Electromagnetic metamaterials over the whole THz range achievements and perspectives H.O. Moser, B.D.F. Casse, O.

. Wilhelmi1, B.T. Saw

Singapore Synchrotron Light Source, National University of Singapore, 5 Research Link, Singapore 117603 1 Present address: FEI Electron Optics BV, Achtseweg Noord 5, 5621 GG Eindhoven, The Netherlands Abstract Using modern micro and nanofabrication techniques, the manufacturing of electromagnetic metamaterials (EM3) with structure sizes < 100 m and critical dimensions < 100 nm has become possible. At this size scale, the resonance frequencies of the structures lie in the THz spectral range. We give an overview of the achievements in and the potential of this field, and discuss new developments towards the micro/nanofabrication of EM3 with nanoscale dimensions and achievable resonance frequencies accordingly higher in the THz range. We then address ways to produce 3D EM3 which derive from stacking, eventually combined with tilted and rotated exposure during X-ray lithography. Finally, we address foundry services offered by the LiMiNT facility at SSLS (Lithography for Micro/Nanotechnology) to customers seeking to have their own micro/nanoscale EM3 manufactured.

1. Introduction In 1964, V.G. Veselago theoretically studied the electrodynamics of materials with simultaneously negative values of their electrical permittivity as well as their magnetic permeability [1]. He showed that the arrangement of the electrical and the magnetic field with respect to the wave vector would have to be described by a left-handed Cartesian co-ordinate frame and described a plethora of unusual consequences such as the inversion of the Doppler and Cerenkov effects, Snells law as well as the radiation pressure. Moreover, he described the image-formation of a parallel slab of such material and discussed the change of role of bi-convex and bi-concave lenses. About 30 years later, J.B. Pendry and co-workers showed in 1996 [2] and 1999 [3] how practical structures might be built to exhibit this left-handed property. As these structures were simple and feasible, they marked the onset for a quest to experimentally demonstrate left-handed materials, then also called electromagnetic metamaterials (EM3). First EM3 structures based on the nested split-ring design [3] were built using a shadow mask/etching technique on printed circuit boards [4]. The outer diameter or width was 6.6 mm [4] and 2.62 mm [5], respectively, leading to a resonance frequency of 4.8 GHz and 11.2 GHz. The EM3 property of this material was shown first in a transmission-type experiment [5], [6], later followed by the demonstration of the negative refraction [7]. As for the application to imaging, a review by J.B. Pendry and co-workers had revealed that the resolution that could be achieved with EM3 may well be much better than the usual diffractionlimited resolution achievable with classical optics, eventually leading to the notion of the perfect lens [8], [9]. Continuing on earlier work [6], [7], C.G. Parazzoli et al. soon after demonstrated a plano-concave lens at 14.7 GHz [10]. A further reduction of size to an outer SRR width of 2.22 mm was achieved by D.R. Smith and co-workers [11] who used transmission and refraction experiments to determine the EM3 behaviour between 13 and 14 GHz. The material was produced by milling of a Cu clad printed circuit board on an automated numerically controlled micro-milling machine as it is in use for the rapid prototyping of printed circuit boards.

