Nanoplasmonic Terahertz Photoconductive Switch on GaAs

Nov 21, 2012 - sensitivity in detection from a GaAs-based PC switch by using a nanoplasmonic structure. As compared to a conventional. GaAs PC switch,...
0 downloads 0 Views 578KB Size
Subscriber access provided by UNIV OF ALABAMA BIRMINGHAM

Communication

Nanoplasmonic Terahertz Photoconductive Switch on GaAs Barmak Heshmat, Hamid Pahlevaninezhad, Yuanjie Pang, Mostafa Masnadi Shirazi, Ryan B. Lewis, Thomas Tiedje, Reuven Gordon, and Thomas E. Darcie Nano Lett., Just Accepted Manuscript • DOI: 10.1021/nl303314a • Publication Date (Web): 21 Nov 2012 Downloaded from http://pubs.acs.org on November 25, 2012

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Nanoplasmonic Terahertz Photoconductive Switch on GaAs 240x114mm (120 x 120 DPI)

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 12

Nanoplasmonic Terahertz Photoconductive Switch on GaAs Barmak Heshmat†, Hamid Pahlevaninezhad†,Yuanje Pang†, Mostafa Masnadi Shirazi‡, Ryan Lewis‡, Thomas Tiedje†, Reuven Gordon*† and Thomas Edward Darcie† †



Department of Electrical and Computer Engineering, University of Victoria, V8P 5C2, Victoria, BC Canada

Department of Electrical and Computer Engineering, University of British Columbia, V6T 1Z4, Vancouver, BC, Canada *

[email protected]

Abstract: Low-temperature (LT) grown GaAs has a sub-picosecond carrier response time that makes it favorable for terahertz photoconductive (PC) switching. However, this is obtained at the price of lower mobility and lower thermal conductivity than GaAs. Here we demonstrate sub-picosecond carrier sweep-out and over an order of magnitude higher sensitivity in detection from a GaAs-based PC switch by using a nanoplasmonic structure. As compared to a conventional GaAs PC switch, we observe 40 times the peak-to-peak response from the nanoplasmonic structure on GaAs. The response is double that of a commercial, antireflection coated LT-GaAs PC switch.

Keywords: Nanoplasmonics; Terahertz Detection; Photoconductive switches. Research in terahertz (THz) imaging1-3 and spectroscopy4-8 has advanced significantly in the last decade, fueled particularly by advances in the output power of THz sources and the efficiency of THz detectors9-18. Photoconductive (PC) switching offers advantages over other methods of THz generation and detection11. In particular, PC switches are compact and low cost, they operate at room temperature, and they can function both as emitters and receivers of THz waves. These PC switches are center-fed microantennas, usually fabricated on a short carrier-lifetime semiconducting material, such as low-temperature (LT) grown GaAs12. The feed current of the microantenna is provided by the short-lifetime photocarriers that are generated through optical excitation of the center gap of the antenna. Research efforts have focused on improving the efficiency of PC switches for THz applications13-15. From the material perspective, low-mobility LT-GaAs replaced high-mobility GaAs due to shorter carrier lifetime more than a decade ago11.

