We present a novel concept to design apertureless plasmonic probes for near-field scanning optical microscopy (NSOM) with enhanced optical power throughput and near-field enhancement. Specifically, we combine unidirectional surface plasmon polariton (SPP) generation along the tip lateral walls with nanofocusing of SPPs through adiabatic propagation towards an apertureless tip. Three key design parameters are considered: the nanoslit width, the pitch period of nanogrooves for unidirectional plasmonic excitation and the pyramidal geometry of the NSOM probe for SPP focusing. Optimal design parameters are obtained with 2D analysis and two realistic probe geometries with patterned plasmonic surfaces are proposed using the optimized designs. The electromagnetic properties of the designed probes are characterized in the near-field and compared to those of a conventional single-aperture probe with same pyramidal shape. The optimized probes feature FWHM around 150nm, comparable with conventional NSOM designs, but over 3 orders of magnitude larger field enhancement, without degrading its spatial resolution. Our ideas effectively combine the resolution of apertureless probes with throughput levels much larger than those available even in aperture-based devices.
©2011 Optical Society of America
The resolution of conventional optical microscopy is governed by Rayleigh diffraction and limited by the wavelength of operation [1,2]. Various pursuits and efforts have been made to overcome this fundamental optical limitation [3–5], especially operating in the near-field of the object to be imaged. In particular, near-field scanning optical microscopy (NSOM) shows great promise to overcome the diffraction limit, by capturing the evanescent portion of the spatial spectrum of an image, before it rapidly decays away from the object. A conventional single aperture NSOM probe features spatial scanning resolution around 100 nm, which is largely determined by the aperture size, rather than by the wavelength of operation. Classic NSOM measurements can therefore break the diffraction limit; however, aperture NSOM probes still suffer from low optical throughput, which ultimately limits the resolution due to a low signal-to-noise ratio .
In this respect, the main challenge for efficient near-field scanning is on one hand to provide highly localized electromagnetic energy localization near the tip of the NSOM probe, in order to illuminate only a sub-wavelength detail of the object of interest, and on the other to support high optical throughput, in order to be able to detect the scattering from the object over the background noise. With conventional aperture-based NSOM tips, the resolution of the measurement, inversely proportional to the aperture size, is also inversely related to the overall power throughput; and low optical throughput remains a challenge for high-resolution near-field scanning application. On the other hand, techniques exploiting SPP waves have indeed shown promising results [7–9] in enhancing the achieved field enhancement at the tip, since they allow focusing and localizing the electromagnetic wave within extremely small regions compared to the wavelength. An apertureless sharp metallic probe tip efficiently generates SPP from exterior illumination with a help of periodic grooves and guide SPPs with a near-field spatial resolution of around 20 nm ; tip-aperture plasmonic probes surrounded by periodic nanogrooves have been shown to provide high optical throughput in the near-field region [11,12]. The possibility of nanofocusing SPPs propagating on a plasmonic surface have been shown using arrays of subwavelength holes  or slits . In addition, interesting plasmonic planar geometries that may focus and direct SPPs towards specific locations with sub-wavelength resolutions have been recently demonstrated both numerically and experimentally [15–17]. Among these techniques, the ones that utilize SPP diffraction for directional excitation appears particularly attractive for NSOM measurements, as it can provide higher electromagnetic energy focusing for the local probe geometry of interest. Also, apertureless plasmonic geometry allowing for highly localized near-field confinement seems more desirable for high resolution NSOM measurements.
In this paper, we put forward new concepts to employ SPP generation and focusing to improve apertureless NSOM measurements: our goal is to extract the energy from the probe before it gets to the bottom of the tip, where the guided energy is deeply below cut-off due to the small cross-section of the probe. By introducing an optimally designed slit at a much larger cross-section, and directing the generated SPPs towards the apertureless tip, we aim at obtaining throughputs much larger than what available in conventional aperture-based probes, and then routing the energy towards an apertureless tip to ensure large resolution. We use two schemes to achieve this ambitious goal: nanofocusing of SPP propagation using proper shaping of the probe tip and enhanced plasmonic generation through nanopatterning of the probe tip, in order to achieve enhanced optical throughput in the near-field region. Specifically, sharp pyramidal probes, which can be fabricated by anisotropic wet-chemical etching process, are combined with gratings supporting unidirectional SPP propagation, in order to excite enhanced electromagnetic modes confined around the probe tip, which may drastically increase light focusing compared to conventional NSOM probes. Two different types of plasmonic probes are designed; each one integrates a slit grating with different unidirectional SPP excitation techniques. To verify our designs, we first explore the two-dimensional (2D) excitation of plasmonic surfaces by a metal slit in a planar geometry; then we apply the optimized design parameters to 3D realistic probe designs to explore the effects of geometrical, unidirectional SPP nanofocusing for NSOM measurements. Both proposed designs achieve electric field enhancement over 1000 times compared to a conventional single aperture probe, without compromising spatial resolution. In addition, the designed probes offer potential operations over a wide spectral range.
