Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Strong bound exciton-photon coupling in ZnO whispering gallery microcavity

Open Access Open Access

Abstract

We report the cathodoluminescence (CL) study of bound-exciton-polaritons in ZnO whispering gallery (WG) microcavity. Thanks to the very high spatial resolution (: 100 nm) of CL technique, a scanning CL mapping along the tapered ZnO nanowire is achieved. We observe a clear anticrossing behavior which demonstrates the strong coupling between cavity mode and bound-excitons. Coupled oscillator model including both bound excitons and free excitons fits well with the experimental results. The energy splitting of bound-exciton-polaritons:2.6 meV is obtained.

© 2013 Optical Society of America

1. Introduction

Exciton-polaritons, generated by the strong coupling between light and excitons, are attracting tremendous attentions in the last decades due to the unique properties exhibited in both the physics of quantum coherence and the novel optical devices. Great advances in this field have been achieved, e.g., polariton lasing, Bose-Einstein condensation of exciton-polaritons, super〉uidity, vortices, et al [14]. These interesting macro quantum phenomena arise from dense polaritons involving free excitons which show extended propagating waveform. Another interesting system is defect-induced localized excitons (bound-excitons) coupling with light, which would be potential for studying the optical evolvement of polaritons from collective to single emitter excitation. It is expected that the strong coupling between bound-excitons and light is realized by a series of virtual coherent emitting and reabsorption process of bound-excitons [5,6]. However, the experimental showing the anticrossing for strong coupling of light-bound exciton is yet to be demonstrated, and the cavity confined bound-exciton polaritons have not been reported before. We note that introducing microcavity to the subject of bound-exciton polaritons would pave a new way for revealing the physical mechanism of bound-exciton polaritons and extends its potential application in photonic storage and communications.

ZnO as a wide band gap material has large exciton binding energy (: 60 meV) and giant exciton oscillator strength, which is ideal semiconductor for the study of light-matter interaction. The slowdown of light due to exciton-polariton propagation in ZnO has been reported [7]. The strong coupling between cavity modes and free excitons in ZnO have been successfully observed in nanowire, of which hexagonal cross section functioned as a WG microcavity [8,9]. In this letter, A tapered ZnO nanowire with radius gradually decreasing from the root to the tip top, is used as a testbed. Resonant modes (WGMs) are clearly observed in the CL spectra. The scanning CL mapping along the tapered nanowire is performed at low temperature (5 K). We find that a clear anticrossing trend appears at energy range of the cavity modes resonating with the bound exciton states. This unambiguously indicates bound-exciton strongly coupled with cavity modes. The plane wave formula for WGMs combined with the multi-oscillators coupled model for polaritons give a good fit to the experimental results. The experimental obtained energy splitting of bound exciton-cavity mode strong coupling is about 2.6 meV.

2. Experimental details

The high quality ZnO tetrapod with tapered nanowire arms are fabricated by using vapor phase transport method [8]. To prepare a single nanowire for CL measurement, we first disperse nanowires in ethanol by sonicating and transfer them onto a Si substrate. The CL system includes a SEM (FEI Quanta 200, resolution 3-4 nm) and a grating monochromator equipped with a liquid nitrogen cooled CCD camera (Jobin Yvon, HR460, 0.3 meV/pixel). CL is collected by the parabolic mirror which has been mounted in the chamber of SEM and then analyzed by the monochromator [10]. For the CL mapping along a single nanowire, the accelerating voltage of 20 kV and the excitation spot of 100 nm are chosen. The CL scanning step is set to ~1 µm, and the diameters of nanowire at different scanning points are measured simultaneously. To observe the emission peaks of bound excitons clearly and suppress the thermal broadening of linewidth of the peaks, the sample, which is placed on a copper holder of a He-〉ow-type cryostat in the chamber of SEM, is cooled to 5 K.

