Open Access
Add to BibSonomy
Abstract
A periodic array of dual-Vivaldi antennas integrated with metal-insulator-metal (MIM) plasmonic waveguides was designed and investigated for its infrared light absorbance efficiency. Full-wave analysis was used to optimize MIM waveguides compatible with parallel and series connected DC leads without sacrificing radiation efficiency. Free-space to MIM waveguide in-coupling efficiency as high as 41% has been obtained in a sub-wavelength unit cell geometry at a wavelength of 1373 nm. Higher efficiency, up to 85%, is predicted with a modified design including a backplane reflector. A nanofabrication process was developed to realize test devices and far-field optical spectroscopy was used as experimental evidence for antenna-waveguide matching.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Coupling light from free space into plasmonic structures with high efficiency and low footprint is critically important for many applications. For example, photonic integrated circuits (PICs) which exhibit great potential in a wide range of technologies including mobile devices, wearable electronics, and the Internet of Things (IoT), are challenged by relatively large footprint. Due to the diffraction limit at the optical wavelengths, contemporary PIC devices occupy large space (compared with state-of-the-art electronic circuits with their copper interconnects) [1]. Clearly, adoption of PIC technology in many applications largely depends on reducing the footprint of the couplers and interconnects.
Plasmonic waveguides were recently suggested as a possible means to resolve this challenge [2–5]. Plasmonic gap waveguides, and specifically metal-insulator-metal (MIM) waveguides, offer minimal modal area, laterally limited almost entirely by the skin depth of their metal layers [6]. Their low crosstalk permits their use in densely packed optical interconnect networks [7]. Although various MIM-based detectors and energy harvesters were reported [8–13], coupling free space power into MIM waveguides is challenging and current solutions achieve in-coupling efficiency lower than 28% [14–17]. Moreover, coupling is typically achieved with grating couplers and adiabatic waveguide tapers, measuring at least several wavelengths in their lateral dimensions. Optimizing and controlling the in-coupling efficiency received so far little attention. Clearly, a new approach for optimizing the in-coupling step is necessary to achieve high overall efficiency. In previous reports, in-coupling of free space radiation to MIM waveguides via a nano antenna was proposed [10] and demonstrated in the far-infrared (10.6 µm) [13] as a means to address the process efficiency. Scaling this concept into the near-infrared will benefit PIC technology as well as many other MIM-based applications, including thermal and solar energy harvesting and detection, nonlinear optical devices, and optical communication.
Here, we present a MIM waveguide design with an improved in-coupling efficiency using an optimized dual-Vivaldi antenna (DVA) structure [18]. The DVA features an ultra-wideband radiation efficiency from 0.78 to 3.23 µm (122% bandwidth) and exhibits high load sensitivity [19], rendering it a particularly suitable platform to achieve impedance matching between the antenna and the load. Additionally, the DVA can readily accommodate MIM waveguides as well as parallel and series connected DC leads without sacrificing radiation efficiency. The scheme, described below, achieves free space to MIM waveguide in-coupling efficiency of 41% at a wavelength of 1373 nm. We present fabricated devices, full-wave analysis results and experimental evidence of absorption to demonstrate efficient in-coupling of light into antenna-matched MIM waveguides. Additionally, we present a modified geometry, fitted with a Gold backplane, which achieves higher efficiency, up to 85%, according to simulation results.
2. Design
The DVA used in our study consists of two tapered slot antennas (Vivaldi antennas) facing each other to produce a broadside radiation pattern (Fig. 1(a)). By overlapping the antenna two halves at the feed points, adding a thin insulating layer in-between, and extruding the overlap area outwards, we realize MIM waveguide “arms” which form together with the DVA a DVA/MIM structure. The arms are designed to feed/receive signals into/from the antenna. Owing to their narrow profile (warm << λ, where warm is the arm width and λ is the excitation wavelength), solid metal arms (without a dielectric material) have little effect on the structural scattering of the antenna when excited by a cross-polarized wave [20,21]. In reception mode, the DVA directs plasmon modes to its terminals where they are efficiently injected into the arms (MIM waveguides).
