Open Access
Add to BibSonomyAbstract
We present the mechanisms underlying the redshifted and blueshifted photoluminescence (PL) of quantum dots (QDs) upon amorphization of phase change material (PCM). We calculated the stress and energy shift distribution induced by volume expansion using finite element method. Simulation result reveals that redshift is obtained beneath the flat part of amorphous mark, while blueshift is obtained beneath the edge region of amorphous mark. Simulation result is accompanied by two experimental studies; two-dimensional PL intensity mapping of InAs/InP QD sample deposited by a layer of PCM, and an analysis on the relationship between PL intensity ratio and energy shift were performed.
© 2014 Optical Society of America
1. Introduction
Semiconductor quantum dots (QDs) have emerged as promising single photon emitters and entangled photon sources for quantum communication applications such as quantum cryptography [1,2] and quantum computing [3]. One of the main challenges for realizing these applications is precise control of energy in QDs. The three main approaches to tune emission energy are externally applied magnetic field [4,5], electric field [6,7] and mechanical strain [8,9].
Magnetic field method is believed to be the most promising approach in tuning energy in wide range. However, complex and large equipment is required to generate strong magnetic field, limiting this approach for practical realization. The advantages of electric field technique include the ability to tune energy over a large range, and its ease of integration into devices. However, a large field is necessary to reduce fine structure splitting (FSS) to zero. At large electric field, significant decrease in excitonic emission intensity is observed due to field-induced reduction of the electron-hole wavefunction overlap [10]. Mechanical strain method can in principle be used to tune energy. This method was first proposed by Seidl et al. [8]. Seidl et al. showed the change in emission energy with an increase in piezo voltage. Since then, many studies are conducted which show the ability of controlling QD energy using mechanical strain. The main advantage of this method is no decrease in excitonic emission intensity. However, the energy tuning range of this method is still limited. Large energy tuning was only observed for QDs integrated into another component, such as QDs embedded in nanomembrane and nanowire [11].
Research studies dealing with energy tuning have been extensively conducted, however the current techniques only allow energy tuning of an ensemble of QDs, rather than individual QDs. H. D. Robinson et al. demonstrated stress application by pushing the tip of a fiber onto sample. As a result, blueshift of QD emission line was observed with an increase in compression [12]. Mintairov et al. demonstrated energy tuning of single QD using a probe tip of a near-field scanning optical microscope [13]. However, broadening and decrease in photoluminescence (PL) emission intensity was observed. This technique also requires a feedback control system, which is inconvenient for practical realization. An alternative method that allows precise energy tuning of selected individual QDs is desirable.
As a modified approach to nanoindentation with a probe tip, we have experimentally demonstrated a novel approach to precisely apply a local mechanical strain on QDs by using volume expansion of a phase change material upon amorphization [14]. GeSbTe (GST), which has been extensively investigated for optical recording technology was used as phase change material. Transformation from crystalline to amorphous phase causes a volume expansion of approximately 10%. Since GST film is confined by the SiO2 layer, the volume change induces a local strain to the QD. It should be noted that the GST film also acts as an optical mask for spatially selective observation of the strain induced QD [15]. Compared to other energy tuning techniques, the main advantage of our method is the ability to apply a local strain on QDs which allows selective energy tuning. Other advantages include reversible and fast energy tuning. The device structure used in this research is also simple and practical. It can be easily implemented in device applications.
In the previous work [14], we showed a significant redshift of the PL peak upon amorphization and its recovery to its original position upon recrystallization of phase change material. In addition, we have also observed an interesting phenomenon; not all of investigated QDs showed redshift upon amorphization. Some of the QDs showed blueshift upon amorphization. The ratio of redshift and blueshift obtained in previous experiments is about 11:2. At this stage of investigation, we consider that further numerical and experimental studies are indispensable in order to gain insight into the mechanisms of redshifted and blueshifted PL spectra of QDs. This requires an understanding of the stress and energy shift distribution in sample upon stress applied by volume expansion of phase change material. In this work, we first calculated the stress distribution and energy shift distribution by employing finite element (FE) method. This result unravels the mechanisms of the redshift and blueshift of PL emission energy of QDs. Then, we performed experimental evaluation of FEM simulation result by micro-PL measurements. Here, the behaviour of redshift and blueshift is highlighted by conducting two different experiments. Experimental results are compared with simulation results.
2. Simulation
2.1 Method
We developed a two-dimensional FE model to investigate the stress distribution in sample as a result of stress application from amorphous mark. Since the proposed method is considered as a modified approach to nanoindentation with a probe tip, an indentation test was conducted. We utilized ANSYS 14.0 simulation software to model the indentation. The geometry of the model is depicted in Fig. 1. Indentation was simulated assuming the apex of the amorphous mark applies compressive stress on the top of InP surface.