While the work reported so far was all done using structures of the split-ring resonator type, H.S. Chen et al. [12] pursued another approach which they called the extended S-geometry SRR which provided multiple passbands due to multiple sizes of the split rings in a frequency range extending from 10.5 to 18 GHz. J.D. Baena et al. analysed [13] spiral-type split-ring resonators with resonance frequencies around 5 GHz that were seemingly similar to Pendrys Swiss roll structure. Simplified split-ring resonators, also called deformed SRR, were studied by Y.J. Hsu et al. in the 11 to 12 GHz range [14] in view of minimising the attainable size of the unit cell of an EM3. Since Veselagos paper, the issue of potential applications of EM3 has challenged the imagination of researchers. The concept of the perfect lens is certainly a very attractive one that is vigourously pursued. Other ideas may be found in work by N. Engheta [15]. Recent short reviews of the field were given by Pendry, Smith and others [16], [17]. Meanwhile, micro- and nano-fabrication have been introduced to manufacture the next generations of smaller and higher frequency structures [18], [19], [20]. Thereby, dimensions can be extended down to the m and nm scales, thus opening up the terahertz range for experiments and applications. Putting in enough development effort may even give access to the petahertz range. In the following, achievements, expectations, and perspectives will be discussed in more detail. 2. Review of existing work on micro/nanofabricated EM3 During the past year, micro/nanofabrication was increasingly used to manufacture EM3 structures in the THz range. T.J. Yen et al. [18] built an array of square nested split-ring resonators from Cu with an outer edge length of 26, 32, and 36 m, respectively, on a 400 m thick quartz substrate, 8%8 mm2 large. They used photolithography enhanced by the photo proliferated process to achieve 3 m thick structures at a good spatial resolution and were able to show the magnetic response of their samples at about 0.8, 0.95, and 1.25 THz by means of frequency-dependent ellipsometry. S. Linden et al. [19] produced square single split-ring resonators from Au with an edge length of 320 nm, gap of 70 nm, 90 nm width of the ring, and 20 nm thickness. The rings were patterned by means of electron beam lithography into PMMA resist spun on an ITO coated glass substrate. The patterned area was 25%25 m2. They measured transmission and reflection spectra for two directions of polarization in the spectral range from 1 to 4 m. When the electric field vector pointed from one gap side to the other they obtained electric coupling of the incoming radiation to the gap and excited oscillations at about 3.2 m wavelength which corresponds to about 95 THz. At SSLS, H.O. Moser et al. [20] manufactured 2.1%2.1 mm2 chips of Ni and Au nested rod-splitring resonators embedded in a matrix of AZ P4620 photoresist. Their structures had metal rods integrated into the design such as to create negative electrical permittivity and to enable the material to become left-handed in the appropriate frequency range given by the negative magnetic permeability. Samples were produced in various sizes, the inner radius of the inner ring ranging from 8.4 to 14 m. Using Fourier transform interferometry, far infrared transmission measurements were performed which showed that transmission peaks occurred close to the spectral location expected from Pendrys formula and from numerical simulation. Fig. 1 shows some of these structures. On the left, there is a close-up of nested circular rod-split-ring structures made of Au electroplated into an AZ P4620 photoresist template. Their resonances were measured to occur in the lower THz range (1-2.5 THz). On the right, PMMA resist was

Fig. 1: Left: Nested circular split-rings with rods made of Au electroplated into an AZ P4620 photoresist template for the lower THz range (1-2.5 THz) (scale bar 20 m). Right: Patterned and developed PMMA resist with nested circular rod-split-ring structures for the higher THz range beyond 50 THz (scale bar 500 nm). The smallest gap ridge measures about 40 nm.

patterned to form about 40 times smaller rod-split-ring structures expected to have their resonances at about 51 THz. 3. Development and perspectives Obviously, further development in the field will aim at structures with higher frequency such as to cover a spectral range as large as possible for experimentation and applications. This will require manufacturing of even smaller structures. Fortunately, all structures discussed so far are amenable to micro/nanofabrication including rod-split-rings, nested split-rings, single split rings, deformed split-rings, spiral split-rings, and extended S-structures. Fig. 2 summarises some of the results that were achieved, of work in progress, and of the perspectives for future development. Pendrys curve for the resonance frequency of a nested circular split-ring resonator shows that inner radii of 30 to 40 nm must be achieved to bring the resonance in the centre of the visible spectral range. The resonance frequencies of the EM3 samples by Moser et al. [20] match Pendrys curve satisfactorily. In the cases of Yen et al. [18] and Linden et al [19], deviations from Pendrys curve do not come unexpected as their geometry differs from the circular nested split-rings. The lower frequency of the structures by Yen et al [18] may be understood by the higher capacitance and inductance of square nested split-rings with other parameters kept constant. In the case of the structures by Linden et al. [19], a simple estimate of the resonance frequency leads to a much higher value than given, so that it may be necessary to consider the interaction with the substrate in order to understand their comparably low resonance frequency. Finally, the open triangles mark structures that are under development at SSLS [21]. 3.1 Fabrication techniques In micro/nanofabrication, the primary pattern generation is done mostly by laser beam or electron beam direct writing depending on required resolution. Typical values of the spatial resolution of a laser writer are 0.8 m, and for an e beam 30 nm. More rarely, but increasingly,