Fig. 1. The DVA/MIM structure. (a) Schematic diagram. The antenna halves are made of two different metals, M1 and M2. The two halves overlap at the terminals where they are separated by a thin dielectric layer to produce MIM waveguides. (b) CST 3D EM simulation of the DVA terminal connected to the MIM waveguide illuminated by a plane wave with a wavelength of 1373 nm. (c) Port coupled power as a function of its input resistance. Inset shows DVA model and port placement. Power coupling efficiency is over 90% its maximum value for port resistance in the range of 50-165 Ω. (d) The real part of the characteristic impedance of the MIM waveguide as a function of the waveguide width. High impedance matching with DVA achieved for arm width up to 53 nm. Inset shows MIM waveguide model. (e) Simulation of the in-coupling efficiency of the Au/Al2O3/Al DVA/MIM for excited by x- and y-polarized waves. The dash-dotted lines mark the FWHM of the two peaks (41% at 1373 nm and 36.6% at 1803nm).
Finite element method (FEM) simulations were performed using CST Studio Suite [22] to optimize the DVA/MIM geometry for maximum in-coupling efficiency. The structure was first simulated with the arms replaced with discrete ports placed at the terminals and having input impedance, Rport. The modeled antenna has width wa and thicknesses tM1 and tM2 of the two metals denoted M1 and M2, respectively. The metallic structure is set on a SiO2 substrate and the unit cell dimensions are Px×Py = 1790×483 nm. Periodic boundary conditions were applied to the x and y bounding planes of the simulation and “open” boundary conditions were applied to the bounding z planes. A sweep of the port input resistance shows that high power transfer to the port (over 90% the maximum value) is achieved for 165 > Rport > 50 Ω (Fig. 1(c)).
Next, a MIM waveguide with dimensions warm, tM1, tI, tM2, and larm (width, layers thickness, and length, respectively) was simulated (Fig. 1(d)) to identify a specific design compatible with the antenna in geometry and impedance. The characteristic impedance of the waveguide was calculated as
3. Fabrication
To validate the design, DVA and DVA/MIM antenna arrays were realized by a three-step electron-beam lithography (EBL) process (Fig. 2(a)). A periodic design was applied to generate Bragg constructive interference to increase device overall cross-section. In Step I, alignment marks were patterned for each antenna array. These are essential to achieve high alignment accuracy between the following steps in the presence of beam drift (measuring ∼20 nm/min in our system). In Step II, all M1 (gold) elements were patterned in alignment with the existing marks. A 2.4 nm Al2O3 coating was then applied to the sample by atomic layer deposition (ALD). The Al2O3 layer was characterized by laser ellipsometry on a flat area on the sample. In Step III, M2 (aluminum) elements were patterned in alignment with the same marks. Misalignment was found to be under 10 nm on most arrays. Figure 2(b) shows the workflow of an EBL step: 1) spin-coating, 2) exposure, 3) development, 4) evaporation, 5) lift-off, and 6) ALD (on step II only). Figure 2(c) shows a dark-field optical microscope image of a DVA/MIM array. The four patterns at the corners are the twice exposed alignment marks (exposed in steps II and III). Figure 2(d) shows a false-colored SEM image of the DVA/MIM elements. The bright lines indicate an overlap of both metals. Additional overlap areas are observed at the base of the DVA. These have little impact on the optical response of the array since the field intensity is low in that area and they are fed with very wide terminals leading to high impedance and mode mismatch.
Fig. 2. Three-step EBL process used to pattern the DVA/MIM arrays. (a) Illustration of the EBL steps – patterning of the alignment marks, the first metal layer (gold), and the second metal layer (aluminum). (b) The EBL process workflow. (c) Dark-field optical microscope image of the DVA/MIM array. The side length of the array is 100 µm. (d) False-colored SEM image of the DVA/MIM array. Gold areas colored in yellow, aluminum in gray and the substrate in teal. The brighter areas are where the metals overlap. Scalebar: 200 nm.
4. Experiments
The effect of the MIM arms on the antenna optical properties was studied by a far-field optical scattering experiment using large nano antenna arrays (100 µm × 100 µm). The arrays (shown in Fig. 3(a)) were excited by a tunable 1550 nm laser beam normally incident on the sample, which was positioned at the beam waist. The chosen horizontal period of 1.79 µm satisfied Bragg’s condition for constructive interference and produced 1st order diffraction lobes at angles of ±60° to the main beam reflection and transmission lobes. An integrating sphere InGaAs power sensor was placed at a distance of 55 cm from the array, well within the far-field region, to collect the scattered lobe.
Fig. 3. Far field scattering results. (a) False color SEM images of open-circuited and short-circuited arrays in gold, arms arrays in gold and aluminum, and a gold/alumina/aluminum DVA/MIM array. Scale bar is 1 µm. (b) Detector power in 1st order Bragg lobe divided by input power vs. excitation wavelength. (c) Corresponding far-field simulation results (far-field lobe power divided by input power).
