APPLIED PHYSICS LETTERS 96, 151105 共2010兲
X-shaped plasmonic antenna on a quantum cascade laser D. Austin,1,a兲 N. Mullin,1 I. Luxmoore,2 I. C. Sandall,1 A. G. Cullis,2 A. Bismuto,3 J. Faist,3 J. K. Hobbs,1 and L. R. Wilson1 1
Department of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, United Kingdom EPSRC National Centre for III-V Technologies, Department of Electronic and Electrical Engineering, University of Sheffield, Sheffield S1 3JD, United Kingdom 3 Institute for Quantum Electronics, ETH Zurich, Zurich 8093, Switzerland 2
共Received 9 December 2009; accepted 15 March 2010; published online 13 April 2010兲 We report an x-shaped plasmonic antenna design patterned onto the gold coated facet of a mid-infrared quantum cascade laser. Using apertureless scanning near-field optical microscopy we measure a single enhanced region in the optical near-field at the center of the x-antenna, with a full-width-at-half-maximum of ⬃100 nm for the operating wavelength of ⬃8.8 m. This design provides complete suppression of near-field signal away from the center, with concomitant improvements in imaging contrast expected. Our experimental results are also in good agreement with finite difference time domain simulations, which show a full-width-at-half-maximum of ⬃80 nm. © 2010 American Institute of Physics. 关doi:10.1063/1.3380660兴 Quantum cascade lasers 共QCLs兲 have now reached a high level of technological maturity, providing wide tunability and complete control over wavelength and emission linewidth at mid-infrared 共mid-IR兲 wavelengths.1–3 QCL emission wavelengths cover the “fingerprint” region corresponding to molecular rotational-vibrational transitions allowing the highly sensitive identification of molecular bonds.4,5 The integration of plasmonic antennas into QCLs provides a route for their implementation in vibrational microscopy applications. The subwavelength nature of an optically enhanced region affords high spatial resolution distinction between regions of differing chemical composition for almost any material when coupled with a near-field imaging system such as the apertureless scanning near-field optical microscope 共a-SNOM兲. Plasmonic antennas typically consist of subwavelength metal rod or bowtie shaped structures. They have therefore only recently become viable for mid-IR and optical wavelengths, owing to the requirement for sophisticated fabrication techniques. In the presence of incident electromagnetic radiation, polarized parallel to the axis of the rods or bowties, the radiation is coupled down to a small volume, many times below the corresponding free space wavelength.6,7 This results in an ultraenhanced near field in the antenna feed gap, due to the accumulation of charge at the antenna ends. In integrating them onto QCLs, the plasmonic antenna must have its axis aligned to the transversemagnetic polarized emission of the laser. Plasmonic antennas have recently been investigated for a number of different applications, including enhanced SNOM imaging,8 nanolithography,9 in addition to the laser facet defined antennas10–12 reported by a Harvard group. Previously in their work, rods or bowtie structures were used to obtain the strongly enhanced near field. However, in both of those designs there was a substantial enhancement observed at the outer ends of the antenna geometry, although this was reduced in the bowtie design. In this letter, we present an x-shaped antenna design to only allow a single enhanced spot and provide complete suppression of side lobes. This approach provides maximum contrast and sensitivity rea兲
Electronic mail:
[email protected].