ACS Paragon Plus Environment

Page 3 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

This allowed higher bandwidth operation and deeper carrier density modulation that was vital to access the hyper-THz (>1 THz) range and higher sensitivity. However, this was achieved at the price of lower mobility. Since the efficiency is directly dependent on mobility12,15, numerous studies have been conducted to compensate for the low mobility in LT-GaAs using more complex materials such as GaBiAs and ion-implanted GaAs16,17. Recent work has focused on nanomaterials, such as semiconducting ErAs/GaAs nano-island superlattices18, quantum wells19, carbon nanotubes15, and metamaterials20. Separately, from the antenna perspective there has been an effort to use interlaced structures12, 21 and nanostructures in the gap of the antenna to improve the efficiency. The size of the center gap of the antenna was conventionally defined by the optical excitation spot size and thermal conduction of the substrate. Conventional designs tend to converge to gap size of 46 µm. Recently, a submicron-sized gap was proposed and successfully implemented in the form of tip-to-tip nanogaps and nanoantennas13, 22, 23. Tip-to-tip nanogaps22 provide high electric bias-field intensity in continuous-wave THz emission, but due to their small scale they have low optical coupling efficiency. Interlaced structures on LT-GaAs, with gaps of a few microns, provide more efficient carrier collection at the price of lower optical coupling and higher undesired dark current in THz emitting PC switches12, 21. Nanoplasmonics (i.e., the nanostructuring of metals to achieve an enhanced optical response due to interaction of metal surface plasmons with the incident light) has been used to improve the performance of detectors24,25, light emitters26, 27, imaging systems28-31, nanoantennas32, 33 and waveguides34-39. A key capability of nanoplasmonics is efficient coupling of light into subwavelength structures40-47. The optical electric-field affects the surface plasmons of the metal nanostructures; if the metallic nanostructure has proper geometry, the induced effects can constructively resonate with the surface plasmons in a desired fashion. Such resonances are tuned to notably affect the transmission or reflection of subwavelength metallic apertures48-55. Here we demonstrate a nanoplasmonic PC switch that efficiently couples light into 100 nm gaps. These small gaps enable fast carrier sweep out from a GaAs substrate for efficient THz detection. We compare the performance of this nanoplasmonic PC switch with a conventional PC switch structure on GaAs, LT-GaAs, and with a commercial LT-GaAs, antireflection coated device, with peak-to-peak response enhancements of 40×, 10×, and 2×, respectively. A schematic of a contemporary THz heterodyne setup is shown in Fig.1 (a). In this setup, the phase of the THz pulse is varied with variations of the length of the optical excitation pulse path using a delay line. The THz pulse is focused by a silicon lens through the back side of the chip. The receiver is connected to a lock-in amplifier that detects the current signal. Conventional designs for THz receiving PC switches are 10-200 µm dipoles with 5-10 µm center gaps fabricated on LTGaAs substrates. In our experiment, this center gap is replaced with a nanoplasmonic interlaced structure with 100 nm gaps (Fig. 1 (b), (c)).

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 12

We used semi-insulating GaAs wafer ([100] orientation, 350 µm thickness, ~1.3×108 ohm.cm resistivity and ~5500 cm2/V.s mobility). The carrier lifetime exceeds 200 ps for bulk GaAs substrates, as confirmed by reflective pump-probe measurements. For the first step of fabrication, a closed-gap structure is fabricated by photolithography (95 nm Au with a 5 nm Cr adhesive layer). The closed gap is then sputtered using a focused ion beam machine (Hitachi FB-2100 Focused Ion Beam system). The interlaced structure is connected to gold electrodes which are connected to larger gold pads. The gold pads facilitate biasing and measurement of the signal from the structure (Fig. 1 (b)).

Fig. 1. (a) Schematic of the THz detection setup. Abbreviations used: MR = Mirror, BS = Beam splitter, DL = Delay line, L= Lens, Rx = THz receiving PC switch, Tx = THz transmitting PC switch, and fs laser = femtosecond Ti-Sapphire pulse laser. (b) 3D model of the plasmonic THz receiving PC switch. (c) SEM image of the interlaced structure. The gaps are 100 nm and the metallic fingers are 600 nm wide each. The inset shows further details of the structure.

To compare the performance of the interlaced structure, we also fabricated a conventional 20 µm dipole with 5 µm center gap on the same GaAs substrate. Additionally, we fabricated the same dipole structure on an LT-GaAs substrate (1µm LTGaAs grown at 250ºC, in-situ MBE annealing at 600ºC for 60 second, 0.5% excess As). A commercial LT-GaAs PC switch (BATOP PCA-800nm) was used as the THz transmitter. The signal is first measured with another similar commercial PC switch at the receiver side (10µm dipole with 5µm center gap with back-mounted silicon lens and substrate antireflection