2. Unidirectional surface plasmon polariton excitation
Unidirectional excitation of SPPs along a plasmonic surface may be obtained by properly designing a grating, or the nanoslit width given a specific angle of illumination: the nanograting design may be tailored to reflect SPPs excited in the unwanted direction with respect to the slit [15–17], using stop-band design concepts; the nanoslit can also efficiently couple incident waves into SPPs preferably in one direction, exploiting the asymmetry in its geometry. Before applying these techniques to specific 3D probe designs, in order to enhance their throughput efficiency, we numerically analyze 2D plasmonic surfaces using finite-integration (FIT) numerical simulations.
2.1 Unidirectional SPP excitation for oblique illumination of a single slit
We first explore the unidirectional excitation of SPPs by illuminating a tilted slit on a metal surface, similar to the lateral wall of an NSOM probe tip realized by focused ion-beam (FIB) milling process. Our idea is to open a slit on the side of the metallic wall and, since we envision a probe design with pyramidal shape, the metal slit, realized by FIB milling process, will be tilted with an angle of illumination parallel to the one of the tilted slit, as sketched in Fig. 1 . Unidirectional SPP excitation from a single non-tilted slit with oblique backside illumination has been already demonstrated in numerical and analytical studies  and its optical transmission has been theoretically analyzed in . Intuitively, the SPP coupling efficiency is a function of the aperture shape; analytical design parameters for the slit geometry may be determined by exploring the wave behavior inside and outside the slit. The asymmetry of the slit geometry in Fig. 1 ensures that in principle it is possible to excite only one side of the surface. For simplicity of analysis, we consider a 2D geometry (in the transverse y direction) with a tilting angle θ. In this case, a monochromatic TM wave may be represented with a plane-wave expansion as a two dimensional Fourier integral with respect to x and y [19,20]. The scalar potential represents the transverse magnetic field supported by the geometry. The angular spectrum of the slit geometry may be represented as:21] in positive x direction, defined as ; where and are relative dielectric constants of metal and dielectric medium, respectively. describes the slit geometry. By tailoring the output face of the slit as a function of the electromagnetic wave transmission, the slit may be optimized to launch a unidirectional plasmonic wave [16–18,22].
2.2 Unidirectional SPP excitation using a groove array
Unidirectional SPPs may also be generated by using a proper groove geometry that provides a bandgap or destructive interference on one side of the slit . This way, we expect to be able to reflect the SPP wave excited in the unwanted direction and focus the SPP propagation in the direction of the probe tip. Due to the presence of a grating array with stop-band properties, SPPs generated on the left side of the slit in Fig. 2 will be reflected back. The condition that enables resonant interference satisfies the following expression
2.3 Two-dimensional numerical simulations
To numerically investigate the directivity of the SPP excitation at the exit of the slit in the two designs described in the previous subsections, we can define the directivity factor of SPP excitation by comparing the intensities of fields at the left and right sides of the slit:
is the magnetic field in component in the y direction calculated at the right side of the slit and is the magnetic field at the left side of the slit.
Two-dimensional (2D) simulations have been carried out to characterize the directivity D and the amplitude of magnetic field generated on the right of the slit exit . The operational frequency of HeNe laser (633 nm) is considered for excitation. Relative dielectric constant of silver at 633 nm is  and surrounding medium is air (). We have analyzed two different slit geometries, consistent with the previous descriptions: a single slit with optimized slit width (design A, Fig. 3(a) , corresponding to the geometry of Fig. 1) and a single slit geometry surrounded by an array of three grooves (design B, Fig. 3(b), corresponding to the geometry of Fig. 2). The thickness of the metal film, 200 nm, is chosen to roughly match the Fabry-Pérot resonance condition, to maximize the transmission through the slit . For slit A, the amplitude and directivity are calculated as a function of the slit width; for slit B, they are evaluated as a function of d for a fixed slit width wslit = 100 nm.
