AuGe surface plasmon enhances photoluminescence of the InAs/GaAs bilayer quantum dot heterostructure

Abstract

We report an improvement in the photoluminescence of a GaAs-capped InAs/GaAs bilayer quantum dot (QD) heterostructure by AuGe nanoparticle deposition on the surface of a thin capped layer. Scanning electron microscopy confirmed the formation of AuGe nanoparticles on the surface at temperatures ranging from 300 to 700 °C. Optical absorption spectroscopy revealed the plasmon resonance peak of AuGe nanoparticles at around 670 nm for the sample annealed at 300 °C, confirming the presence of the plasmonic effect. Raman spectroscopy revealed a QD phonon peak at ∼238.5 cm⁻¹ for the sample annealed at 300 °C, indicating InAs QDs in the heterostructure. Compared to the uncovered sample, enhancements were observed in the PL spectra of the AuGe-deposited samples annealed at 300 °C and 400 °C (with enhancement factors of 2.58 and 2.18, respectively). The observed enhancement is attributed to photon trapping by scattering from the cross section of the dipole radiation field. Increasing the annealing temperature from 300 °C to 700 °C blue-shifted the photoluminescence peaks at 18 K because of In/Ga inter-diffusion. A decrease in activation energy was observed with the increase in annealing temperature from 300 °C to 700 °C, attributed to poor confinement potential and high electron concentration at the sample surface. Our findings contribute to the realization of high-efficiency plasmonic-based InAs QD detectors for optical communication in the 1300 nm optical window.

---

1. Introduction

Self-organized InAs quantum dot (QD) heterostructures grown on GaAs substrates in the Stranski–Krastanov (SK) mode have been extensively studied because of their superior performance over quantum well based devices in optical communication.^1–4 Mostly, these InAs QDs devices are optical to electrical signal conversion detectors. The size distribution of SK InAs QDs, however, is not uniform, producing low efficiency devices. The use of an InAs/GaAs bilayer QDs heterostructure with a strain-reducing capping layer has been suggested as a solution to the problem.^5,6 In this structure, two consecutive layers of InAs QDs are separated by a GaAs spacer layer. The bottom (seed) InAs layer generates strain-induced vertical coupling of QDs between two layers, which helps in the growth of large dots with uniform size distribution in the top (active) InAs layer.^5,6 However, GaAs capping strongly compresses the InAs QDs formed in the SK growth mode, causing a reduction in device efficiency and an undesirable blue shift of the photoluminescence (PL) maximum to 1200 nm—instead of the 1300 nm wavelength required to minimize losses in optical waveguides.

Improvement in device efficiency can be achieved by exploiting surface plasmon resonance in metals, which increases light trapping or light scattering in the device.^7–11 Plasmonic enhancements have been reported to improve the PL of various semiconductor materials^12–16 and detector photocurrent or photoresponse.^17–22 Zhou et al. improved the PL of GaAs by simply coating a thin layer of Au nanoparticles on its surface.¹⁶ Furthering this line of research, we propose AuGe as an alternative to Au nanoparticles for plasmonic-based IR detectors because of its longer-wavelength resonance peak and higher-and-constant absorbance for the wavelength range of 1000–1500. In addition, we propose using a thin capping layer to reduce the blue shifting.

In this paper, we report the effects of AuGe nanoparticle formation on the properties of bilayer InAs/GaAs QDs capped with a thin GaAs layer. In particular, we investigate PL as a function of the annealing temperature of AuGe nanoparticles. We also examine the vibration properties of the samples. These studies will be helpful to gain a detailed understanding of plasmonic based InAs QD detectors.

2. Experimental procedure

Self-assembled InAs/GaAs bilayer QD heterostructure samples were grown by solid source molecular beam epitaxy (MBE) on a GaAs substrate (see Fig. 1). The seed and active InAs layers were grown to thicknesses of 2.5 and 3.2 monolayers (MLs), respectively. InAs was deposited at a slow growth rate of 0.028 ML per s to achieve a dense population of coherent islands. The seed layer was grown at 500 °C and the active layer was grown at a lower temperature of 460 °C to reduce the degree of In/Ga intermixing during capping.²³ The thickness of the GaAs spacer layer was fixed at 7.5 nm, and a 15 nm-thick GaAs cap layer was deposited on the active QD layer. The GaAs cap and spacer layers were grown at 575 °C to remove intrinsic defects caused by in situ annealing on the QDs. Next, a 10 nm-thick AuGe film was deposited on the cap layer of bilayer QD heterostructure by electron beam evaporation. Finally, AuGe were annealed on the samples at temperatures ranging from 300 to 700 °C by rapid thermal annealing to create AuGe nanoparticles. The annealing was performed with GaAs proximity capping to prevent degradation of sample quality by arsenic-out diffusion from the surface.²⁴

Fig. 1 Schematic illustration of the InAs/GaAs bilayer QD heterostructure with AuGe-deposited nanoparticles on the surface.

