The 2006 analysis by Arslanov derived a closed-form optimal taper profile for the metal-coated, subwavelength-aperture probe of a scanning near-field optical microscope, predicting roughly ten-fold higher throughput than a linear taper and localization of the aperture near-field intensity within 15-20 nm at a 100 nm aperture - but the optimization was performed for an isolated probe, with the imaged sample absent from the model, and only a passing remark that deconvolution methods would be needed to convert the compact near field into imaging resolution. The unresolved problem is to extend this optimization to the coupled probe-sample system: to include sample-induced back-reflection and the excited rapidly decaying reflected modes in the modal description, to determine how the optimal profile, throughput, and sub-aperture localization change with sample reflectivity and probe-sample distance, and to establish whether the predicted gains survive at the imaging stage. No published treatment of the coupled optimization, experimental validation of the analytic profile, or reconstruction scheme for its altered point-spread function exists: the problem remains open, though dormant as aperture-probe near-field microscopy has become a niche.

Background

Scanning near-field optical microscopy (SNOM) circumvents the diffraction limit of conventional optical microscopy by exploiting the evanescent (non-propagating) electromagnetic field that exists within a few tens of nanometers of a subwavelength light source. In the aperture-probe implementation, light is guided down a metal-coated tapered optical fiber and exits through an aperture of diameter 50-100 nm at the apex. Since Bethe's classic analysis of diffraction by a small hole in a perfectly conducting screen (transmitted power scales as the fourth power of the ratio of hole diameter to wavelength), the throughput of such probes - the fraction of launched power actually emitted at the aperture - is extremely small (typically 10−510^{-5} to 10−610^{-6}), which limits signal-to-noise and exposes the probe to thermal damage. The emitted aperture field, an evanescent spot with longitudinal and transverse polarization components, acts as the effective resolution-defining light source when the probe is scanned close to a sample. Arslanov (2006) asked which taper profile maximizes probe performance. Using a coupled-mode (cross-section) treatment of the taper as a nonuniform waveguide - the field expanded in local waveguide modes, with approximate impedance boundary conditions on the metal coating, and coupling between the forward main mode, higher-order decaying modes, and back-propagating reflected modes - a variational minimization of mode conversion, wall absorption, and reflection yielded a closed-form optimal profile. Numerical evaluation predicted roughly ten-fold higher output intensity than a linear (conical) taper, localization of the fundamental transverse-magnetic (TM, the polarization whose magnetic field lies in the plane transverse to propagation) mode intensity within a 15-20 nm region at a 100 nm aperture (at wavelength 500 nm), and an optimal final wall angle near 71 degrees. Crucially, this optimization was carried out for an isolated probe: the sample being imaged is not part of the model. That omission is not innocuous. Probe-sample coupling is a well-documented feature of near-field optics: in reflection-mode SNOM, probe-sample interactions alter the measured signal (Cline and Isaacson 1995), and the 'allowed and forbidden light' framework of Novotny (1997) shows that a substantial part of the light emitted by a near-field source interacts with a nearby object and can be scattered back toward it. A sample placed within the evanescent zone of the aperture can therefore reflect energy back into the probe, excite additional rapidly decaying modes, and change both the throughput and the structure of the near field - the very quantities the optimization was designed to control. Whether the isolated-probe optimum survives this coupling, and how it must be modified when the optimization target becomes the coupled probe-sample system, was not analyzed. Exploiting the predicted sub-aperture localization also requires image-reconstruction tools: a SNOM image is essentially the sample convolved with the probe's effective point-spread function (the spatial response of the microscope to a point-like object), and point-spread-function-based deconvolution for SNOM predates the source work (Pieralli 1994), but no scheme has been formulated for the altered, more compact - and, in a coupled treatment, sample-dependent - point-spread function of the optimal probe.

Problem Statement

Determine the optimal form of a subwavelength-aperture near-field optical probe in the presence of the sample it images - that is, extend the isolated-probe optimization of Arslanov (2006), or an equivalent rigorous electromagnetic treatment, to the coupled probe-sample system. Concretely: (a) formulate the modal or full-wave description of the tapered, metal-coated aperture probe including a nearby sample of specified reflectivity at a controlled probe-sample distance, such that sample-induced back-reflection into the probe and the excited rapidly decaying reflected modes enter the field amplitudes and the energy balance; (b) with that model, formulate a physically justified coupled optimization objective - stated explicitly and motivated by imaging-relevant quantities such as the usable near-field intensity at the sample or the achievable image contrast, rather than by isolated-probe throughput alone - and determine how the optimal taper profile and its associated performance figures - throughput and the sub-aperture localization of the near-field intensity - change relative to the isolated-probe optimum, with respect to that objective, as functions of sample reflectivity and probe-sample distance, or establish with quantified error bounds the regime in which the isolated-probe optimum remains valid, where the stated regime must include imaging-relevant coupled settings - a sample of nonzero reflectivity at probe-sample distances within the evanescent zone of the aperture (tens of nanometers) - and not merely the decoupling limit of vanishing reflectivity or asymptotically large distance; and (c) establish whether the compact near-field distribution of the optimal probe yields a genuine resolution improvement in imaging.