ACS Paragon Plus Environment

Page 5 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

coating). There is a difference in the size of the commercial dipole (10 µm) and our fabricated dipole (20 µm). However, since the THz wave wavelength is 330 µm, both dipoles can be treated as “short dipole”. In order to correctly illuminate the center gap, the receiver is biased first to maximize the photocurrent measured in the THz receiving PC switch. After optical alignment the receiver is connected to the lock-in amplifier and the silicon lens at the back of the receiver (Fig. 1 (a)) is aligned for maximum signal. Results for the commercial receiver are shown in green in Fig. 2 (a). The pulse has sharp sub-picosecond features that translate into hyper-THz frequency components in the frequency domain (Fig. 2 (b)). The receiver PC switch is then replaced with our PC switches and the results are measured again. For each measurement the silicon lens is aligned to obtain the maximum signal. As it is seen in Fig. 2 (b) the ordinary dipole on GaAs has lower bandwidth. The positive and negative peaks of the detected pulse depend on the photocarrier rise time and fall time respectively56. The rise time in GaAs is similar to that of LT-GaAs and therefore the GaAs PC switch can detect the positive side of the THz pulse. However, due to long photocarrier fall time, the negative peak of the pulse is not captured completely and the results are relatively low in frequency and amplitude. Short photocarrier fall time in LT-GaAs enables the PC switch to capture both positive and negative peaks and this enhances the bandwidth and sensitivity of the device as seen in Fig. 2 (blue curve). The commercial device shows better performance compared to our conventional (not nanoplasmonic) dipole on LT-GaAs due to antireflection coating which enhances the optical coupling of its gap by roughly 30%.

Fig.2 (a) The same THz pulse is measured with: plasmonic interlaced structure on GaAs (peak to peak 20.5nA), commercial PC switch (peak to peak 10nA), dipole on LT-GaAs (peak to peak 2nA), and dipole on GaAs (peak to peak 0.5nA). (b) Signal power spectrum (20log10 (|F(I)|)+cte) in dB. Noise level is -80dB. The nanoplasmonic PC switch shows superior performance compared to the other devices. As seen in Fig. 2, the detected signal in the plasmonic PC switch is double that of commercial device on LT-GaAs, one order of magnitude higher than our LT-GaAs PC switch, and around 40 times more than that of ordinary 20 µm dipole on GaAs.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 12

There are two mechanisms that result in the enhanced response for GaAs. Higher detection bandwidth is obtained due to short carrier sweep-out time which causes deeper photocarrier density modulations. There is an active region adjacent to each electrode that defines the number of photocarriers that are collected. This active region is equivalent to the photocarrier lifetime multiplied by saturation velocity. For LT-GaAs with approximately 0.8 ps carrier lifetime and 1.3×107 cm/s electron saturation velocity57 the active distance from the electrodes is around 100 nm. This implies that photocarriers that are generated further than 100 nm from the edge of the electrodes are on average more likely to recombine before reaching the electrodes. On the other hand for GaAs with more than 200 ps carrier lifetime, the photocarriers are yet present long after the THz field is gone. This results in reduced bandwidth and blurring of the sharp peaks in the detected signal. Transit time across the 100 nm gap between the fingers in the interlaced structure is much shorter than their original lifetime. Based on GaAs electron saturation velocity, the 100 nm gap size is just about the right size to artificially mimic the 0.8ps carrier lifetime of the LT-GaAs. The nanoplasmonic resonances in the slits lead to high total optical transmission that increases the efficiency. The 100 nm gap size also increases the THz field intensity across the gap and this increases the detected current. Secondary peaks appear approximately 6 ps after the main peak in the detected signal, corresponding to the roundtrip time of the THz wave internal reflections inside the 350 µm thick GaAs wafer. The longer tail in the detected signal is in part caused by longer lifetime photocarriers that are excited deeper than 200 nm into the substrate. Due to longer distance from the gold electrodes, these photocarriers (a small fraction of around 10%) are swept out later than the carriers at the surface. To design and study an interlaced structure that is resonant with the incident beam, we used FDTD simulations (FDTD Lumerical software, mesh accuracy 1 nm×1 nm) and searched for the resonant transmission peak with finger and gap width variations. In the simulations, the periodicity was swept from 300nm to 1000 nm in 100 nm steps, the gap width was swept from 50 nm to 200 nm in 50 nm steps, and the thickness was swept from 50 nm to 200 nm in 50 nm steps. Based on this analysis, the best dimensions were chosen as determined by maximal absorption in the semiconductor. Fig. 3 (a) shows the 2D E-field profile of a single finger, in a periodic structure with period of 700 nm which results in 100 nm gaps between the fingers (Fig. 1 (c)). Here z is the vertical co-ordinate to the substrate; x is the transverse co-ordinate of the periodic structure perpendicular to the axis of the slits as shown in Fig. 1 (b). When the optical pulse illuminates the structure from the top, as shown in Fig 3(a), the optical E-field excites the surface plasmons of the gold electrode surface. The structure periodicity and geometry are chosen in such a way to trigger the resonant enhanced transmission from the Au-Air region of the structure58. This significantly enhances the optical transmission through the 100 nm gap based on gap plasmons behavior for a metallic slit59. The fringed decaying E-field that is present due to edge effects and high curvature of the structure is then absorbed and attenuated in the substrate based on the exponential Beer-Lambert profile.