The surface morphology of the AuGe nanoparticles was observed using field emission scanning electron microscopy (FE-SEM, Raith-150 two e-beam system). The plasmon resonance peak was measured by optical absorption spectroscopy using a UV-Vis-NIR spectrometer (Lambda 750). The vibrational properties of the bilayer QDs resulting from AuGe nanoparticle deposition and annealing were investigated through Raman spectroscopy using an Ar + laser with an excitation wavelength of 514.5 nm. Temperature-dependent (18–300 K) PL studies of the annealed samples were performed using a 532 nm green diode-pumped solid-state (DPSS) continuous wave laser.

3. Results and discussion

Surface morphology

FE-SEM images of AuGe nanoparticle films were taken before and after annealing in the temperature range of 300–700 °C in an Ar atmosphere (see Fig. 2). No AuGe nanoparticles were found on the AuGe deposited and un-annealed film. AuGe nanoparticles are distributed uniformly on annealing and their shapes are close to spherical. Uniformly distributed and dense particles are observed for the 300 °C sample. The nanoparticles of average size are less dense at 500 °C. The distribution of the nanoparticles becomes less uniform with a further increase in temperature to 600 °C, and small trenches are observed at the GaAs surface. The trenches are also observed for the sample annealed at 700 °C. It is caused because some As desorption from the film at higher temperatures.

Fig. 2 FE-SEM images of (a) un-annealed, (b) 300 °C, (c) 400 °C, (d) 500 °C, (e) 600 °C and (f) 700 °C annealed AuGe-deposited samples.

Plasmon resonance peak

Fig. 3 shows the optical absorption spectra of AuGe and Au nanoparticles prepared on a sapphire substrates by annealing of AuGe and Au films, each of 10 nm thickness, at temperatures of 300 °C and 500 °C respectively. The resonance peaks of the AuGe and Au nanoparticles are located at around 670 nm and 535 nm, respectively. The AuGe nanoparticles are more suitable than Au nanoparticles for plasmonic based IR detectors due to longer wavelength resonance peak of AuGe nanoparticles. Additionally, because GaAs has a high refractive index, the localized surface plasmon resonance peak is expected to shift to a longer wavelength. The excitonic transitions range in bilayer InAs QD is 1200–1300 nm, indicating the advantage of using AuGe nanoparticles in comparison to Au. Moreover, AuGe nanoparticle produce higher-and-constant absorbance in comparison to Au nanoparticles for wavelength range 1000–1500 nm, indicating its usefulness for plasmonic-based IR detector application.²²

Fig. 3 Room temperature absorbance spectra of AuGe and Au nanoparticles prepared on sapphire substrates.

Vibrational properties

Fig. 4 shows the Raman spectra of uncovered sample and the samples with AuGe nanoparticles annealed at different temperatures. All samples showed an intense peak around 293 cm⁻¹, attributed to the longitudinal-optical (LO) phonon peak of the GaAs matrix.²⁵ A weak peak around 272 cm⁻¹ was also observed in samples annealed at 300–700 °C, attributed to the GaAs transverse-optical (TO) phonon peak.²⁶ The TO phonon peak intensity and FWHM increased with the annealing temperature from 300 °C to 700 °C. The presence of TO phonon peak indicates structural disorder in the GaAs cap layer reported for GaAs-based QD heterostructures.²⁷ The sample annealed at 700 °C showed high TO phonon peak intensity owing to the rather high In/Ga inter-diffusion in addition to redistribution of the InAs material between islands of the active layer. The FE-SEM image also confirmed this structural disorder. Besides these two peaks, additional peaks were observed at approximately 238.5, 242, 241, and 239.5 cm⁻¹ for the uncovered sample, annealed at 300 °C, 400 °C, and 500 °C, respectively, attributed to InAs QDs in the heterostructure.^27,28 No QD phonon peak was observed for the sample annealed at 700 °C, owing to the redistribution and dissolution of the InAs material with the spacer/cap layer. Thus, QD phonon frequency peaks resulting from InAs QD were confirmed.

Fig. 4 Raman spectra of self-assembled InAs/GaAs bilayer QD heterostructures for the uncovered sample, and AuGe-deposited samples annealed at different temperatures.