3. Results and discussion

Figure 1(a) shows the SEM image of ZnO tetrapods with tapered arms (nanowire) dispersed on Si substrate. One can see that the arms of ZnO tetrapods usually have a length of tens of micrometers. The tapered arm marked by a white rectangle in the Fig. 1(a), is used for CL measurement. Its enlarged SEM image is shown in Fig. 1(b), from which we can see that the radius of tapered nanowire gradually decreases from:400 nm at the root to:270 nm at the tip top. The CL spectra, measured on both the middle part of nanowire (P1) and the root part (P2) at 5 K, are depicted in Fig. 1(c). The typical spectrum is characterized by excitonic transitions in the ultraviolet energy region. The free excitons (XA, XB) are visible at 3.370 eV and 3.375 eV. The donor-bound excitons (DXs) are responsible for the several sharp peaks at energy region from 3.355 eV to 3.364 eV. The more detailed peak features of these bound excitons are shown in Fig. 1(d). The main peaks I8 (3.360 eV), I9 (3.355 eV) can be identified as the emission from the shallow donors Indium and GaZn [11]. The I8a has not been identified yet. The peaks around 3.364 eV labeled as D0XB are assigned to the same donors as those for I8a and I8, but with the hole derived from the B valence band [12]. In addition, two-electron-satellite transitions (TES, 3.324 eV and 3.308 eV) of the DXs and the first order of phonon replica of DX (DX-1LO, 3.289 eV) can be also observed [13]. Compared to the spectrum (P2), a shoulder-like peak appeared at 3.357 eV in the spectrum (P1) is clearly observed. This distinctive peak is actually a resonant mode, which comes from a natural formed WG microcavity in the hexagonal cross section plane of ZnO nanowire due to the total internal re〉ection effect [14].

 figure: Fig. 1

Fig. 1 (a) The SEM image of dispersed ZnO tetrapods on Si substrate. (b) The enlarged SEM of a single ZnO tapered nanowire (one arm of tetrapod) extracted from the white rectangle region from (a). (c) The CL spectra measured on the position P1 (middle part of nanowire) and P2 (the root part of nanowire). (d) Cavity mode (WGM) with linewidth:1.6 meV is clearly observed in the spectrum (P1).

Download Full Size | PDF

Thanks to the natural formed nanowire with a tapered radius, we can perform a scanning CL mapping along the c-axis of nanowire to further examine the origin of the shoulder-like peak, because the WG modes would be strongly affected by the radius of the nanowire. And the most important is that by this strategy we can control precisely the energy of the cavity mode to be in resonance with the donor-bound exciton states, and to explore the strong interaction behavior of bound exciton and cavity modes. In our experiment, by scanning the excitation spot with a step of 1 µm, the CL mapping is obtained as shown in Fig. 2. It is obvious that at least four cavity modes are visible and they gradually shift to high energy side with decreasing the radius of nanowire (white belts labeled by yellow stars). The energy shift is not linear versus the radius, but shows an anticrossing-like behavior against the excitons states. This result is similar to the PL mapping of ZnO tapered nanowires at room temperature [7]. The strong coupling between cavity modes and excitons at UV region should be considered to explain this off-linear behavior of WG modes. In addition, we can also clearly resolve several linear white belts being independent on radius (labeled by red lines), which are ascribed to the intrinsic exciton-related transitions as we have analyzed in Fig. 1(c). However, at the bound-excitons energy region, as shown in Fig. 3, one can see a dislocation of cavity modes at bound excitons energies, showing an anticrossing dispersion. To make this dispersion more visible, we extract cavity mode peaks carefully with multi-peaks fitting and second-order differential method, as shown in Fig. 3(b). The anticrossing behavior is undoubtedly observed by eliminating the strong affection of bound excitons emission. This provides a straightforward evidence for the strong coupling between cavity modes and bound-exciton states.

 figure: Fig. 2

Fig. 2 The scanning CL mapping along the c-axis of ZnO nanowire, the modulated WG modes (N = 9 −12, yellow stars) can be clearly observed. Excitons-related transitions are also labeled by red lines.

Download Full Size | PDF

 figure: Fig. 3

Fig. 3 (a) The scanning CL mapping along the c-axis of ZnO nanowire at bound-excitons energy region. A log-scale for intensity is used. (b) The expanded view of anticrossing region is shown. The cavity modes peaks are carefully extracted by multi-peaks fitting and second-order differential treatment. The dashed red curves (colored curves in (b)) are calculated by multi-oscillators coupling model with both free and bound excitons included.