It is seen how we can optimize the design parameters to maximize the SPP excitation in terms of maximum directivity and high field transmission. The optimal slit width for design A is 371 nm, and this value agrees well with the value calculated from the cavity modal expansion presented in . It is interesting to notice that, as expected, the metal slit geometry couples the incident wave into SPPs in a bidirectional manner when the slit width is smaller than 170 nm, about a quarter wavelength of the incident wavelength, as expected from a small aperture in a metallic screen . For design B, we have chosen a smaller slit width (100 nm) and we direct the SPPs by properly designing the groove array on one side of the slit. The optimized value d = 484 m, and pgroove = 308.5 nm are chosen, which agrees well with the considerations in the previous subsection. For comparison, we show the H-field distributions for a simple 100 nm slit aperture and the optimized designs A and B in Fig. 4 (a), (b) and (c) , respectively.
It is seen how the small aperture slit in Fig. 4(a) excites SPPs on both sides of the aperture, but by either increasing the aperture or tailoring the grooves around it, it is possible generate unidirectional SPP propagation. Comparing the configurations presented in Fig. 4, design A shows about 2 times higher transmission than type B, due to the larger slit aperture. Roughly it is expected that 3D probe design employing design A will feature 2 times higher electromagnetic field intensity (4 times higher electrical energy density) compared to design B. We apply these concepts to an optimized 3D probe design in the next section.
3. Nanofocusing probe design using unidirectional SPP generation
Here we apply the previous concepts to optimized 3D probe designs based on unidirectional SPP excitation. We consider a metal-coated () pyramidal geometry, used to provide SPP nanofocusing towards the tip for NSOM probe measurements. A pyramidal probe with a tip angle of 70.6◦ is considered, which may be obtained by anisotropic wet-etching process. The probe is metal-coated with a silver film 200 nm thick and has a sharp tip apex, with diameter of 100 nm. The optimized design parameters obtained in Section 2 are applied to the design of proper slit apertures on the lateral walls of the 3D probes, as depicted in Fig. 5 , in order to excite SPP propagating and focusing towards the probe tip. Since the geometry is 3D here, due to symmetry considerations a continuous slit as the one depicted in Fig. 5 will excite SPP modes whose axial component of electric field excited by a linearly polarized source will destructively interfere at the apex of the probe. For this reason, we consider here a radially polarized light source at 633 nm in order to nanofocus constructively the SPP propagation on all sides of the probe towards the tip. This would ensure maximized near-field enhancement of the designed probes. A radially polarized light source may be configured by applying proper boundary conditions in our 3D simulations. Schematic drawings of the designed probes and of a canonical single aperture probe excited by a radially polarized light source are shown in Fig. 5.
4. Near-field light enhancement at the probe tip through coupled SPPs
Three different types of NSOM probes, i.e., probe A, probe B and a classic single aperture probe are numerically characterized in the near-field of the tip, consistent with Fig. 5. The calculated electric field intensity and the electric energy density of each probe are shown in Fig. 6 . For the electromagnetic characterization, full width and half maximum (FWHM) is defined by the distribution of the electric energy density near the probe tip. The electric energy density is compared to a canonical single aperture probe at 20 nm distance from the tip. Compared to the conventional single aperture probe, type A and type B probes feature electric field enhancements by a factor of 2119 and 1023, respectively, consistent with the 2D results reported in Section 2. As predicted above, type A shows the best performance in terms of optical throughput in the near-field region. Electrical energy density of type A is boosted more than times, without compromising FWHM. This value is about 4 times higher than that of type B probe, and this value is in well agreement with the predicted values from 2D simulations. The calculated FWHMs of the three probes are: 148.84 nm (canonical aperture probe), 138.02 nm (type A) and 171.3 nm (type B). In Fig. 7 , the calculated FWHM of each probe corresponding to various tip-sample distances is shown. Within 100 nm from the probe tip, the calculated FWHM of probe type A and B is under 200 nm, and the two designed probes still hold huge energy density enhancement (over six orders of magnitude compared to the simple aperture probe).
From these numerical results, it is found that the FWHM of the designed probe is largely determined by the tip size, as already reported in , rather than by the type of excitation. The SPPs mainly contribute to dramatically enhance the field throughput at the aperture, with substantial advantages in terms of signal-to-noise ratio, without compromising the measurement resolution. It is expected that with smaller tip sizes, the SPPs may be focused even more efficiently. In our different designs, we have achieved FWHM around 50 nm for both type A and type B probes using sharper probe tips of 30 nm. Even though higher field enhancement and confinement may be achieved through the utilization of sharper tip geometries, in this study, we have focused on a 100nm probe tip, which is reliably attainable and reproducible with current nanofabrication technology . This will pave the way to the near-future experimental demonstration of these concepts, which we are planning in our group.