References

  1. N. M. Arslanov, 'The optimal form of the scanning near-field optical microscopy probe with subwavelength aperture', Journal of Optics A: Pure and Applied Optics 8, 338-344 (2006), DOI: 10.1088/1464-4258/8/3/018, https://doi.org/10.1088/1464-4258/8/3/018 (Green-OA preprint: arXiv:physics/0509209, https://arxiv.org/abs/physics/0509209)
  2. Narkis M. Arslanov, Sergey A. Moiseev, 'Ultrahigh interference spatial compression of light inside the subwavelength aperture of a near-field optical probe', Journal of the Optical Society of America A 24, 831 (2007), DOI: 10.1364/JOSAA.24.000831, https://doi.org/10.1364/JOSAA.24.000831 (preprint: arXiv:physics/0509187, https://arxiv.org/abs/physics/0509187)
  3. N. M. Arslanov, 'Optimal form of the near-field microscopy probe narrowing for TM1m modes', SPIE Proceedings (2006), DOI: 10.1117/12.675150, https://doi.org/10.1117/12.675150
  4. Narkis M. Arslanov, Sergey A. Moiseev, Ali A. Kamli, 'A transversely localized light in a waveguide: the analytical solution and its potential application', Laser Physics 27, 025103 (2017), DOI: 10.1088/1555-6611/aa4fff, https://doi.org/10.1088/1555-6611/aa4fff (preprint: arXiv:1512.04005, https://arxiv.org/abs/1512.04005)
  5. Stéphane Collin, Fabrice Pardo, Jean-Luc Pelouard, 'Waveguiding in nanoscale metallic apertures', Optics Express 15, 4310 (2007), DOI: 10.1364/OE.15.004310, https://doi.org/10.1364/OE.15.004310
  6. Valeria Lotito, Christian Hafner, Urs Sennhauser, Gian-Luca Bona, 'Novel SNOM Probes Based on Nanofocusing in Asymmetric Structures', book chapter in Plasmonics - Principles and Applications (InTech, 2012), DOI: 10.5772/50806, https://doi.org/10.5772/50806
  7. Lucy Hale, Oleg Mitrofanov, Janine Keller, Tom Siday, Rodolfo I. Hermans, Johannes Haase, John L. Reno, Igal Brener, Giacomo Scalari, Jerome Faist, 'Probe-Sample Interaction in Aperture-type THz Near-Field Microscopy of Complementary Resonators', 44th International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz) (2019), DOI: 10.1109/IRMMW-THz.2019.8874572, https://doi.org/10.1109/IRMMW-THz.2019.8874572
  8. J. A. Cline, M. Isaacson, 'Probe-sample interactions in reflection near-field scanning optical microscopy', Applied Optics 34, 4869 (1995), DOI: 10.1364/AO.34.004869, https://doi.org/10.1364/AO.34.004869
  9. Lukas Novotny, 'Allowed and forbidden light in near-field optics. I. A single dipolar light source', Journal of the Optical Society of America A 14, 91 (1997), DOI: 10.1364/JOSAA.14.000091, https://doi.org/10.1364/JOSAA.14.000091
  10. H. Nakamura, T. Sato, H. Kambe, K. Sawada, T. Saiki, 'Design and optimization of tapered structure of near-field fibre probe based on finite-difference time-domain simulation', Journal of Microscopy 202, 506 (2001), DOI: 10.1046/j.1365-2818.2001.00804.x, https://doi.org/10.1046/j.1365-2818.2001.00804.x
  11. T. Fujii, Y. Taguchi, T. Saiki, Y. Nagasaka, 'Near-field fluorescence thermometry using highly efficient triple-tapered near-field optical fiber probe', Review of Scientific Instruments 83, 114902 (2012), DOI: 10.1063/1.4769057, https://doi.org/10.1063/1.4769057
  12. Liang Wang, Xianfan Xu, 'High transmission nanoscale bowtie-shaped aperture probe for near-field optical imaging', Applied Physics Letters 91, 033110 (2007), DOI: 10.1063/1.2752542, https://doi.org/10.1063/1.2752542
  13. C. Pieralli, 'Statistical estimation of point spread function applied to scanning near-field optical microscopy', Optics Communications 108, 187 (1994), DOI: 10.1016/0030-4018(94)90649-1, https://doi.org/10.1016/0030-4018(94)90649-1
  14. H. A. Bethe, 'Theory of Diffraction by Small Holes', Physical Review 66, 163 (1944), DOI: 10.1103/PhysRev.66.163, https://doi.org/10.1103/PhysRev.66.163