Scattering results from open-circuited antenna, short-circuited antenna (Au, Au arms, and Al arms), and MIM-antenna arrays are markedly different from each other as evident in the spectral response curves in Fig. 3(b). The open-circuited Au array exhibits high scattering monotonically decreasing as a function of wavelength while the short-circuited Au array has a lower scattering and monotonically increasing response. These results echo the results of Yifat et al. [19] who compared the scattered power from open-circuited and short-circuited arrays. The differences between the experimental and simulation results in open-circuited arrays is attributed to the high sensitivity of the spectral response of the device to the gap width. Two additional short circuit antenna designs were investigated. These were designed with “arms” having the same lateral geometry as the matched array and are made of a single metal film (gold or aluminum). As expected, the arrays with the arms exhibited similar scattering curves to that of the short-circuited array (without arms). The response of the matched MIM waveguide consisting of two metal halves (gold and aluminum) and an alumina layer exhibited markedly lower scattered power compared to the open-circuited and short-circuited arrays. The lower scattering is attributed to the impedance matching of the design. In both numerical simulation and experiment, the open-circuited, short-circuited, and DVA/MIM arrays exhibit distinct features, while the different varieties of short-circuited configurations, with and without arms, are closely related in shape and magnitude, and both display higher scattering as compared to the DVA/MIM array. This affirms the conclusion that the difference in geometry (the presence of arms) is unimportant relative to the impedance matching between the antenna and the waveguides. Besides validating the experimental results, the power loss within the MIM waveguide, evident in the simulation sheds light on the underlying absorption mechanism.
The simulation shows that a modified design, in which the antenna is set over a Gold backplane separated by a λ/4 SiO2 spacer, achieves significantly improved efficiency, up to 85%, with parameters: λ = 1363 nm, Px = 1790 nm, Py = 483 nm, wa = 483 nm, tM1 = 40 nm, tI = 3 nm, tM2 = 40 nm, warm = 80 nm, SiO2 spacer thickness: 500 nm, Gold backplane thickness: 100 nm.
5. Conclusions
We have designed an efficient in-coupling scheme of light to MIM waveguides using a DVA. A large-scale DVA/MIM array was fabricated in a three-step electron beam lithography process, achieving very low misalignment. The array was illuminated by a near infrared (1460<λ<1640 nm) laser beam and the power scattered to its 1st order Bragg lobe was recorded. Additional arrays comprising DVAs with different electrical loads: open and short circuit, and single metal “arms” arrays (Gold, Aluminum) having the same geometry as the DVA/MIM were characterized in the same manner. The measured scattered power spectrum of the DVA/MIM array was distinctly lower than that of the open circuit and short circuit arrays. This is consistent with a matched absorptive load as evident in our numerical simulation. The single metal “arms” arrays measurements exhibit close similarity (in shape and magnitude) to the short-circuited array, confirming that such difference in geometry has little effect on the scattering efficiency and losses (e.g. ohmic and defect). The agreement between the experimental results and the simulation results suggests that most of the illuminating beam power was absorbed within the MIM waveguides. Analysis of the simulation indicates that the absorption can be attributed to plasmon decay. However, other processes beyond the realm of classical linear simulation could also contribute to the power absorption or conversion. For example, optical rectification by thermionic emission or electron tunneling could have taken place as well within the MIM junction, considering the high field enhancement, thin insulator layer, and the built-in asymmetry.
The scattering experiment presented here was designed to evaluate the coupling of incident illumination into the MIM waveguides. To achieve this, the parameters of the MIM waveguides were chosen for maximum absorption in order to eliminate back reflection from the unterminated edges of the waveguides. This is not the typical case, where high propagation lengths are desirable. With similar materials, excitation and geometries, considerably higher propagation lengths are possible [25,26].