Fig. 2: Inner radius of (inner) ring versus frequency for split-ring resonators over the spectral range from 1 THz to 1 PHz. Symbols mark measured results except for the triangles which belong to structures in progress. The straight line is Pendrys formula for the resonance frequency of a circular nested split-ring resonator.

ion beam direct writing is used promising both, a better resolution and substantially larger penetration depth than the e beam [22]. All these writing pattern generators enable almost arbitrary shapes of the structures to be written, i.e., there is a lot of freedom to design resonators such as the nested split-ring [18], the rod-split-ring [20], the deformed SRR [14], spirals [13], the extended S resonator [12], and more. To produce small quantities of experimental test structures the primary pattern generation might be adequate. However, for larger quantities needed for extended experiments and for applications, the parallel-processing lithography with masks, and further replication techniques with metals and polymers come into play. Typically, the structures are produced in a plastic matrix by creating voids, filling them with metal, and releasing the metal-filled matrix from the substrate. The materials may also be produced on a substrate foil in which case the matrix could be removed by subsequent plasma treatment. Unidirectional structures that are topologically sets of cylinders with a common direction of the axis can be mass produced by hot embossing or injection moulding of the matrix followed by metal deposition into the voids. 3.2 3D concepts Presently, the basic building blocks of EM3 structures are more or less flat metal structures in a plane. Obviously, they are highly anisotropic, and ways of assembling flat EM3 structures such that they offer full coupling for the incident electric and magnetic fields in two or three orthogonal directions have been devised in the GHz range such as examples [4] [6] are showing. This usually involves either building a basic unit cell and then repeatedly accumulating

it, or cutting, folding, piling or stacking larger parts from an array of EM3 structures such as to reach the 3-dimensionality. We want to address an alternative way, namely, to make 3D orthogonal structures within the same matrix. Usually, when doing lithographic exposure, the photon beam impinges on the mask-substrate stack perpendicularly. However, the angle of incidence can be varied so that inclined exposure results. Moreover, the stack can be rotated to different positions between exposures, and it may even be rotated continuously during exposure. This tilting, rotating, and even wobbling in lithographic exposures was proposed many years ago [23], [24]. Fig. 3 shows a few implementations and the resulting structures, and, obviously, the variety of 3D structures that can be produced in this way is large. Applied to the micro/nanofabrication of EM3, this method could be used to build structures in the same matrix layer in which the axes of the resonators cover two or even three perpendicular directions so that the incident field always can couple efficiently to the resonators. The need for stacking would then be reduced to a simple piling up of the matrix layers one by one.

Fig. 3: Illustration of the plethora of structures that can be achieved using tilting, rotating, and wobbling between or during exposures.

Fig. 4 shows a simple example of a double-angle inclined exposure into SU-8 negative resist.

Fig. 4: Double-angle tilted exposure in SU-8. The tilt angle is 30 (scale bar 50 m).

Figs. 5 and 6 illustrate concepts of building inclined structures based on multiply inclined and rotated deep X-ray lithography. Piling up matrix foils with the embedded metal structures would lead to volume EM3.

Fig. 5: Left: Schematic perspective view of a single square split ring with a double-angle tilted exposure. The split-ring at the top is the gold absorber on the mask membrane. Dashed lines are the projection lines for two subsequent tilted exposures. The inclined cylindrical split rings indicate the metal structures obtained from metal deposition into the developed resist structures. Right: Schematic top view of a single square split-ring with three subsequent exposures featuring triple angle rotation plus single angle tilt. The picture shows again the gold absorber on the mask membrane in the centre, and three inclined cylindrical split-rings obtained from metal deposition into the developed resist structures.