Photoluminescence

Fig. 5(a) shows the PL spectra at 18 K for the uncovered sample and the samples with AuGe nanoparticles annealed at different temperatures. The uncovered sample has two visible peaks: a long-wavelength peak at around 1308 nm attributed to a ground state (GS) emission corresponding to the 3.2-ML active layer QDs and a short-wavelength peak at around 1234 nm attributed to first excited state transitions originating from the 3.2-ML active layer QDs. Compared to the uncovered sample, enhancements were observed in the PL spectra for the AuGe-deposited samples annealed at 300 °C and 400 °C. The highest increment of PL intensity, compared to the uncovered sample, was observed for the sample annealed at 300 °C for a peak at ∼1301 nm. The mentioned PL results are average of five different runs in PL measurement system. The PL intensity is showing fluctuation of ±0.05% in five different runs and showing negligible fluctuation in terms of wavelength.

Fig. 5 (a) PL spectra for the uncovered sample and AuGe-deposited samples annealed at various temperatures. (b) Laser excitation power dependent PL of the sample annealed at 300 °C (c). Temperature-dependent PL spectra of the sample annealed at 300 °C. (d) Variation of activation energy (calculated from Arrhenius plot).

The observed enhancement is attributed to photon trapping by scattering from the cross section of the dipole radiation field, which would be an order of magnitude larger than the physical cross section of the AuGe nanoparticles.^7,29 These studies indicate that the dipole radiation pattern significantly penetrated the semiconductor material. The improvement in PL intensity by 532 nm excitation indicates significant scattering of photons from the surface plasmon generated in the AuGe nanoparticles, concurring with the PL enhancement reported by Pfeiffer and colleagues for GaAs QD.¹² In their study, the maximum PL enhancement was observed for laser excitation wavelengths higher than the surface plasmon absorption peak wavelength. However, in the present work, PL enhancement was observed for an excitation wavelength lower than the absorption peak wavelength.

The PL peak position is blue-shifted with an increase in annealing temperature from 300 °C to 700 °C. In these bilayer QD samples, which have a thin spacer layer (7.5 nm), the active layer QDs are in a larger strain state owing to improper propagation of the strain field from the seed layers.^30,31 This results from desorption of indium through the thin spacer layer and a consequent reduction in the effective volume of the buried seed QDs while growing the active layer QDs. In our samples, the strained active islands assist in material diffusion during annealing. This accounts for the degradation of the material quality of samples due to In/Ga inter-diffusion and leads to a blue shift of the emission wavelength of the samples with increasing annealing temperature. We observed a gradual blue shift from around 1308 nm (uncovered sample) to around 1287 nm corresponding to the GS emission wavelength of the 3.2-ML active layer QDs for samples annealed at temperatures of up to 600 °C, as shown in Fig. 4(a). For the sample annealed at 700 °C, a high blue-shifted broad peak was observed, which indicates high In/Ga inter-diffusion in addition to redistribution of the InAs material between the islands of the active layer.^30,31 The improvement in the PL of the uncapped InAs bilayer QDs was not investigated in this study, because uncapped QDs exhibit extremely weak emission.

The enhancement of the PL emission can be understood using the Purcell enhancement factor F,³² which can be expressed in terms of the surface-plasmon-induced spontaneous recombination rate (R_p), the radiative recombination rate (R_r), and the nonradiative recombination rate (R_n) as

The above approximation is obtained by considering R_n ≪ R_p and R_n ≪ R_r. Alternatively, the ratio R_p/R_r can be expressed in terms of the PL intensity as R_p/R_r = I_p/I_o, where I_p is the peak intensity of the plasmonic-enhanced PL and I_o is the peak intensity of the PL spectrum before the addition of nanoparticles to the sample (uncovered sample).³³ Use of this equation gave F values of 2.58 and 2.18 for samples annealed at 300 °C and 400 °C, respectively.

To determine the origin of different peaks, laser-excitation power dependent PL of the sample annealed at 300 °C was investigated (see Fig. 5(b)). The PL spectra at 18 K were collected for laser powers of 0.5, 1, 5, 15, and 25 mW. Three peaks at ∼1301 nm, ∼1232 nm and ∼1098 nm can be observed in the 18 K PL spectra when the excitation power is increased to 25 mW. The PL intensity decreases as the excitation power is reduced. Only the first peak at 1301 nm remains when the excitation power is reduced to 0.5 mW. This peak at 1301 nm can be interpreted as the optical transition between the ground states of the 3.2-ML QD active layer. The second peak at 1232 nm starts appearing at 15 mW when the excitation power is increased from 0.5 to 25 mW, indicating the origin of transitions between the first excited states in the 3.2-ML QDs. The third peak at 1098 nm has been clearly observed at the highest excitation power of 25 mW. The origin of the third peak is attributed to ground states transitions of the 2.5-ML QDs seed layer.