Download Full Size | PDF

To further understand this phenomenon, a theoretical simulation is performed. We can use a plane wave model to describe the WG modes in ZnO hexagonal cross section. The resonant energy Ec as a function of the circumscribing radius of the hexagonal cavity R can be written as [8,14]

Ec=hc33nbgR[N+6πarctan(β3nbg24)]

Here, nbg is the background refractive index, h and c are the Plank constant and the speed of light in vacuum. The integer N is the mode number. The factor β denotes polarization. For TE polarization (Ec-axis), β = n, and for TM polarization (E//c-axis), β = 1/n. In UV region, the polarization of emission mainly shows TE preference because A- and B-excitons are TE polarized, and their recombination and phonon replicas would dominate the emission [15]. Therefore, the observed cavity modes here can be assigned to the TE polarized WG modes.

At room temperature, these TE polarized cavity modes will strongly interact with A- and B-excitons, and thus form hybrid quasi-particles, excitonic polaritons. A simple coupled oscillators model including A-, B-excitons, and cavity modes can well describe this blueshifted behavior of TE polarized resonant modes (the lower branch of TE polarized exciton polaritons) in ZnO. However, at low temperature, most of free excitons are captured by localized donors, and become bound excitons, as we have observed at CL spectra. These localized states will play an important role in the strong coupling of cavity modes and excitons, and thus finally determine the eigenstates of system. For TE polarized dispersions, both bound excitons (I9, I8a, I8, D0XB) and free excitons (XA, XB) should be included in the fitting. Here, the quasi-degenerated I8a and I8 are simplified to an effective bound exciton I8eff. Therefore, the eigenstates of system can be obtained from the diagonalization of the Hamiltonian matrix, which is given by

H^=[EcgAgBg1g2g3gAEA0000gB0EB000g100E100g2000E20g30000E3]
where Ec is the energy of cavity mode, which is determined by Eq. (1). EA (3.370 eV), EB (3.376 eV) are free A- and B-excitons energies and gA, gB are the corresponding coupling strengths. E1 (I9, 3.355 eV), E2 (I8eff, 3.360 eV), and E3 (D0XB, 3.364 eV) are bound excitions energies with the corresponding coupling strengths g1, g2 and g3. The exact energies of excitons can be obtained from the spectra of Fig. 1 (d). Combined Eq. (1) with Matrix (2), we can give a best fit to the experimental result (Fig. 3) with parameters gA = 0.021 eV, gB = 0.103 eV, g1 = 0.002 eV, g2 = 0.009 eV and g3 = 0.002 eV. From this best fitting, we can see that the coupling strengths (gA, gB) of free excitons mainly contribute to the strong coupling of light-exciton. However, under this free exciton-light strong coupling effect, the bound exciton-light strong interaction is still visible, we obtain experimentally that the energy splitting Ωbd for I8eff bound-exciton polariton, which shows an obvious anti-crossing behavior, is about 2.6 meV.

Without the free excitons and other bound excitons, the Rabi splitting (Ω) of I8eff can be calculated from fitting parameter g2 with Ω = 2g2, which is 0.018 eV. To further verify the reasonability of the fitting parameter. We compared our value with the oscillator strengths (f) of I8a and I8, which is obtained by using time-of-flight spectroscopy [7]. The Rabi splitting can be written asΩ=2ωexωLT, where ωexand ωLT are exciton energy and bulk value of longitudinal-transverse splitting, respectively [16]. According to the expression of dispersion of exciton polariton in [7], Chen et al., and [8], Sun et al., oscillator strength f can be described asf=2ωLT/ωex, thus we can get the relationship between Ω and f asf=Ω/ωex. With this equation, the f (I8ef) is calculated as 2.86 × 10−5. Considering the f (I8ef) here contains both oscillator strengths of I8a and I8, this value is almost consistent with f (I8) = 2.6 × 10−5 and f (I8a) = 1.3 × 10−5 in [7], Chen et al.. A little bit deviation may be due to much higher bound exciton-light coupling induced by improved light confinement in ZnO WG microcavity or higher density of donor impurities in ZnO nanowire.