0003-6951/2010/96共15兲/151105/3/$30.00
quired for future imaging of chemical and biological materials. The QCL devices themselves 共sample number N974A兲 were double high-reflection 共HR兲 coated 共360 nm Al2O3 followed by 180 nm Au兲. The threshold was measured at ⬃1.15 A at RT 共operating characteristics shown in Fig. 1兲 with an active region of standard bound-to-continuum design.13 A Focused ion beam system 共JEOL Fabrika based upon a JEOL 6500F SEM with an Orsay Physics Ga+ ion column attached兲 was then used to pattern the x-antenna onto the emission-side facet. Finite difference time domain 共FDTD兲 simulations 共using the commercial software package XFDTD 6.5 by Remcom兲 were used to design the x-antenna, as shown in Fig. 1. The x-antenna is completely analogous to bowtie antennas/ apertures in theory and design, so an initial estimate for the size of the antenna is obtained from the equation L = / 2n for rod-type antennas, where is the wavelength and n is the refractive index of the underlying substrate. L is generally taken as the vertical length of each bowtie section. However,
FIG. 1. 共Color online兲 FDTD simulations of the x-antenna. 共a兲 Near-field of the x-antenna, on a linear scale, with the E-field enhancement factor indicated on the right, with the maximum shown at the top 共⬃31.7⫻兲. The simulations are normalized to an incoming plane wave of 1 V amplitude, thus automatically giving a measure of the enhancement factor. The horizontal and vertical directions are defined here for the purposes of simplicity. 共b兲 共Black line兲 A linescan of the near-field across the horizontal of the antenna shown in 共a兲, with the length 共in m兲 along the x-axis, and E-field enhancement along the y-axis. It is seen, and can be assumed from the geometry, that only a single enhanced spot is expected. 共Other line兲 A linescan along the vertical of the x-antenna. From this it is seen that a different profile would be expected along the two polarizations. It is seen that a high confinement is achieved in both directions, with the strong enhancement in the antenna gap, which is 80 nm across in the calculations.
96, 151105-1
© 2010 American Institute of Physics
151105-2
Austin et al.
this length is not well defined for bowtie antennas, which show a relatively broadband response,14 with multiple resonances observed for large tip angles. It is redshifted with increasing tip angle, and in this case the surrounding metal film adds an effective capacitive load to the length. Also, in this case, n would be taken as between the refractive index values of the dielectric layer and semiconductor. The resonant antenna length is known to differ close to optical frequencies due to plasma oscillations in the metal, effectively becoming much shorter. Often, this is represented by a scaling factor15 but this can be ignored at the frequencies of operation used in this work. The gap size also has an effect on the near-field enhancement possible for a particular antenna length. A larger field intensity and stronger confinement is observed for sharper antenna tips and a reduced antenna tip angle, due to an increased density of charge. For the same reason, it is found that a reduced gap size increases the intensity. The x-antenna in this work represents a compromise between design and fabrication. Our simulations predict that ⬃30% improvements in the electric field enhancement will be expected in future work, by reduction in the antenna gap size to less than 50 nm and antenna length to ⬃1.1 m at which another of the resonances is observed. In the simulations, a sinusoidal plane wave polarized along one length of the antenna is propagated perpendicular to the structure. Five to six periods of the wave were used to provide adequate convergence, with spatial discretization of cell sizes ranging from 10 nm near the gap to 20 nm nearer the edges. This was found to provide adequate resolution to simulate the antenna length. Perfectly matched layers were used on all boundaries. For the x-antenna, these simulations showed a maximum electric field enhancement of ⬃30⫻ across the feed gap, as shown in Fig. 1. The confinement 关full-width-at-half-maximum 共FWHM兲兴 was found to be ⬃80 nm in the horizontal and ⬃180 nm in the vertical directions 共see Fig. 1 for the convention of vertical and horizontal used here兲. The a-SNOM 共Refs. 16–18兲 is an ideal instrument for imaging in the mid-IR region, providing an ultrahigh resolution which is wavelength independent. In the conventional SNOM technique, light passes through a subwavelength aperture which is scanned across the surface of a sample, limiting the achievable resolution to about a tenth of the wavelength of the light while maintaining a usable transmission through the aperture.19 In contrast, the apertureless method involves detection of the scattered radiation from an oscillating probe, with the scattering from the tip apex collected for imaging. In our configuration, the ⬃ 8.8 m QCL is used as the source and the a-SNOM gold coated tip is scanned across the nanopatterned facet, allowing for simultaneous topographic force microscopy analysis and direct near field imaging of the plasmon-enhanced region during QCL operation. The a-SNOM used in this work was based around a Veeco Dimension 3100 atomic force microscope 共AFM兲 with a NanoScope IV controller. The AFM was operated in tapping mode using NanoWorld Arrow FMR cantilevers. These were coated with a 50 nm gold layer by thermal evaporation, and their resonant frequency was approximately 55 kHz 共after coating兲. The tip amplitude was chosen to be approximately 30 nm. The QCL was operated on a Peltier-cooled copper mount, allowing the devices to be driven well above thresh-
Appl. Phys. Lett. 96, 151105 共2010兲
FIG. 2. 共Color online兲 A representation of the a-SNOM setup, at the edge of the QCL device. 共a兲 The tip oscillates at ⬃55 kHz. 共b兲 A mirror is placed to the rear of the device, to collect the scattered signal. 共c兲 Two aspheric ZnSe lenses are used to focus the light onto the Mercury-cadmium-telluride detector element 共d兲.