5. Discussion and conclusions
We have presented here optimized plasmonic probe designs for enhanced near-field optical throughput. Without using a conventional aperture geometry  or external illumination , we have numerically proven that simple nanopatterns, such as a slit and slit with an array of reflective grooves [15,16], may be used for near-field enhancement purposes, in combination with SPP nanofocusing associated with the NSOM probe tip. Under radially polarized illumination, the designed plasmonic probes show extremely high field enhancement at the apex of probe tip. The electric field enhancement is over three orders of magnitude higher than a classic single aperture probe for both proposed designs. The FWHM of all designed probes is comparable with the conventional single aperture probe within 100 nm from the probe tip, while providing huge electrical field enhancement. It is interesting to notice that the designed probes, especially type A, whose functionality does not depend on destructive interference due to gratings, provide the possibility of wideband operation if the slit width is properly selected. Even if the frequency of illumination is deviated from the optimal design frequency, the slit will still couple most part of the impinging wave into unidirectional SPPs and the generated SPPs will be focused at the apex of the probe by the sharp metallic probe tip. Figure 8(a) and (b) shows the spectral characteristics of the three probe designs in the near-field region.
It is obvious that design B probes are more sensitive to frequency variation, due to frequency selective properties of the groove array, consistent with the periodic features of the directivity in Fig. 3(b). Both designs, however, can provide large throughput enhancement (of 4 and 6 orders of magnitude, respectively) over a wide frequency range without degrading FWHM. It is worth mentioning that, although type A probes offer consistently superior performance, in their practical realization type B designs can be advantageous. For instance, in configurations and experimental realizations in which it is difficult to control the slit angle or when perfectly collimated incident light is not obtainable, type B probe designs may offer more robust operation.
In conclusion, we have reported an extensive numerical study to optimize NSOM probe designs for nanoscale slits and grooves that may couple the impinging wave into unidirectional SPPs. We have shown that these concepts may provide large near-field enhancement compared to opening an aperture at the probe tip. This makes physical sense, since the probe geometry is effectively a waveguide below cut-off, whose throughput is largely affected by the small aperture size. By opening a slit away from the tip, where the waveguide cross section is larger, most of the impinging energy may be coupled to SPP modes, which do not have a cut-off and can propagate and focus at the probe tip. The optimized slit shape and groove design provide unidirectional SPP excitation, which further enhances the coupling efficiency towards the tip. The proposed designs can achieve extremely high near-field enhancement while minimizing the unwanted scattering of usual apertureless plasmonic probe that require external illumination. Unlike our proposed designs, a conventional apertureless scanning probe with direct tip illumination often shows limited performance when strong background noise is considered, associated with the probe body scattering and direct sample reflected scattered light. To overcome these drawbacks, recent progress has been made based on nonlocal probe tip geometry  and highly polarized SPPs launched from gratings, enabling a significant increase of signal to noise ratio; however, additional signal discrimination method based on the orientation of polarization of scattered field is still required in this configuration. In contrast, our approach allows minimal background noise and may not require any signal discrimination technique, which usually comes at the expense of NSOM sensitivity. In addition, our designs are particularly appealing for nanolithographic processes and may be realized within current nanofabrication technology. Pyramidal hollow tips suited for scanning probe can be formed by wet chemical etching of (100) and (111) single crystal silicon in TMAH etchant . Tip formation with anisotropic etching technology and thermal oxidation yields a sharp tip diameter of less than 100 nm. Further apex sharpening with thermal oxidation and film deposition technology may be possible and result in tips with apex radius below 30 nm. Armed with in situ nanofabrication technology, we may realize a scanning probe featuring high optical resolution with high optical power throughput that is hardly achieved from conventional aperture-based scanning probes.
This research was performed at Biomedical Engineering, Microelectronics Research Center (MRC) at UT Austin. The authors thank Dr. Kazunori Hoshino and Dr. Ashwini Gopal for stimulating discussions on nanophotonic structure design and nanofabrication. We gratefully acknowledge the financial support from National Science Foundation (NSF CAREER Award Grant No. 0846313, PI: Zhang, NSF CAREER Award Grant No. 0953311, PI: Alù) and DARPA Young Faculty Award (N66001-10-1-4049, PI: Zhang).