ACS Paragon Plus Environment

Page 7 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

As seen in Fig. 3 (b), the structure shows a peak in the substrate E-field loss at excitation wavelength of 830 nm. E-field loss is the difference between the optical transmissions at the substrate surface (z=0 nm) and depth of 200 nm. 830 nm is the center wavelength of the excitation pulse generated by the Ti-Sapphire femtosecond laser (pulse width = 80 fs and 80 MHz repetition rate). The electric field loss in 200 nm depth of substrate (y axis) is used as a criterion of transmission in each frequency.

Fig. 3. FDTD simulation results. (a) 2D E-field profile (log10 (|E|2) scale) of a single finger cross section. The structure is repeated with 700 nm periodicity for ten periods in the gap (Fig. 1 (c)). (b) The electric field loss spectrum in 200 nm depth of the substrate. The reason for calculating the loss to a depth of 200 nm was to focus on the fast photocarriers swept out by the THz field that decays as the square of depth. We have estimated that a depth of 200 nm accounts for 90% of the photocarriers swept out within less than 2 ps, and therefore, the depth to 200 nm is used to ensure that this surface layer dynamics is accurately captured. Palik permittivity values were used for GaAs60 and Johnson and Christy values were used for gold61. To confirm the presence of nanoplasmonic resonances, we measured the reflection of the interlaced structure for two different optical excitation polarizations ( R⊥ ; the reflection when the polarization is perpendicular to the slits and R|| ; the reflection when the E-field polarization is parallel with the slits). We then compared the results with FDTD simulation results. Based on the simulations the ratio R|| / R⊥ was found to be 2.6. This ratio was measured to be 2.1 for our fabricated interlaced structure. The measured polarization dependent transmission indicates a resonant transmission. The measured polarization dependent transmission indicates a resonant transmission. As seen in Fig. 4, the measured reflection shows reasonable agreement with the simulation results. The mismatch between the theory and measurement can be due to surface roughness that is induced in FIB process. The roughness in the gold surface can induce scattering that weakens the effect62. It is likely that there is some level of Ga contamination from the FIB process that would influence both the optical properties and the electrical properties of the samples measured. While not seeing pronounced changes in the conductivity and reflectivity during routine measurements, we have not characterized the influence of contamination fully. Based on previous

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 12

experimental reports63, however, for thickness of damaged layers in GaAs (maximum of 40 nm depth for 30 keV and 1000 pA) the Ga ion contamination depth with our milling settings (40 keV and 57 pA) should be less than 10 nm.

Fig. 4. The polar plot of the reflection ratio. The dotted blue line is based on the FDTD result and the red squared dots are the measured results. Unlike the THz transmitting PC switch, where the gap is biased by the DC voltage supply, the THz receiving PC switch is biased by the received THz waves. The 100 nm gaps between the fingers can lead to a significant dark current in the THz emitting PC switch, which can burn the device. The fingers for each of the dipole electrodes are electrically parallel and thus the dark current for each gap is added. Consequently, using the same interlaced nanoplasmonic structure as a THz emitter will likely give lower maximum powers. A better solution for the transmitter is to divide the DC bias between the fingers, using a series electrical configuration. This is possible with having parallel strips (or rods) in the gap instead of interlaced electrodes64, 65. While dark current is not a notable concern for THz receiving PC switches, capacitance of the narrow gap can set an upper limit for detection bandwidth. Increasing the number of fingers or reducing the gap size intensifies this undesired capacitive effect. For our design with 10 periods, the capacitive time constant is estimated to be less than 180 fs. This induces a -3 dB point at 550 GHz. Considering the generated THz pulse spectrum, this can explain the relative decrease in detection enhancement for frequencies higher than 550GHz (red curve in Fig. 2 (b)). Such capacitive bandwidth limitation may be further improved by engineering the THz impedance of the dipole. This requires more complex design relative to common single dipole structures22. On the fabrication side, the molecular beam epitaxy (MBE) process to grow LT-GaAs is not necessary as simple GaAs wafers are used. Also the high resolution photolithography that might be needed for 4-5 µm center gap dipoles can be replaced with lower resolution photolithography processes. However, the nanostructure itself needs to be fabricated via focused ion beam machines. The fabrication is not as challenging as many multilayered nanostructure fabrications; the continuous zigzag like pattern can be sputtered at one run. One might consider direct e-beam writing as another fabrication approach that is free from Ga contamination.