To investigate the carrier confinement potential of the uncovered and annealed AuGe-deposited samples, we performed temperature-dependent PL measurements for the sample annealed at 300 °C (see Fig. 5(c)). We observed a shift in the emission peak toward a longer wavelength with the increase in ambient temperature, indicating a change in the band gap of the InAs QD layer with temperature. Variation of the integrated PL intensity with temperature can be expressed using the well-known equation:³⁴

where I_o is the integrated PL intensity at 18 K, C is the ratio of the thermal escape rate to the radiative recombination rate, E_A is the activation energy, and k is the Boltzmann constant. The graph of activation energy calculated from Arrhenius plots of the uncovered and AuGe-deposited annealed samples are shown in Fig. 5(d). The calculated activation energy is 256 meV, 184 meV, 177 meV, 146 meV, 137 meV and 109 meV for uncovered sample and the samples annealed at 300 °C, 400 °C, 500 °C, 600 °C, and 700 °C respectively. The decrease of E_A with an increase in annealing temperature has been attributed to the poor confinement potential caused by In/Ga inter-diffusion at the dot/barrier layer interface.³⁰ Another reason for decreased activation energy might be the presence a higher electron concentration at the sample surface for the annealed AuGe-deposited samples. Even at low temperature annealing such as 300 °C, there is a large difference in the activation energies of the uncovered sample and annealed sample. There is very less In/Ga inter-diffusion at 300 °C, indicating high electron concentration is responsible for the difference in activation energy at this particular temperature.

4. Conclusion

To improve the performance of InAs/GaAs QD detectors for optical communication applications, we investigated the effect of AuGe nanoparticles formation on properties of bilayer InAs/GaAs QDs capped with a thin GaAs layer. An enhancement in photoluminescence is reported for InAs/GaAs bilayer QDs coupled to AuGe nanoparticles. Optical absorption spectroscopy illustrates the plasmon resonance peak of AuGe nanoparticles at around 670 nm for the sample annealed at 300 °C, confirming the presence of plasmonic feature. Clear AuGe nanoparticles images were also produced by FE-SEM measurement. Raman spectroscopy demonstrates QD phonon peak in the spectra, indicating InAs QDs in the structure. The highest increment of PL intensity was observed for the sample annealed sample at 300 °C (enhancement factor of 2.58) for the peak at ∼1301 nm. The observed enhancement is interpreted in terms of photon scattering from the cross section of the large dipole radiation field. The calculated activation energy results indicate a higher electron concentration at surface of 300 °C annealed sample. Our results of AuGe nanoparticles study will help to realize plasmonic-based high-performance InAs/GaAs QD detectors for optical communication applications.

Acknowledgements

We acknowledge the Department of Science and Technology, Government of India for its financial support. Partial funding was received from (i) the Ministry of Communications & Information Technology, Government of India, through the Centre of Excellence in Nanoelectronics, IIT Bombay and (ii) Riber France. We are also very thankful to SERI DST, India along with others for their support.