References and links
1. E. Hecht and A. R. Ganesan, Optics (Pearson Education, 2001).
2. E. Abbe, “Beiträge zur theorie des mikroskops und der mikroskopischen wahrnehmung,” Archiv. Mikros. Anat. 9, 413 (1873). [CrossRef]
3. E. Betzig, J. K. Trautman, T. D. Harris, J. S. Weiner, and R. L. Kostelak, “Breaking the diffraction barrier: optical microscopy on a nanometric scale,” Science 251(5000), 1468–1470 (1991). [CrossRef] [PubMed]
5. A. Lewis and K. Lieberman, “Near-field optical imaging with a non-evanescently excited high-brightness light source of sub-wavelength dimensions,” Nature 354(6350), 214–216 (1991). [CrossRef]
6. R. Bachelot, F. H'dhili, D. Barchiesi, G. Lerondel, R. Fikri, P. Royer, N. Landraud, J. Peretti, F. Chaput, G. Lampel, J.-P. Boilot, and K. Lahlil, “Apertureless near-field optical microscopy: A study of the local tip field enhancement using photosensitive azobenzene-containing films,” J. Appl. Phys. 94(3), 2060–2072 (2003). [CrossRef]
8. V. S. Volkov, S. I. Bozhevolnyi, S. G. Rodrigo, L. Martín-Moreno, F. J. García-Vidal, E. Devaux, and T. W. Ebbesen, “Nanofocusing with channel plasmon polaritons,” Nano Lett. 9(3), 1278–1282 (2009). [CrossRef] [PubMed]
9. K. C. Vernon, D. K. Gramotnev, and D. F. P. Pile, “Adiabatic nanofocusing of plasmons by a sharp metal wedge on a dielectric substrate,” J. Appl. Phys. 101(10), 104312 (2007). [CrossRef]
10. C. C. Neacsu, S. Berweger, R. L. Olmon, L. V. Saraf, C. Ropers, and M. B. Raschke, “Near-field localization in plasmonic superfocusing: a nanoemitter on a tip,” Nano Lett. 10(2), 592–596 (2010). [CrossRef] [PubMed]
12. Y. Wang, Y. Y. Huang, and X. J. Zhang, “Plasmonic nanograting tip design for high power throughput near-field scanning aperture probe,” Opt. Express 18(13), 14004–14011 (2010). [CrossRef] [PubMed]
13. L. Yin, V. K. Vlasko-Vlasov, J. Pearson, J. M. Hiller, J. Hua, U. Welp, D. E. Brown, and C. W. Kimball, “Subwavelength focusing and guiding of surface plasmons,” Nano Lett. 5(7), 1399–1402 (2005). [CrossRef] [PubMed]
14. W. Chen, D. C. Abeysinghe, R. L. Nelson, and Q. Zhan, “Plasmonic lens made of multiple concentric metallic rings under radially polarized illumination,” Nano Lett. 9(12), 4320–4325 (2009). [CrossRef] [PubMed]
15. F. López-Tejeira, S. G. Rodrigo, L. Martín-Moreno, F. J. García-Vidal, E. Devaux, T. W. Ebbesen, J. R. Krenn, I. P. Radko, S. I. Bozhevolnyi, M. U. González, J. C. Weeber, and A. Dereux, “Efficient unidirectional nanoslit couplers for surface plasmons,” Nat. Phys. 3(5), 324–328 (2007). [CrossRef]
16. H. Kim and B. Lee, “Unidirectional surface plasmon polariton excitation on single slit with oblique backside illumination,” Plasmonics 4(2), 153–159 (2009). [CrossRef]
17. S. B. Choi, D. J. Park, Y. K. Jeong, Y. C. Yun, M. S. Jeong, C. C. Byeon, J. H. Kang, Q.-H. Park, and D. S. Kim, “Directional control of surface plasmon polariton waves propagating through an asymmetric Bragg resonator,” Appl. Phys. Lett. 94(6), 063115 (2009). [CrossRef]
19. J. Soohoo and G. E. Mevers, “Cavity mode analysis using the fourier transform method,” Proc. IEEE 62(12), 1721–1723 (1974). [CrossRef]
20. R. Petit, Electromagnetic Theory of Gratings (Springer-Verlag, New York, 1980).
21. H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings (Springer-Verlag, Berlin, 1988).
22. S. Astilean, Ph. Lalanneb, and M. Palamarua, “Light transmission through metallic channels much smaller than the wavelength,” Opt. Commun. 175(4-6), 265–273 (2000). [CrossRef]
23. E. D. Palik and G. Ghosh, Handbook of Optical Constants of Solids, E.D. Palik, ed. (Academic, Orlando, Fla., 1985).
24. H. A. Bethe, “Theory of diffraction by small holes,” Phys. Rev. 66(7-8), 163–182 (1944). [CrossRef]
25. J. S. Hyun, J. S. Moon, J. H. Parka, J. W. Kima, Y. D. Kimaand, and J. H. Boo, “Fabrication of near-field optical probes using advanced functional thin films for MEMS and NEMS applications,” Mater. Sci. Eng. 149, 292–298 (2008).