old 共⬃30%兲 at room temperature, with a pulse duration of ⬃80 ns and repetition rate of 500 kHz, while ensuring stability of device output. This frequency was chosen to provide discrimination between the desired near-field optical signal 共to give a sinusoid demodulated at ⬃110 kHz for the second harmonic of the tip oscillation frequency兲 and the unwanted scattered laser output, using a lock-in amplifier referenced to the cantilever drive signal. A mirror was placed behind the device as in Fig. 2. The reflected signal was passed through two aspheric ZnSe lenses, and onto a liquid nitrogen cooled mercury cadmium telluride detector. Demodulation at the second harmonic frequency is often used to reduce the effect on the signal from scattering from the tip shaft and cantilever.20 Figure 3 shows the AFM topographic and a-SNOM optical images for the x-antenna. The line scans across the antenna show the intense near-field spot localized to a region 共FWHM兲 ⬃110 nm 共 / 80兲 in the horizontal and ⬃140 nm 共 / 60兲 in the vertical directions. This is the overall smallest confinement observed for plasmonic antennas in the mid-IR. The gap in the fabricated antenna was ⬃80 nm. The enhanced region appears larger than this and is probably due to the interaction with the gold a-SNOM tip, which is a somewhat invasive tool to measure the near-field optical signal as the tip effectively acts as a second antenna for the near-field. For an example of how the tip may affect this near field when in close proximity to the antenna, further FDTD calculations were performed in this configuration, with some necessary limitations. The tip was considered to be in a stationary position in each calculation, in a parallel plane 10 nm above the surface of the x-antenna. This simplified model indicates a further enhancement factor of the near field of ⬃3⫻ when the tip is directly over the antenna feed gap 共as shown in Fig. 4兲. The total field at the a-SNOM tip was plotted against its relative position to the antenna center. Furthermore, it is seen in Fig. 4共d兲 that the near field is “smeared out” across the surface of the antenna, as the tip travels across the surface, but only along the horizontal direction Fig. 4共d兲 also shows the near field along the vertical axis, which was largely unaffected by the tip. This was somewhat borne out by the linescans of Fig. 3 which showed secondary peaks in areas parallel to the antenna axis, but no clear secondary peaks along the perpendicular, and a larger horizontal FWHM than expected from the simulations of Fig. 1. There were also obvious defects in the fabricated antenna due to an inhomogeneous HR coating over the QCL facet 共as seen in
151105-3
Appl. Phys. Lett. 96, 151105 共2010兲
Austin et al.
FIG. 4. 共Color online兲 FDTD simulations of the tip-antenna interaction. 共a兲 The AFM tip is here centered at the gap, and 10 nm above the surface. It is seen that the gold AFM tip pulls the field from the center of the gap. 共b兲 Linescans across the antenna surface. The electric field is seen to be further enhanced by the presence of the AFM tip, but is also responsible for a broadening of the FWHM, especially along the plane of polarization. The black line is the perpendicular polarization. 共c兲 The AFM tip is here 0.5 m away from the center of the antenna, along the plane of polarization. 共d兲 The surface near-field with the AFM tip perpendicular to the antenna length, 0.5 m away from the center.
show both a 共horizontally兲 broadened and increased near field signal due to the “lightning rod” effect of the a-SNOM tip. This work was funded by EPSRC 共Grant No. EP/ G004307/1兲, and the QCL devices were grown and fabricated at ETH Zurich. 1
FIG. 3. 共Color online兲 Experimental results of the x-antenna. 共a兲 Operating characteristics of the QCL used. The L-I shows a 1.15 A threshold at room temperature, with the inset giving the associated spectrum, centered at ⬃8.8 m. 共b兲 SEM image of two x-antennas on the QCL facet. The top edge of the facet is on the left of the image. Multiple antennas were patterned, to account for the unknown mode profile of the laser. 共c兲 AFM topographic image of the x-antenna. 共d兲 Simultaneously obtained a-SNOM optical image of the antenna in 共c兲. A clear bright spot is seen at the center of the structure, as expected from simulation. 共e兲 Three-dimensional image of the optical signal of 共d兲. 共f兲 Horizontal linescan of the antenna, with the confinement factor of about / 80 FWHM. 共g兲 Vertical linescan of the antenna, with the confinement factor of about / 60. There is some correspondence with the simulations in this regard, together with the differing profiles across the gap and antenna lengths.