ACS Paragon Plus Environment

Page 9 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

The interlaced nanoplasmonic THz-receiving PC switch shows superior performance relative to its LT-GaAs and GaAs peers. The device uses the plasmonic resonances of the interlaced structure to efficiently couple the excitation pulse into narrow 100 nm gaps. These gaps sweep out the generated photocarriers in the subpicosecond time range allowing comparable bandwidth operation on otherwise long carrier lifetime GaAs. Future work includes using higher resistance substrates to improve the output power level in the THz transmitting PC switches for both 810 nm and 1550 nm optical pumping. Supporting Information Additional figures (Figures S1 and S2). This material is available free of charge via the Internet at http://pubs.acs.org. Acknowledgements The authors acknowledge support from Natural Sciences and Engineering Research Council (NSERC) Canada. References (1) Huber, A. J.; Keilmann, F.; Wittborn, J.; Aizpurua, J.; and Hillenbrand, R., Nano Lett., 2008, 11, 3770. (2) Krishnakumar, V.; and Nagalakshmi, R.; Nano Lett. 2008, 8, 3882–3884. (3) Lee, W. M.; and Hu, Q., Opt. Lett., 2005, 30, 2563. (4) Beard, M. C.; Turner, G. M.; and Schmuttenmaer, C. A., J. Phys. Chem. B, 2002, 106, 7146–7159. (5) Vitiello, S.; Coquillat, D.; Viti, L.; Ercolani, D.; Teppe, F.; Pitanti, A.; Beltram, F.; Sorba, L.; Knap, W.; and Tredicucci, A., Nano Lett., 2012, 12, 96–101. (6) Horiuchi, N; Nat. Phot.; 2012, 6, 82–83. (7) Mandal, P. K.; and Chikan, V, Nano Lett., 2007, 7, 2521–2528. (8) Tonouchi, M., Nat. Phot., 2007, 1, 97. (9) Hu, Q., Nat. Phot., 2009, 3, 64. (10) Yu, N.; Wang, Q.; Capasso, F., Laser and Phot. Rev., 2012, 6, 24. (11) Dragoman, T. D.; Dragoman, M., Prog. in Q. Elec., 2004, 28, 1-66. (12) Brown, E. R.; Smith, F. W.; McIntosh, K. A., J. Appl. Phys., 1993, 73, 1480. (13) Park, S.; Jin, K.; Yi, M.; Ye, C.J.; Ahn, J.; and Jeong, K., Nano Lett., 2012 , 6, 2026–2031. (14) Large, N.; Abb, M.; Aizpurua, J.; and Muskens, O., Nano Lett., 2010 ,10 ,1741–1746. (15) Heshmat, B.; Pahlevaninezhad, H.; Beard, M. C.; Papadopoulos, C.; and Darcie, T.E., Opt. Exp., 2011, 19, 15077-15089. (16) Bertulis, K.; Krotkus, A.; Aleksejenko, G.; Pačebutas, V.; Adomavičius, R.; and Molis, G., App. Phys. Lett., 2006, 88, 201112.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 12