As mentioned in the Introduction, the combination of the DVA with a MIM waveguide is attractive for PIC applications. It is also particularly attractive in the realm of optical rectennas, devices capable of direct conversion of free space radiation to DC power [25]. Optical rectennas can theoretically achieve 100% efficiency at a single wavelength [26] and solar rectennas, harvesting the Sun’s energy, are only limited by the Landsberg efficiency limit of 93.3% [27,28]. However, state-of-the-art optical rectennas exhibit substantially lower overall efficiencies [8,29,30]. Kale described the overall efficiency η of MIM rectennas as a product of four efficiency factors [32]:
where ηa and ηs are the antenna aperture efficiency and its ohmic losses, respectively, typically exhibiting values between 0.5 and unity. Under favorable conditions, such as proper selection of metals, insulator composition and thickness, and DC bias, the quantum efficiency of a MIM diode ηj may approach unity [29]. However, the coupling efficiency, ηc, accounting for the impedance matching between the antenna and the diode, remains an unmet challenge [8,29,31,33]. In this work, we have demonstrated a platform which optimizes the in-coupling efficiency ηi = ηaηsηc while leaving sufficient degrees of freedom to optimize the remaining efficiency factor, ηj. It is important to note that the MIM waveguide characteristic impedance is only weakly dependent on the insulator thickness and dielectric constant and similar results can be obtained with different metal compositions. Also, the DVA/MIM design parameters and material composition can be modified to obtain a wideband high in-coupling efficiency in the visible and the far-infrared ranges. Moreover, the DVA/MIM device supports diverse circuit designs. The arms can be extended to connect to neighboring elements in parallel without significant impact on the device performance. At the edge of the array they can be connected to DC pads to apply bias voltage or to collect the generated current. Similarly, the top and bottom leads of the device can be used to connect elements in series.To conclude, in this investigation we used a DVA-coupled to MIM plasmonic waveguides as a platform for efficient in-coupling of free space near-infrared radiation into a MIM structure. A methodology was developed to optimize the structure by separately optimizing the antenna with discrete ports, then the load – a MIM waveguide, and finally the complete DVA/MIM device. Simulation results show that up to 85% of the power incident on the array can be coupled into the MIM arms, significantly exceeding previous reports and clearing a path to a multitude of applications.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results in this paper are available upon request to the corresponding author.
References
1. M. Born and E. Wolf, Principles of Optics, 7th ed. (Cambridge University, 1999).
2. J. Ma, D. Zeng, Y. Yang, C. Pan, L. Zhang, and H. Xu, “A review of crosstalk research for plasmonic waveguides,” Opto-Electronic Adv. 2(4), 18002201 (2019). [CrossRef]
3. J. B. Khurgin and G. Sun, “The case for using gap plasmon-polaritons in second-order optical nonlinear processes,” Opt. Express 20(27), 28717–28723 (2012). [CrossRef]
4. S. I. Bozhevolnyi, V. S. Volkov, E. Devaux, and T. W. Ebbesen, “Channel plasmon-polariton guiding by subwavelength metal grooves,” Phys. Rev. Lett. 95(4), 046802 (2005). [CrossRef]
5. E. Moreno, S. G. Rodrigo, S. I. Bozhevolnyi, L. Martın-Moreno, and F. J. Garcıa-Vidal, “Guiding and focusing of electromagnetic fields with wedge plasmon polaritons,” Phys. Rev. Lett. 100(2), 023901 (2008). [CrossRef]
6. M.L. Brongersma and P.G. Kik, Surface Plasmon Nanophotonics, (Springer, 2007).
7. E. Feigenbaum and M. Orenstein, “Modeling of complementary (void) plasmon waveguiding,” J. Lightwave Technol. 25(9), 2547–2562 (2007). [CrossRef]
8. A. Sharma, V. Singh, T. L. Bougher, and B. A. Cola, “A carbon nanotube optical rectenna,” Nat. Nanotechnol. 10(12), 1027–1032 (2015). [CrossRef]
9. J. Kern, R. Kullock, J. Prangsma, M. Emmerling, M. Kamp, and B. Hecht, “Electrically driven optical antennas,” Nat. Photonics 9(9), 582–586 (2015). [CrossRef]
10. S. Grover, O. Dmitriyeva, M. J. Estes, and G. Moddel, “Traveling-wave metal/insulator/metal diodes for improved infrared bandwidth and efficiency of antenna-coupled rectifiers,” IEEE Trans. Nanotechnol. 9(6), 716–722 (2010). [CrossRef]
11. Z. Zhu, S. Joshi, and G. Moddel, “High performance room temperature rectenna IR detectors using graphene geometric diodes,” IEEE J. Sel. Top. Quantum Electron. 20(6), 70–78 (2014). [CrossRef]
12. P. C. D. Hobbs, R. B. Laibowitz, F. R. Libsch, N. C. Labianca, and P. P. Chiniwalla, “Efficient waveguide-integrated tunnel junction detectors at 1.6 µm,” Opt. Express 15(25), 16376 (2007). [CrossRef]
13. B. Pelz, A. Belkadi, and G. Moddel, “Traveling-wave metal-insulator-metal diodes for infrared rectennas,” in IEEE 43rd Photovolt. Spec. Conf. (Portland, OR, 2016), pp. 1034–1038.