Fig. 6: Schematic view of a one-layer array of double-angle tilted cylindrical split-rings. Piling up such layers on top of each other creates three-dimensional EM3. Adjacent layers may be arranged such that the split-rings are parallel or they might be crossed for reduced anisotropy.

Such 3D structures are favourably produced by parallel-processing lithography as hot embossing or injection moulding is no longer possible because these structures cannot be de-moulded. These examples show that there is significant freedom in designing the shape of EM3 if lithography is used for manufacturing them. This may lead to entirely new geometries. 4. SSLS LiMiNT as a foundry and research facility SSLS has installed a complete micro/nanofabrication facility that is effectively a one-stop shop for micro/nanofabrication based on the LIGA process (LIGA is a German acronym and stands for Lithography, Electroplating (Galvanoformung), and Plastic Moulding (Abformung))[25], [26], [27]. The facility is open to users, preferably in a service mode. SSLS is offering to

Fig. 7: At SSLS, the LiMiNT facility is set up in a class 1000 cleanroom and represents a one-stop shop for micro/nanomanufacturing based on the LIGA process.

fabricate customer-defined structures on a contractual basis and is teaching users to reach autonomy as well. 5. Conclusions Construction of electromagnetic metamaterials by means of micro/nanofabrication techniques is in full swing. As the minimum structural dimensions can be expected to reach 10 nm and below, in the future, the frequency range of EM3 is likely to extend to the near infrared and even to the visible. Mass production methods such as parallel processing lithography and replication techniques such as hot embossing and electroplating promise quantities of EM3 sufficient for ample experimentation and the development and marketing of devices. The LiMiNT facility at SSLS offers its services to broaden and speed up this development.

Acknowledgment The authors thank Professor Lim Hock and Gan Yeow Beng of the Temasek Laboratories, NUS, for stimulating discussions. They also thank SSLS LiMiNT staff J.R. Kong and Shahrain bin Mahmood for process development and optimization as well as J.H.W. Wong for help in preparing illustrations. The work was performed at SSLS under A*STAR/MOE RP3979908M, A*STAR 0121050038, and NUS Core Support C-380-003-003-001 grants.

References

[1] V.G. Veselago, The Electrodynamics of Substances with Simultaneously Negative Values of and , Sov. Phys. Uspekhi 10(4), 509(1968) (Usp. Fiz. Nauk 92, 517(1964)). [2] J.B. Pendry, A.J. Holden, W.J. Stewart, I. Youngs, Extremely low frequency plasmons in metallic meso structures, Phys. Rev. Lett. 76, 4773(1996). [3] J.B. Pendry, A.J. Holden, D.J. Robbins, W.J. Stewart, Magnetism from Conductors, and Enhanced Non-linear Phenomena, IEEE Trans. on Microwave Theory and Tech. 47(11), 2075(1999). [4] D.R. Smith, W.J. Padilla, D.C. Vier, S.C. Nemat-Nasser, S. Schultz, Composite Medium with Simultaneously Negative Permeability and Permittivity, Phys. Rev. Lett. 84(18), 4184(2000).