References

1. J. Philips, P. Bhattacharya, S. W. Kennerly, D. W. Beekman and M. Dutta, IEEE J. Quantum Electron., 1999, 35, 936–943.
2. D. Pan, E. Towe and S. Kennerly, Appl. Phys. Lett., 1998, 73, 1937–1939.
3. D. L. Huffaker, G. Park, Z. Zou, O. B. Shchekin and D. G. Deppe, Appl. Phys. Lett., 1998, 73, 2564–2566.
4. A. R. Kovsh, A. E. Zhukov, D. A. Livshits, A. Y. Egorov, V. M. Ustinov, M. V. Maximov, Y. G. Musikhin, N. N. Ledentsov, P. S. Kop'ev, Z. I. Alferov and D. Bimberg, Electron. Lett., 1999, 35, 1161–1163.
5. P. Howe, B. Abbey, E. C. Le Ru, R. Murray and T. S. Jones, Thin Solid Films, 2004, 464, 225–228.
6. Y. I. Mazur, Z. M. Wang, G. G. Tarasov, V. P. Kunets, G. J. Salamo, Z. Y. Zhuchenko and H. Kissel, J. Appl. Phys., 2005, 98, 053515.
7. J. Lee, P. Hernandez, J. Lee, A. O. Govorov and N. A. Kotov, Nat. Mater., 2007, 6, 291–295.
8. K. R. Catchpole and A. Polman, Opt. Express, 2008, 16, 21793–21800.
9. C. Wu, C. He, H. Lee, H. Chen and S. Gwo, J. Phys. Chem. C, 2010, 114, 12987–12993.
10. K. Y. Yang, K. C. Choi and C. W. Ahn, Appl. Phys. Lett., 2009, 94, 173301.
11. S. Pillai, K. R. Catchpole, T. Trupke and M. A. Green, J. Appl. Phys., 2007, 101, 093105.
12. M. Pfeiffer, K. Lindfors, C. Wolpert, P. Atkinson, M. Benyoucef and A. Rastelli, et al., Nano Lett., 2010, 10, 4555–4558.
13. J. Lin, A. Mohammadizia, A. Neogi, H. Morkoc and M. Ohtsu, Appl. Phys. Lett., 2010, 97, 221104.
14. J. S. Biteen, L. A. Sweatlock, H. Mertens, N. S. Lewis, A. Polman and H. A. Atwater, J. Phys. Chem. C, 2007, 111, 13372–13377.
15. Y. Lu, C. Chen, K. Shen, D. Yeh, T. Tang and C. C. Yang, Appl. Phys. Lett., 2007, 91, 183107.
16. X. Zhou, Q. Wei, L. Wang, B. Joshi, Q. Wei and K. Sun, Appl. Phys. A, 2009, 96, 637–641.
17. S. Nunomura, A. Minowa, H. Sai and M. Mie Kondo, Appl. Phys. Lett., 2010, 97, 063507.
18. D. Derkacs, S. H. Lim, P. Matheu, W. Mar and E. T. Yu, Appl. Phys. Lett., 2006, 89, 093103.
19. I. M. Pryce, D. D. Koleske, A. J. Fischer and H. A. Atwater, Appl. Phys. Lett., 2010, 96, 153501.
20. V. E. Ferry, M. A. Verschuuren, H. B. T. Li, E. Verhagen, R. J. Walters, R. E. I. Schropp, A. A. Harry and P. Albert, Opt. Express, 2010, 18, A237–A245.
21. W. Liu, X. Wang, Y. Li, Z. Geng, F. Yang and J. Li, Sol. Energy Mater. Sol. Cells, 2011, 95, 693–698.
22. S. A. Maier, Plasmonics: Fundamentals and Applications, Springer, New York, 2007.
23. E. C. Le Ru, P. Howe, T. S. Jones and R. Murray, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 67, 165303.
24. R. Leon, Y. Kim, C. Jagadish, M. Gal, J. Zou and D. Cockayne, Appl. Phys. Lett., 1996, 69, 1888–1890.
25. V. Swaminathan and A. T. Macramder, Material Aspects of GaAs and InP Based Structures, Prentice-Hall, Englewood Cliffs, 1991.
26. S. Reich, A. R. Goni, C. Thomsen, F. Heinrichsdorff, A. Krost and D. Bimberg, Phys. Status Solidi B, 1999, 215, 419–424.
27. J. Ibáñez, A. Patane, M. Henini, L. Eaves, S. Hernandez, R. Cusco, L. Artus, Y. G. Musikhin and P. N. Brounkov, Appl. Phys. Lett., 2003, 83, 3069–3071.
28. L. Artus, R. Cusco, S. Hernandez, A. Patane, A. Polimeni, M. Henini and L. Eaves, Appl. Phys. Lett., 2000, 77, 3556–3558.
29. H. Benisty, R. Stanley and M. Mayer, J. Opt. Soc. Am. A, 1998, 15, 1192–1201.
30. S. Chakrabarti, N. Halder, S. Sengupta, J. Charthad, S. Ghosh and C. R. Stanley, Nanotechnology, 2008, 19, 505704.
31. S. Chakrabarti, N. Halder, S. Sengupta, J. Charthad, S. Ghosh and C. R. Stanley, J. Nanoelectron. Optoelectron., 2008, 3, 277–280.
32. E. M. Purcell, Phys. Rev., 1946, 69, 681.
33. J. Kümmerlen, A. Leitner, H. Brunner, F. R. Aussenegg and A. Wokaun, Mol. Phys., 1993, 80, 1031–1046.
34. F. Y. Tsai and C. P. Lee, J. Appl. Phys., 1998, 84, 2624–2627.

---