Here we note that the CL from bound excitons is much stronger than that of bound-exciton polariton modes. This can be easily understood by considering the unique properties of WG microcavity formed by total internal re〉ection. In this kind of cavity, only light with specific wave vectors can couple into the WG modes and forming bound-exciton polaritons, as depicted in [8], Sun et al., and in [15], Sun et al.. Most of photons emitted by bound excitons escape directly from nanowire and contribute a very strong CL of bound excitons.

4. Conclusions

In conclusion, we have shown the direct evidence for bound exciton-cavity mode strong coupling in ZnO WG microcavity. By performing CL scanning along a tapered ZnO nanowire, the anticrossing behavior between cavity modes and bound excitons are clearly observed. The plane wave formula for WG modes combined with multi-oscillators coupling model for light-exciton strong interaction gives a good fit to the experimental results. An energy splitting of about 2.6 meV is obtained from experiments.

Acknowledgments

We would like to thank Jinkyoung Yoo and Le Si Dang for CL measurements in Institut Néel, Grenoble. This work was supported by the National Natural Science Foundation of China (Grant Nos: 11104302, 91121007, 11225419), Shanghai Committee of Science and Technology, China (Grant No: 11ZR1442300), the 973 projects of China (No. 2011CB925600).

References and links

1. S. Christopoulos, G. B. von Högersthal, A. J. D. Grundy, P. G. Lagoudakis, A. V. Kavokin, J. J. Baumberg, G. Christmann, R. Butté, E. Feltin, J. F. Carlin, and N. Grandjean, “Room-temperature polariton lasing in semiconductor microcavities,” Phys. Rev. Lett. 98(12), 126405 (2007). [CrossRef]   [PubMed]  

2. J. Kasprzak, M. Richard, S. Kundermann, A. Baas, P. Jeambrun, J. M. J. Keeling, F. M. Marchetti, M. H. Szymańska, R. André, J. L. Staehli, V. Savona, P. B. Littlewood, B. Deveaud, and S. Dang, “Bose-Einstein condensation of exciton polaritons,” Nature 443(7110), 409–414 (2006). [CrossRef]   [PubMed]  

3. A. Amo, D. Sanvitto, F. P. Laussy, D. Ballarini, E. del Valle, M. D. Martin, A. Lemaître, J. Bloch, D. N. Krizhanovskii, M. S. Skolnick, C. Tejedor, and L. Viña, “Collective fluid dynamics of a polariton condensate in a semiconductor microcavity,” Nature 457(7227), 291–295 (2009). [CrossRef]   [PubMed]  

4. K. G. Lagoudakis, M. Wouters, M. Richard, A. Baas, I. Carusotto, R. André, L. S. Dang, and B. Deveaud-Plédran, “Quantized vortices in an exciton-polariton condensate,” Nat. Phys. 4(9), 706–710 (2008). [CrossRef]  

5. A. V. Kavokin, G. Malpuech, and W. Langbein, “Theory of propagation and scattering of exciton-polaritons in quantum wells,” Solid State Commun. 120(7–8), 259–263 (2001). [CrossRef]  

6. G. Xiong, J. Wilkinson, K. B. Ucer, and R. T. Williams, “Time-of-flight study of bound exciton polariton dispersive propagation in ZnO,” J. Phys. Condens. Matter 17(46), 7287–7296 (2005). [CrossRef]  

7. S. L. Chen, W. M. Chen, and I. A. Buyanova, “Slowdown of light due to exciton-polariton propagation in ZnO,” Phys. Rev. B 83(24), 245212 (2011). [CrossRef]  

8. L. Sun, Z. Chen, Q. Ren, K. Yu, L. Bai, W. Zhou, H. Xiong, Z. Q. Zhu, and X. Shen, “Direct observation of whispering gallery mode polaritons and their dispersion in a ZnO tapered microcavity,” Phys. Rev. Lett. 100(15), 156403 (2008). [CrossRef]   [PubMed]  