[5] R.A. Shelby, D.R. Smith, S.C. Nemat-Nasser, S. Schultz, Microwave transmission through a two-dimensional, isotropic, left-handed metamaterial, Appl. Phys. Lett. 78(4), 489(2001). [6] K. Li, S.J. McLean, R.B. Greegor, C.G. Parazzoli, M.H. Tanielian, Free-space focused-beam characterisation of left-handed materials, Appl. Phys. Lett. 82(15), 2535(2003) [7] C.G. Parazzoli, R.B. Greegor, K. Li, B.E.C. Koltenbah, M. Tanielian, Experimental Verification and Simulation of Negative Index of Refraction Using Snells Law, Phys. Rev. Lett. 90, 107401(2003). [8] J.B. Pendry, Negative Refraction Makes a Perfect Lens, Phys. Rev. Lett. 85, 183966(2000) [9] J.B. Pendry, S.A. Ramakrishna, Focusing light using negative refraction, J. Phys.: Condens. Matter 15, 6345(2003) [10] C.G. Parazzoli, R.B. Greegor, J.A. Nielsen, M.A. Thompson, K. Li, M.H. Tanielian, Performance of a negative index of refraction lens, Appl. Phys. Lett. 84(17), 3232(2004). [11] D.R. Smith, P. Rye, D.C. Vier, A.F. Starr, J.J. Mock, T. Perram, Design and Measurement of Anisotropic Metamaterials that Exhibit Negative Refraction, IEICE Trans. Electron. E87-C, 359(2004) [12] H.S. Chen, L.X. Ran, J.T. Huangfu, X.M. Zhang, K.S. Chen, T.M. Grzegorczyk, J.A. Kong, Metamaterial exhibiting left-handed properties over multiple frequency, J. Appl. Phys. 96, 5338(2004). [13] J.D. Baena, R. Marques, F. Medina, J. Martel, Artificial magnetic metamaterial design by using spiral resonators, Phys. Rev. B 69, 014402(2004) [14] Y.J. Hsu, Y.C. Huang, J.S. Lih, J.L. Chern, Electromagnetic resonance in deformed split ring resonators of lefthanded meta-materials, J. Appl. Phys. 96(4), 1979(2004). [15] N. Engheta, in Advances in Electromagnetics of Complex Media and Metamaterials, Kluwer Academic, Dordrecht, London, 2003. [16] J.B. Pendry, D.R. Smith, Reversing Light with Negative Refraction, Physics Today 57(6), 37(2004) [17] D.R. Smith, J.B. Pendry, M.C.K. Wiltshire, Metamaterials and Negative Refractive Index, Science 305, 788(2004) [18] T. J. Yen, W.J. Padilla, N. Fang, D.C. Vier, D.R. Smith, J.B. Pendry, D.N. Basov, X. Zhang, Terahertz Magnetic Response from Artificial Materials, Science 303, 1494(2004). [19] S. Linden, C. Enkrich, M. Wegener, J.F. Zhou, T. Koschny, C.M. Soukoulis, Magnetic Response of Metamaterials at 100 THz, Science 306, 1351(2004). [20] H.O. Moser, B.D.F. Casse, O. Wilhelmi, B.T. Saw, Terahertz Response of a Microfabricated Rod-Split-RingResonator Electromagnetic Metamaterial, Phys. Rev. Lett. 94, 063901(2005). [21] B.D.F. Casse, H.O. Moser, O. Wilhelmi, B.T. Saw, Micro- and Nano-Fabrication of Electromagnetic Metamaterials for the Terahertz Range, this conference. [22] Th. Osipowicz, H.O. Moser, Proton beam direct writing of advanced high-aspect-ratio masks for deep X-ray lithography, Grant ARF R-144-000-130-112, Ministry of Education, Singapore, 2005. [23] H.O. Moser, W. Ehrfeld, M. Lacher, H. Lehr, Fabrication of Three-dimensional Microdevices from Metals, Plastics, and Ceramics, Proceedings 1st Japanese-French Congress on Mechatronics, Besanon, Oct. 20-22, 1992, Institut des Microtechniques de Franche-Comt. [24] W. Bacher, P. Bley, H.O. Moser, Potential of LIGA technology for optoelectronic interconnects, in Optoelectronic Interconnects and Packaging, SPIE Critical Reviews of Optical Science and Technology, Vol. CR62, pp. 442-460, 1996. [25] H.O. Moser et al., Status of and materials research at SSLS, Nucl. Instrum. and Meth. B , (2005), in press. [26] E.W. Becker, W. Ehrfeld, D. Muenchmeyer, H. Betz, A. Heuberger, S. Pongratz, W. Glashauser, H.J. Michel, R.v. Siemens, Production of Separation-Nozzle Systems for Uranium Enrichment by a Combination of X-ray Lithography and Galvanoplastics, Naturwissenschaften 69, 520(1982). [27] E.W. Becker, W. Ehrfeld, P. Hagmann, A. Maner, D. Muenchmeyer, Fabrication of microstructures with high aspect ratios and great structural heights by synchrotron radiation lithography, galvanoforming, and plastic moulding (LIGA process), Microelectron. Eng. 4, 35(1986).