With increasing growth T from 440°C to 520°C, the PL peak energy is shifted to short wavelength (blueshift) and the PL intensity increases at 10 K while at 300 K the samples grown at 460°C and 520°C exhibit the strongest and the weakest PL intensity, respectively. The blueshift of the PL peak with increasing growth T is ascribed to the formation of InGaP QSs due to intermixing of Ga in InGaP wetting layer with InP QSs at high growth T (≥480°C). The PL decay times of all QS samples are slightly dependent on emission wavelength at 10 K and show the longest time near the PL peak while the decay times at high T (140 and 80 K) show a strong dependence on wavelength. The increase of PL decay times at high T with increasing wavelength is attributed to the relaxation of carriers from smaller QSs to larger QSs. The PL and TRPL results indicate the optimum growth T of 460°C because the sample grown at this T shows the strongest PL intensity at 300 K and the longest decay time at 10 K. It is found that the optical and structural properties of InP/InGaP QSs can be controlled by adjusting the growth temperature.

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2013R1A1A2A10058310). This study was supported by 2015 Research Grant from Kangwon National University (No. 52015395). Time-resolved photoluminescence measurements were performed at the Central Lab of Kangwon National University.

Fig. 1. PL spectra of InP/InGaP QS samples measured at (a) 10 K and (b) 300 K. T440, T460, T480, and T520 represent the QS samples grown at 440°C, 460°C, 480°C, and 520°C, respectively.
Fig. 2. PL decay curves of InP/InGaP QS samples measured at the PL peak position at 10 K. The inset shows the PL decay times and amplitudes of InP/InGaP QSs estimated by a double exponential decay function; I(t)=A1exp(?t1)+A2exp(?t2).
Fig. 3. Emission-wavelength-dependent PL decay curves of (a) T460 and (b) T520 measured at 10 K.
Fig. 4. PL decay times of T440 (up triangles), T460 (circles), T480 (down triangles), and T520 (squares) as functions of emission wavelength taken at 10 K. The normalized PL spectra (solid lines) measured at 10 K are also displayed.
Fig. 5. Emission-wavelength-dependent PL decay curves of (a) T460 and (b) T520 measured at 140 and 80 K, respectively.
  1. Hatami, F, Masselink, WT, and Harris, JS (2006). Nanotech. 17, 3703.
    CrossRef
  2. Ha, SK, Song, JD, Han, IK, Ko, DY, Kim, SY, and Lee, EH (2011). J Korean Phys Soc. 59, 3089.
    CrossRef
  3. Reithmaier, JP, Somers, A, Deubert, S, Schwertberger, R, Kaiser, W, Forchel, A, Calligaro, M, Resneau, P, Parillaud, O, Bansropun, S, Krakowski, M, Alizon, R, Hadass, D, Bilenca, A, Dery, H, Mikhelashvili, V, Eisenstein, G, Gioannini, M, Montrosset, I, Berg, TW, van der Poel, M, Mork, J, and Tromborg, B (2005). J Phys D: Appl Phys. 38, 2088.
    CrossRef
  4. Ugur, A, Hatami, F, Masselink, WT, and Cryst, J (2011). Growth. 323, 228.
    CrossRef
  5. R?del, R, Bauer, A, Kremling, S, Reitzenstein, S, H?fling, S, Kamp, M, Worschech, L, and Forchel, A (2012). Nanotech. 23, 015605.
    CrossRef
  6. Podemski, P, Kudrawiec, R, Misiewicz, J, Somers, A, Schwertberger, R, Reithmaier, JP, and Forchel, A (2006). Appl Phys Lett. 151902, 89.
  7. Kim, SY, Song, JD, Han, IK, and Kim, TW (2012). J Nanosci Nanotech. 12, 5519.
    CrossRef
  8. Oh, JW, Cho, I-W, Ryu, M-Y, and Song, JD (2015). Appl Sci Converg Tech. 24, 67.
    CrossRef
  9. Cho, NK, Ryu, SP, Song, JD, Choi, WJ, Lee, JI, and Jeon, H (2006). Appl Phys Lett. 133104, 88.
  10. Ilahi, B, Salem, B, Aimez, V, Sfaxi, L, Maaref, H, and Morris, D (2006). Nanotechnol. 17, 3707.
    CrossRef
  11. Yang, H, Kim, DJ, Colton, J, Park, T, Meyer, D, Jones, A, Thalman, S, Smith, D, Clark, K, and Brown, S (2014). Appl Surf Science. 296, 8.
    CrossRef
  12. Kim, HY, Ryu, MY, and Kim, JS (2012). J Lumine, pp. 132
  13. Karachinsky, LY, Pellegrini, S, Buller, GS, Shkolnik, AS, Gordeev, NY, Evtikhiev, VP, and Novikov, VB (2004). Appl Phys Lett. 7, 84.
  14. Byun, HR, Ryu, M-Y, Song, JD, and Lee, C-L (2015). J Korean Phys Soc. 66, 811.
    CrossRef

Share this article on :

Stats or metrics