In this calculation, we assumed that the GST layer is amorphized throughout its whole thickness. Hence, the height of the indenter is 40 nm. The indenter has a flat apex with width of 400 nm. The width of the oval surface at the edge region of the indenter is 100 nm. A quantum dot (InAs QD) with width of 20 nm and height of 1 nm was placed 100 nm below the surface of InP surface along y-axis. These values are taken based on the microscopic observation of the QD sample used in experiment. Young’s Modulus and Poisson’s ratio of InP, InAs and Ge2Sb2Te5 are listed in Table 1. Our previous experimental results exhibit approximately 2 meV redshift of PL peak energy upon amorphization of GST. Therefore, the indentation force was chosen to regenerate a 2 meV redshift at 100 nm beneath the top of InP surface.
Table 1. List of material parameters used in calculation.
Energy shift was then calculated by using Eq. (1). This equation provides the shift of QD emission energy induced by [001] compression [13].
where Y and v are Young’s modulus and Poisson’s ratio, respectively. and represent mean stress components and deviatoric stress components, respectively. The first term in Eq. (1) shows the energy shift induced by hydrostatic stress, while the second term in Eq. (1) corresponds to the energy shift induced by uniaxial stress. The value of a (−6.45) and b (−2.11) is derived from [16].2.2 Results and discussion
Figure 2 presents the stress distribution in model using stress contour plot. The dashed white line shows the location of the QD. The bottom image shows the magnified stress distribution of the black dashed rectangle area shown in (a). In this model, the indentation force is approximately 100 MPa. Based on the FEM simulation result, when redshift 2 meV is obtained at position y = 100 nm, compressive stress induced by amorphous mark on QD is at least 50 MPa.
Fig. 2 (a) Stress distribution in simulation model. The white line shows the position of QD in model. (b) Magnified stress distribution of the black dashed rectangle area shown in (a).
The energy shift distribution in model obtained through calculation is shown in Fig. 3. In Fig. 3, one can see that the energy shift is negative (redshift) at the shallow area beneath the flat part of amorphous mark. On the contrary, the energy shift is positive (blueshift) beneath the edge region of indenter. This simulation result agrees well with our previous experimental results which show both redshift and blueshift as a result of compressive stress applied by volume expansion of phase change material.
Fig. 3 Energy shift distribution in simulation model. Redshift is observed at the shallow area beneath the flat part of indenter, while blueshift is observed at the edge region of the indenter.
In order to identify the stress components acting on the QD, the stress components along y = 100 nm in x-y plane are plotted in Fig. 4. As can be seen in Fig. 4, the hydrostatic stress is almost constant for all contact area. Significant increase in axial stress can be observed at the edge region of amorphous mark. Hydrostatic stress dominates the energy shift of InAs/InP system at the shallow area beneath the flat part of amorphous mark, resulting in a redshift of energy bandgap. Axial stress dominates the energy shift of InAs/InP system at the edge region of amorphous mark, resulting in a blueshift of energy bandgap.
Fig. 4 Stress components along y = −100 nm in 2D model. The red line is the hydrostatic stress along y = −100 nm, while the green line represents the axial stress along y = −100 nm. The inset shows the region where stress values are taken.
3. Experiment
The validity of FEM simulation was evaluated by performing two experiments. In these experiments, QDs were measured in a microphotoluminescence setup. First, a two-dimensional PL intensity mapping over an amorphous mark was conducted. Secondly, a study on the relationship between PL intensity ratio and energy shift was performed. Low density self-assembled InAs/InP QDs grown by solid-source molecular beam epitaxy were utilized as sample [17]. On top of this sample, a layer of GeTe with thickness 40 nm is deposited, followed by deposition of a 100 nm SiO2 layer. The QD density of this sample is 5 × 108 (QDs/cm2).
3.1 Experimental details
We first placed the sample on the cold finger of a Helium flow cryostat, ensuring temperature below 8K. All measurements were conducted at low temperature. To induce nanoscale amorphization, a subnanosecond Q-switched laser was focused through a microscope objective lense (N.A = 0.65) at excitation wavelength 532 nm onto the sample surface (spot size ~1 μm), creating an amorphous mark with diameter less than 1 μm. During PL measurement, QDs were excited using a He-Ne laser at excitation wavelength 632.8 nm, and PL signal was detected by a spectrometer and a liquid-nitrogen cooled InGaAs diode array.
In the first experiment, a two-dimensional PL intensity mapping over a 2 μm × 2 μm amorphous region was performed. Experimental conditions for the PL mapping are as follows. PL spectra were measured at every 200 nm step across a 2 μm × 2 μm area. The integration time per spectrum was 20 s and excitation power density was 80 W/cm2. Two-dimensional PL intensity mapping was performed before and after the creation of amorphized spot. Finally, energy shift of PL peak energy of QDs at the center and the edge of the spot before and after amorphization was measured.