(17) Tan, H.; Jagadish, C.; Korona, K.; Jasinski, J.; Kaminska, M.; Viselga, R.; Marcinkevicius, S.; and Krotkus, A., IEEE J. of Sel. Top. in Q. Eelec., 1996, 2, 66. (18) Chen, H. T.; Padilla, W. J.; Zide, J. M.; Bank, S. R.; Gossard, A. C.; Taylor, A. J.; and Averitt, R. D., Opt. Lett., 2007, 32, 1620. (19) Roehle, H.; Dietz, R. J. B.; Hensel, H. J.; Böttcher, J.; Künzel, H.; Stanze, D.; Schell, M.; and Sartorius, B., Opt. Exp., 2010, 18, 2296-2301. (20) Alu, A.; and Engheta, N., IEEE Trans. on Ant. and Prop., 2007, 55, 3027-3039. (21) Chen, Y.; Williamson, S.; Brock, T.; Smith, F. W.; and Calawa, A. R., Appl. Phys. Lett., 1991, 59, 1984. (22) Tanoto, H.; Teng, J. H.; Wu, Q. Y.; Sun, M.; Chen, Z. N.; Maier, S. A.; Wang, B.; Chum, C. C.; Si, G. Y.; Danner, A. J.; and Chua, S. J., Nat. Phot., 2012, 6, 121–126. (23) Park, H. R.; Bahk, Y.; Ahn, K.; Park, Q.; Kim, D. S.; Moreno, L.; Vidal, F. J.; and Abad, J.B., ACS Nano, 2011, 5, 8340. (24) Chang, C.; Sharma, Y. D.; Kim, Y.; Bur, J. A.; Shenoi, R. V.; Krishna, S.; Huang, D.; and Lin, S., Nano Lett., 2010, 10, 1704–1709. (25) Hofmann, C. E.; Abajo, F. J. G.; and Atwater, H. A., Nano Lett., 2011, 11, 372–376. (26) Fan, P.; Chettiar, U.; Cao, L.; Afshinmanesh, F.; Engheta, N., and Brongersma, M., Nat. Phot., 2012, 6, 380-385. (27) Zhai, T.; Zhang, X.; Pang, Z.; Su, X.; Liu, H.; Feng, S.; and Wang, L., Nano Lett., 2011, 11, 4295–4298. (28) Weber-Bargioni, A.; Schwartzberg, A.; Cornaglia, M.; Ismach, A.; Urban, J. J.; Pang, Y.; Gordon, R.; Bokor, J.; Salmeron, M. B.; Ogletree, D. F.; Ashby, P.; Cabrini, S.; Schuck, P. J, Nano Lett., 2011, 11, 1201. (29) Shimizu, T. K.; Pala, R. A.; Fabbri, J. D.; Brongersma, M. L.; and Melosh N. A., Nano Lett., 2006, 6, 2797-803. (30) Yu, Z.; White, J. S.; Barnard, E. S.; Brongersma, M. L.; and Fan, S.; Nano Lett., 2009, 9, 235–238. (31) Diest, K.; Dionne, J.; Spain, M.; and Atwater, H.; Nano Lett., 2009, 9, 2579. (32) Ginzburg, P.; Nevet, A.; Berkovitch, N.; Normatov, A.; Lerman, G. M.; Yanai, A.; Levy, U.; and Orenstein, M., Nano Lett., 2001, 11, 220-224. (33) Lin, T. R.; Chang, S. W.; Chuang, S. L.; Zhang, Z.; Schuck, P. J., App. Phys. Lett., 2010, 97, 063106. (34) Lerman, G; Yanai, A.; and Levy, U., Nano Lett., 2009, 9, 2139-2143. (35) Berini, P.; Charbonneau, R.; and Lahoud, N.; Nano Lett., 2007, 7, 1376–1380. (36) Fang, Y.; Li, Z.; Huang, Y.; Zhang, S.; Nordlander, P.; Halas, N. J.; and Xu, H., Nano Lett., 2010, 10, 1950– 1954. (37) Oulton, R.; Sorger, V.; Genov, D. A.; Pile, D. F. P.; Zhang, X., Nat. Phot., 2008, 2, 496. (38) Sorger, V. J., Ye, Z.; Oulton, R. F.; Wang, Y.; Bartal, G.; Yin, X.; and Zhang, X.; Nat. Comm., 2011, 2, 331.