14. M. Khodami and P. Berini, “Grating couplers for (Bloch) long-range surface plasmons on metal stripe waveguides,” J. Opt. Soc. Am. B 36(7), 1921–1930 (2019). [CrossRef]
15. C. Chen and P. Berini, “Grating couplers for broadside input and output coupling of long-range surface plasmons,” Opt. Express 18(8), 8006–8018 (2010). [CrossRef]
16. P.S. Davids, R.D. Kekatpure, D.W. Peters, and E. Shaner, Thermal Infrared Plasmonics (Sandia National Labs, 2011).
17. B. Arigong, J. Ding, H. Ren, R. Zhou, H. Kim, Y. Lin, and H. Zhang, “Design of wide-angle broadband Luneburg lens based optical couplers for plasmonic slot nano-waveguides,” J. Appl. Phys. 114(14), 144301 (2013). [CrossRef]
18. Z. Iluz and A. Boag, “Dual-Vivaldi wideband nanoantenna with high radiation efficiency over the infrared frequency band,” Opt. Lett. 36(15), 2773–2775 (2011). [CrossRef]
19. Y. Yifat, Z. Iluz, D. Bar-lev, M. Eitan, Y. Hanein, A. Boag, and J. Scheuer, “High load sensitivity in wideband infrared dual-Vivaldi nanoantennas,” Opt. Lett. 38(2), 205–207 (2013). [CrossRef]
20. L. Novotny, “Effective wavelength scaling for optical antennas,” Phys. Rev. Lett. 98(26), 266802 (2007). [CrossRef]
21. Y. Yifat, Z. Iluz, M. Eitan, I. Friedler, Y. Hanein, A. Boag, and J. Scheuer, “Quantifying the radiation efficiency of nano antennas,” Appl. Phys. Lett. 100(11), 111113 (2012). [CrossRef]
22. CST Studio Suite, “CST Microwave Studio,” http://www.cst.com (2019).
23. J. Huang, T. Feichtner, P. Biagioni, and B. Hecht, “Impedance matching and emission properties of nanoantennas in an optical nanocircuit,” Nano Lett. 9(5), 1897–1902 (2009). [CrossRef]
24. H. B. Michaelson, “The work function of the elements and its periodicity,” J. Appl. Phys. 48(11), 4729–4733 (1977). [CrossRef]
25. J. A. Dionne, L. A. Sweatlock, and H. A. Atwater, “Plasmon slot waveguides: Towards chip-scale propagation with subwavelength-scale localization,” Phys. Rev. B 73(3), 035407 (2006). [CrossRef]
26. R. Zia, M. D. Selker, P. B. Catrysse, and M. L. Brongersma, “Geometries and materials for subwavelength surface plasmon modes,” J. Opt. Soc. Am. A 21(12), 2442–2446 (2004). [CrossRef]
27. R. L. Bailey, “A proposed new concept for a solar-energy converter,” J. Eng. Power 94(2), 73–77 (1972). [CrossRef]
28. S. Grover and G. Moddel, Rectenna Solar Cells (Springer, New York, 2013), p. 47.
29. R. Corkish, M.A. Green, T. Puzzer, and T. Humphrey, “Efficiency of antenna solar collection,” in Proc. 3rd World Conf. Photovolt. Energy Convers. C, 2682 (2003).
30. S. Grover and G. Moddel, “Applicability of metal/insulator/metal (MIM) diodes to solar rectennas,” IEEE J. Photovoltaics 1(1), 78–83 (2011). [CrossRef]
31. B. Berland, Photovoltaic Technologies Beyond the Horizon : Optical Rectenna Solar Cell (NREL, Golden, CO, 2003).
32. B. M. Kale, “Electron tunneling devices in optics,” Opt. Eng. 24(2), 242267 (1985). [CrossRef]
33. E. Donchev, J.S. Pang, P.M. Gammon, A. Centeno, F. Xie, P.K. Petrov, J.D. Breeze, M.P. Ryan, D.J. Riley, and N.M. Alford, “The rectenna device: From theory to practice (a review),” MRS Energy Sustain. 1, E1 (2014). [CrossRef]