9. L. Sun, H. Dong, W. Xie, Z. An, X. Shen, and Z. Chen, “Quasi-whispering gallery modes of exciton-polaritons in a ZnO microrod,” Opt. Express 18(15), 15371–15376 (2010). [CrossRef]   [PubMed]  

10. B. Piechal, J. Yoo, A. Elshaer, A. C. Mofor, G.-C. Yi, A. Bakin, A. Waag, F. Donatini, and L. S. Dang, “Cathodoluminescence of single ZnO nanorod heterostructures,” Phys. Status Solidi B 244(5), 1458–1461 (2007). [CrossRef]  

11. B. K. Meyer, H. Alves, D. M. Hofmann, W. Kriegseis, D. Forster, F. Bertram, J. Christen, A. Hoffmann, M. Straßburg, M. Dworzak, U. Haboeck, and A. V. Rodina, “Bound exciton and donor-acceptor pair recombinations in ZnO,” Phys. Status Solidi B 241(2), 231–260 (2004). [CrossRef]  

12. M. Schirra, A. Reiser, G. M. Prinz, A. Ladenburger, K. Thonke, and R. Sauer, “Cathodoluminescence study of single zinc oxide nanopillars with high spatial and spectral resolution,” J. Appl. Phys. 101(11), 113509 (2007). [CrossRef]  

13. S.-K. Lee, S. L. Chen, D. Hongxing, L. Sun, Z. Chen, W. M. Chen, and I. A. Buyanova, “Long lifetime of free excitons in ZnO tetrapod structures,” Appl. Phys. Lett. 96(8), 083104 (2010). [CrossRef]  

14. T. Nobis, E. M. Kaidashev, A. Rahm, M. Lorenz, and M. Grundmann, “Whispering gallery modes in nanosized dielectric resonators with hexagonal cross section,” Phys. Rev. Lett. 93(10), 103903 (2004). [CrossRef]   [PubMed]  

15. L. Sun, Z. Chen, Q. Ren, K. Yu, W. Zhou, L. Bai, Z. Q. Zhu, and X. Shen, “Polarized photoluminescence study of whispering gallery mode polaritons in ZnO microcavity,” Phys. Status. Solidi C 6(1), 133–136 (2009). [CrossRef]  

16. M. A. Kaliteevski, S. Brand, R. A. Abram, A. Kavokin, and L. Dang, “Whispering gallery polaritons in cylindrical cavities,” Phys. Rev. B 75(23), 233309 (2007). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (3)

Fig. 1
Fig. 1 (a) The SEM image of dispersed ZnO tetrapods on Si substrate. (b) The enlarged SEM of a single ZnO tapered nanowire (one arm of tetrapod) extracted from the white rectangle region from (a). (c) The CL spectra measured on the position P1 (middle part of nanowire) and P2 (the root part of nanowire). (d) Cavity mode (WGM) with linewidth:1.6 meV is clearly observed in the spectrum (P1).
Fig. 2
Fig. 2 The scanning CL mapping along the c-axis of ZnO nanowire, the modulated WG modes (N = 9 −12, yellow stars) can be clearly observed. Excitons-related transitions are also labeled by red lines.
Fig. 3
Fig. 3 (a) The scanning CL mapping along the c-axis of ZnO nanowire at bound-excitons energy region. A log-scale for intensity is used. (b) The expanded view of anticrossing region is shown. The cavity modes peaks are carefully extracted by multi-peaks fitting and second-order differential treatment. The dashed red curves (colored curves in (b)) are calculated by multi-oscillators coupling model with both free and bound excitons included.

Equations (2)

Equations on this page are rendered with MathJax. Learn more.

E c = hc 3 3 n bg R [N+ 6 π arctan(β 3 n bg 2 4 )]
H ^ =[ E c g A g B g 1 g 2 g 3 g A E A 0 0 0 0 g B 0 E B 0 0 0 g 1 0 0 E 1 0 0 g 2 0 0 0 E 2 0 g 3 0 0 0 0 E 3 ]
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.