In the second experiment, more than 50 amorphous marks were formed. PL measurement was conducted before and after the formation of each amorphous mark. The energy shift of PL peak energy and corresponding PL intensity ratio (PL intensity at amorphous state/PL intensity at crystalline state) of QDs below amorphous mark were measured.
3.2 Results and discussion
3.2.1 PL mapping intensity of amorphous mark
Figure 5 represents an integrated (0.9668 eV~1.0255 eV) PL intensity mapping diagram of the amorphized spot. High PL intensity region (red-yellow-green) is assumed to be the amorphous mark and the low PL intensity region (blue) is the crystalline region. PL spectra of QDs before and after the formation of the amorphous mark were compared at 3 points; A, B and C. Point A is at the edge of the amorphous mark, point B is at the center of the amorphous mark and point C is at the crystalline region. Figures 6(a)-6(c) exhibit PL spectra of QDs before and after amorphization at point A, B and C respectively. The horizontal axis presents the emission energy, and the vertical axis presents the PL intensity. At point A, a blueshift (3.2 meV) of PL peak energy was observed, while at point B, the PL peak energy was redshifted as large as 1.1 meV. PL peak energy at point C did not exhibit energy shift. These results indicate that the PL peak energy of QD at the center of the amorphous mark was shifted to lower energy (redshift). As opposed to this, PL peak energy of QDs at the edge of the amorphous mark was shifted to higher energy (blueshift). These experimental results show good agreement with FEM simulation result.
Fig. 6 PL spectra measurement of QDs before and after the formation of an amorphous mark. The black dashed circle represents the PL peak of observation. The inset shows the location of observed QD. The arrow shows the direction of the peak shift. After the formation of an amorphous mark, emission energy (a) blueshifts 3.2 meV at point A, (b) redshifts 1.1 meV at point B, and (c) point C does not exhibit energy shift.
3.2.2 Relationship between PL intensity ratio and Energy Shift
Figure 7 shows the relationship between PL intensity ratio and energy shift of QDs below amorphous region. PL intensity ratio is defined as the ratio of PL intensity at amorphous state to PL intensity at crystalline state. As apparent from Fig. 7, when the energy shift is negative, the energy shift increases with an increase in PL intensity ratio. On the contrary, when the energy shift is positive, the result shows no correlation between PL intensity ratio and energy shift. A possible reason for the different behavior for redshift and blueshift is the position of the QDs.
The PL intensity of QDs below the center of amorphous mark depends on the degree of amorphization. For example, if the crystalline phase change material is fully amorphized (in the depth direction), the PL intensity of QDs should be larger compared to partial amorphization. As the degree of amorphization increases, the stress applied becomes larger, which leads to larger energy shift. This explanation agrees well with the redshift case.
On the other hand, QDs below the edge of amorphous region should exhibit lower PL intensity than QDs below the center of amorphous region. At excitation wavelength (He-Ne laser), the absorption coefficient of crystal is larger than that of amorphous. This results in the reduction of the light intensity that reaches the QD sample. When QDs are located below the edge of amorphous region, it is expected that they show no dependence on the degree of amorphization. This corresponds to the blueshift case.
4. Conclusion
In conclusion, we have investigated the mechanisms underlying the redshifted and blueshifted PL emission of QDs. The stress and energy shift distribution in sample were calculated by using FEM. We found that redshift was obtained at the shallow area beneath the flat part of the amorphous mark, while blueshift was exhibited at the edge region of the amorphous. The combined action of hydrostatic and axial stress is responsible for the direction of energy shift. We have also experimentally demonstrated the validity of this simulation result by performing two experiments; (a) a two-dimensional PL intensity mapping over an amorphous mark, and (b) a study on the correlation between PL intensity ratio and energy shift. The experimental results agree with the simulation result. The findings of this study unravel the mechanisms of the redshift and blueshift, and might be the key to precise control of both redshift and blueshift using volume expansion of phase change material.
Acknowledgment
This research was supported by JSPS KAKENHI Grant Number 23360141 and Core-to-Core Program, Advanced Research Networks. We would like to thank Toshimichi Shintani for depositing PCM film on QD sample.