the SEM image of Fig. 3兲, perhaps leading to differences between the simulated and measured signal. To conclude, an x-shaped plasmonic antenna has been fabricated on the facet of a mid-IR QCL. Experimental verification of the enhanced near field expected from FDTD simulation was observed through an a-SNOM imaging system, with complete suppression of side-lobes. Strong 共⬃100 nm兲 confinement was observed in both horizontal and vertical directions, many times below the wavelength of the QCL. FDTD simulations of the tip-antenna interaction
Y. Bai, S. R. Darvish, S. Slivken, W. Zhang, A. Evans, J. Nguyen, and M. Razeghi, Appl. Phys. Lett. 92, 101105 共2008兲. 2 D. G. Revin, J. W. Cockburn, M. J. Steer, R. J. Airey, M. Hopkinson, A. B. Krysa, L. R. Wilson, and S. Menzel, Appl. Phys. Lett. 90, 151105 共2007兲. 3 X. J. Wang, J. Y. Fan, T. Tanbun-Ek, and F. S. Choa, Appl. Phys. Lett. 90, 211103 共2007兲. 4 M. R. McCurdy, Y. A. Bakhirkin, and F. K. Tittel, Appl. Phys. B: Lasers Opt. 85, 445 共2006兲. 5 R. Lewicki, G. Wysocki, A. A. Kosterev, and F. K. Tittel, Opt. Express 15, 7357 共2007兲. 6 K. B. Crozier, A. Sundaramurthy, G. S. Kino, and C. F. Quate, J. Appl. Phys. 94, 4632 共2003兲. 7 R. D. Grober, R. J. Schoelkopf, and D. E. Prober, Appl. Phys. Lett. 70, 1354 共1997兲. 8 J. N. Farahani, H.-J. Eisler, D. W. Pohl, M. Pavius, P. Fluckiger, P. Gasser, and B. Hecht, Nanotechnology 18, 125506 共2007兲. 9 A. Sundaramurthy, P. J. Schuck, N. R. Conley, D. P. Fromm, G. S. Kino, and W. E. Moerner, Nano Lett. 6, 355 共2006兲. 10 E. Cubukcu, E. A. Kort, K. B. Crozier, and F. Capasso, Appl. Phys. Lett. 89, 093120 共2006兲. 11 N. Yu, E. Cubukcu, L. Diehl, M. A. Belkin, K. B. Crozier, F. Capasso, D. Bour, S. Corzine, and G. Hofler, Appl. Phys. Lett. 91, 173113 共2007兲. 12 N. Yu, E. Cubukcu, L. Diehl, D. Bour, S. Corzine, J. Zhu, G. Hofler, B. B. Crozier, and F. Capasso, Opt. Express 15, 13272 共2007兲. 13 J. Faist, D. Hofstetter, M. Beck, T. Aellen, M. Rochat, and S. Blaser, IEEE J. Quantum Electron. 38, 533 共2002兲. 14 H. Fischer and O. J. F. Martin, Opt. Express 16, 9144 共2008兲. 15 L. Novotny, Phys. Rev. Lett. 98, 266802 共2007兲. 16 B. Knoll and F. Keilmann, Nature 共London兲 399, 134 共1999兲. 17 A. Bek, R. Vogelgesang, and K. Kern, Rev. Sci. Instrum. 77, 043703 共2006兲. 18 B. Knoll and F. Keilmann, Appl. Phys. A: Mater. Sci. Process. 66, 477 共1998兲. 19 B. Hecht, B. Sick, and U. P. Wild, V. Deckert, R. Zenobi, O. J. F. Martin, and D. W. Pohl, J. Chem. Phys. 112, 7761 共2000兲. 20 B. Knoll and F. Keilmann, Opt. Commun. 182, 321 共2000兲.