ACS Paragon Plus Environment

Page 11 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters (39) Tanemura, T.; Balram, K. C.; Ly-Gagnon, D.; Wahl, P.; White, J. S.; Brongersma, M. L.; and. Miller, D. A. B, Nano Lett., 2011, 11, 2693–2698. (40) Chettiar, U. K.; Garcia, R. F.; Maier, S.; and Engheta N., Opt. Exp., 2012, 10, 16104-16112. (41) Hsieh, B-Y.; Jarrahi, M., J. of App. Phys., 2011, 109, 084326. (42) Genevet, P.; Tetienne, J.; Gatzogiannis, E.; Blanchard, R.; Kats, M. A.; Scully, M. O.; and Capasso F.; Nano Lett., 2010, 10, 4880–4883. (43) Knight, M. W.; Grady, N. K.; Bardhan, R.; Hao, F.; Nordlander, P.; and Halas, N. J., Nano Lett., 2007, 7, 2346– 2350. (44) Grady, N. K.; Knight, M. W.; Bardhan, R.; and Halas, N. J., Nano Lett., 2010, 10, 1522–1528. (45) Xiang, S.; Nahata, A., IEEE Trans. on Ter. Sci. and Tech., 2011, 1,412-415. (46) Matsui, T.; Agrawal, A.; Nahata, A.; and Vardeny, Z. V., Nat. 2007, 446, 517-520. (47) Nguyen, T. D.; Vardeny, Z. V.; and Nahata, A., Opt. Exp., 2010, 18, 25441-25448. (48) Sondergaard; T., Bozhevolnyi; S. I., Novikov; S. M., Beermann; J., Devaux; E., and Ebbesen, T. W., Nano Lett, 2010, 10, 3123–3128. (49) Ebbesen; T. W., Lezec; H. J., Ghaemi; H. F., Thio; T., and Wolff, P. A, Nature, 1998, 391, 667–669. (50) Nagpal; P., Lindquist; N. C., Oh; S. H., and Norris, D. J., Science, 2009, 325, 594–597. (51) Seo; M. A., Park; H. R., Koo; S. M., Park; D. J., Kang; J. H., Suwal; O. K., Choi; S. S., Planken; P. C. M., Park; G. S., Park; N. K., Park; Q. H., and Kim, D. S., Nat. Photon., 2009, 3, 152-156. (52) Novitsky; A, Zalkovskij; M, Malureanu; R, Jepsen; P. U., and Lavrinenko, A., Opt Lett., 2012, 15, 3903-5. (53) Bulgarevich; D. S., Watanabe; M, and Shiwa; M, New J. of Phys., 2012, 14, 0530001. (54) Seo; M., Kyoung; J., Park; H., Koo; S., Kim; H., Bernien; H., Kim; B. J., Choe; J. H., Ahn; Y. H., Kim; H., Park; N., Park; Q., Ahn; K., and Kim; D. S., Nano Lett., 2010, 10, 2064–2068. (55) Park; H. R., Park; Y. M., Kim; H. S., Kyoung; J. S., Seo; M. A., Park; D. J., Ahn; Y. H., Ahn; K. J., and Kim; D. S., Appl. Phys. Lett., 2010, 96, 121106. (56) Sun, F. G.; Wagoner, G. A.; and Zhang, X. C., Appl. Phys. Lett., 1995, 67, 1656–1658. (57) Hsin; Y. M., Tang; W.B., and Hsu, H.T.; Solid-State Elec., 2003, 49, 295-300. (58) Aguanno; G. D., Mattiucci; N., Bloemer; M. J., Ceglia; D., Vincenti; M. A., and Alù, A., JOSA B, 2011, 28, 253264. (59) Min; Q., and Gordon, R., Opt. Exp., 2008, 16, 9708-9713. (60) Palik; E.D., Handbook of optical constants of solids, (Academic, 1985). (61) Johnson; P. B., and Christy, R. W., Phys. Rev. B, 1972, 6, 4370-4379.

ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 12

(62) Cai, W.; and Brongersma, M. L., Nat. Nanotech., 2010, 5, 485-486. (63) Rubanov; S., and, Munroe, P.R., J. of Microsc. Microanal. 2005, 11, 446–455. (64) Berry; C. W., Hashemi; M. R., Unlu; M., and Jarrahi, M, published online at http://arxiv.org/abs/1209.1680, 2012. (65) Park; S. G., Jin; K. H., Yi; M., Ye; J. C., Ahn; J., and Jeong, K. H., ACS Nano, 2012, 6, 2026–2031.

ACS Paragon Plus Environment