References and links
1. E. Waks, K. Inoue, C. Santori, D. Fattal, J. Vuckovic, G. S. Solomon, and Y. Yamamoto, “Secure communication: Quantum cryptography with a photon turnstile,” Nature 420(6917), 762 (2002). [CrossRef] [PubMed]
2. A. Poppe, A. Fedrizzi, R. Ursin, H. R. Böhm, T. Lörunser, O. Maurhardt, M. Peev, M. Suda, C. Kurtsiefer, H. Weinfurter, T. Jennewein, and A. Zeilinger, “Practical quantum key distribution with polarization entangled photons,” Opt. Express 12(16), 3865–3871 (2004). [CrossRef] [PubMed]
3. D. Loss and D. P. DiVincenzo, “Quantum computation with quantum dots,” Phys. Rev. A 57(1), 120–126 (1998). [CrossRef]
4. R. M. Stevenson, R. J. Young, P. See, D. G. Gevaux, K. Cooper, P. Atkinson, I. Farrer, D. A. Ritchie, and A. J. Shields, “Magnetic-field-induced reduction of the exciton polarization splitting in InAs quantum dots,” Phys. Rev. B 73(3), 033306 (2006). [CrossRef]
5. H. Kim, D. Sridharan, T. C. Shen, G. S. Solomon, and E. Waks, “Strong coupling between two quantum dots and a photonic crystal cavity using magnetic field tuning,” Opt. Express 19(3), 2589–2598 (2011). [CrossRef] [PubMed]
6. K. Kowalik, O. Krebs, A. Lemaitre, S. Laurent, P. Senellart, and P. Voison, “Influence of an in-plane electric field on exciton fine structure in InAs-GaAs self-assembled quantum dots,” Appl. Phys. Lett. 86, 041907 (2005). [CrossRef]
7. M. M. Vogel, S. M. Ulrich, R. Hafenbrak, P. Michler, L. Wang, A. Rastelli, and O. G. Schmidt, “Influence of lateral electric fields on multiexcitonic transitions and fine structure of single quantum dots,” Appl. Phys. Lett. 91(5), 051904 (2007). [CrossRef]
8. S. Seidl, M. Kroner, A. Högele, K. Karrai, R. J. Warburton, A. Badolato, and P. M. Petroff, “Effect of unaxial stress on excitons in a self-assembled quantum dot,” Appl. Phys. Lett. 88(20), 203113 (2006). [CrossRef]
9. C. Bonato, E. Nieuwenburg, J. Gudat, S. Thon, H. Kim, M. P. Exter, and D. Bouwmeester, “Strain tuning of quantum dot optical transitions via laser-induced surface defects,” Phys. Rev. B 84(7), 075306 (2011). [CrossRef]
10. B. D. Gerardot, S. Seidl, P. A. Dalgamo, R. J. Warburton, D. Granados, J. M. Garcia, K. Kowalik, O. Krebs, K. Karrai, A. Badolato, and P. M. Petroff, “Manipulating exciton fine structure in quantum dots with a lateral electric field,” Appl. Phys. Lett. 90(4), 041101 (2007).
11. M. Bouwes Bavinck, M. Zieliński, B. J. Witek, T. Zehender, E. P. A. M. Bakkers, and V. Zwiller, “Controlling a nanowire quantum dot band gap using a straining dielectric envelope,” Nano Lett. 12(12), 6206–6211 (2012). [CrossRef] [PubMed]
12. H. D. Robinson, M. G. Muller, B. B. Goldberg, and J. L. Merz, “Local optical spectroscopy of self-assembled quantum dots using a near-field optical fiber probe to induce a localized strain field,” Appl. Phys. Lett. 72(17), 2081 (1998). [CrossRef]
13. A. M. Mintairov, K. Sun, J. L. Merz, C. Li, A. S. Vlasov, D. A. Vinokurov, O. V. Kovalenkov, V. Tokranov, and S. Oktyabrsky, “Nanoindentation and near-field spectroscopy of single semiconductor quantum dots,” Phys. Rev. B 69(15), 155306 (2004). [CrossRef]
14. M. Takahashi, N. S. Humam, N. Tsumori, T. Saiki, P. Regreny, and M. Gendry, “Local control of emission energy of semiconductor quantum dots using volume expansion of a phase-change material,” Appl. Phys. Lett. 102(9), 093120 (2013). [CrossRef]
15. N. Tsumori, M. Takahashi, R. Kubota, P. Regreny, M. Gendry, and T. Saiki, “Near-infrared nano-spectroscopy of semiconductor quantum dots using a phase-change mask layer,” Appl. Phys. Lett. 100(6), 063111 (2012). [CrossRef]
16. C. Priester, G. Allan, and M. Lannoo, “Band-edge deformation potentials in a tight-binding framework,” Phys. Rev. B Condens. Matter 37(14), 8519–8522 (1988). [CrossRef] [PubMed]
17. R. Kubota, T. Saiki, P. Regreny, A. Benamrouche, and M. Gendry, “Low-density InAs quantum dots grown on InP(001) using solid-source molecular beam epitaxy with a post-growth annealing process,” Jpn. J. Appl. Phys. 49(4), 041201 (2010). [CrossRef]
