In a recent paper, Intonti et al. attributed ͓Phys. Rev. Lett. 87, 076801 ͑2001͔͒ a 3-meV peak in the autocorrelation spectrum of low-temperature nanophotoluminescence spectra of a single disordered GaAs quantum well to level repulsion, i.e., to the statistical analog of an avoided crossing due to overlapping wave functions in the disorder potential. Our data, which reproduce their findings very nearly, are taken to an additional test employing filter functions, which clearly shows that the 3-meV peak is associated with lowenergy states-in striking contrast to the level repulsion scenario. By a careful analysis of the high-energy states, however, we are able to identify a second peak around 1.5 meV which we attribute to level repulsion. The experiments are compared with simple model calculations, which support our interpretation.
We use scanning near-field optical microscopy and spectroscopy to investigate the near field of onedimensional periodic arrays of gold nanowires on top of a slab waveguide. The Bragg resonance of these one-dimensional metallic photonic crystal slabs can coincide with the particle-plasmon resonance of the metal nanowires, giving rise to an avoided crossing. We find a rich behavior when systematically varying the polarization of the incident light, the lattice constant, and the slab thickness. Numerical simulations of the transverse magnetic component of the near field qualitatively reproduce the experiment, in agreement with the expectation from the Bethe-Bouwkamp theory. Furthermore, these calculations give insight into those electromagnetic field components which cannot easily be measured.
Combining a low-temperature scanning near-field optical microscope with a picosecond streak camera allows us to measure the complete wavelength-time behavior at one spot on the sample within about 13 min at excitation powers of 100 nW. We use this instrument to measure the variation of relaxation times in disordered single-GaAs quantum wells with sample position. © 2002 American Institute of Physics. ͓DOI: 10.1063/1.1477274͔Soon after the invention of scanning probe microscopy, many researchers got excited about combining scanning probe techniques with the established methods of femto-or picosecond optical spectroscopy-a combination, which would open the door to a whole new family of experiments. Several pioneering experiments were performed indeed, [1][2][3][4] however, much of the enthusiasm was largely damped over the years because of enormous technical difficulties which inhibited a routine application of these techniques. Here, we combine a low-temperature scanning near-field optical microscope ͑using uncoated optical fiber tips, allowing for large throughput͒ with a picosecond streak camera. While this combination has only modest spatial ͓200-300 nm resolution at 800 nm photoluminescence ͑PL͒ wavelength, see Fig. 1͔ and temporal ͑2-10 ps͒ resolution, it does allow for routine operation. Compared to, e.g., using solid immersion lenses, our setup allows for more flexibility: Large field of view, coarse sample positioning up to 5 mm and ability to also work at around 400 nm wavelength. 5,6 Figure 1 depicts the instrument. The sample is cooled by a variable-temperature He-flow cryostat with symmetric heat exchanger, while the fiber tip is held at room temperature. 5-7The scanning near-field optical microscope ͑SNOM͒ uses uncoated fiber tips, which are fabricated using a two-step selective etching process, 8 leading to an apex radius around 20 nm. The 150 fs optical pulses at exc ϭ700 nm wavelength ͑⇔ប exc ϭ1.77 eV photon energy͒ with an average power of P exc ϭ100 nW are sent into the single-mode optical fiber. We use the constant height mode 8 with a typical tipsample separation of 100 nm, i.e., the feedback loop is switched off. The excited PL of the sample is collected by the fiber tip and sent into a 0.5 m grating spectrometer connected to the streak camera. Without temporal resolution ͑i.e., without the streak camera͒, we typically detect several hundreds of counts per second in one wavelength channel ͑corresponding to 0.012 nm width͒ in the PL peaks under these conditions. As this number of counts is spreadout over time, i.e., over several hundreds of channels, one obviously needs photon-counting capability and significant integration ͑exposure͒ times when introducing the streak camera. The time resolution is 2 ps without spectrometer and 10 ps for the actual experiments ͑e.g., Fig. 2͒.The two samples investigated in this letter are highquality single-GaAs quantum wells ͑SQW͒. One is a 3.5 nm thin, 180 s growth-interrupted GaAs SQW with superlattice barriers ͓4 monolayers ͑ML͒ of AlAs, no interruption, 8...
Introducing spatial autocorrelation analysis of nanophotoluminescence images of atomically rough single GaAs quantum wells, we show that the bright spots in the images are not randomly distributed but rather tend to “repel each other” along certain directions. Simple computer simulations on the anisotropic transport of excitons in the disorder potential and on the resulting images can reproduce this surprising behavior.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.