Exploiting plasmonics in active photonic devices
Developments in plasmonics hold great promise, not only for advancing our understanding in fundamental physics but also for facilitating new active and passive photonic devices. Over the past decade a diverse range of plasmonics applications has emerged, which exploit the extraordinary optical properties that can be conveniently manipulated by carefully engineered metallic structures. Even a structure as simple as a metallic nanoparticle, such as a nanosphere, a nanorod, or a core-shell configuration, can serve as a tiny antenna capable of capturing and concentrating light waves through the use of localized surface plasmons (SPs). Periodic metallic structures support both localized and delocalized SPs simultaneously, which provides additional design possibilities. The focus of our recent work has been the physics and applications of plasmonic devices based on periodic metallic structures, mainly in the context of free-space devices.
The most important challenge with respect to plasmonic metallic structures is the heat dissipation that arises as SPs are excited, which has traditionally been regarded as a hindrance. However, wavelength-dependent heat dissipation can be exploited in plasmonic thermal emitters to enable a unique radiation spectrum. It can be considered as a type of absorption: as SPs are excited, the absorption spectrum of plasmonic structures becomes highly dependent on heat dissipation. A plasmonic structure consists of metallic and lossless dielectric material, so that the extinction coefficient is zero and all the absorption is due to heat dissipation. According to Kirchhof's law of thermal radiation, the emissivity is equal to the absorptivity and therefore also closely related to the heat dissipation. We have shown that the emissivity can be manipulated by changing the geometrical parameters that affect the plasmonic properties of the structure.1
Figure 1 shows the simulated thermal radiation spectra of some plasmonic structures. The black and red solid lines represent the thermal radiation of the T-shaped array and plasmonic multilayer structure, respectively. The ideal black-body radiation at 500°C is shown as a dashed line. The plasmonic multilayer structure gives three thermal radiation peaks at 2.2, 3.1, and 6μm. In contrast, the radiation spectrum of the T-shaped array shows a single peak around 3.5μm because unwanted resonant modes are suppressed.
From the point of view of thermal emitter applications, it does not matter whether the absorption is due to a metal or other materials: all absorbed radiation will eventually contribute to the emission. However, in solar cell applications it is very important which component of the plasmonic structure is absorbing.
Plasmonic structures have been widely used for light harvesting. The plasmonic multilayer structure of an amorphous silicon (a-Si) solar cell significantly enhances absorption at λ=550nm, where λ is the wavelength of light (see Figure 2). However, this is accompanied by heat dissipation due to the SP resonant nature of the structure. In the system we studied, the metallic material of the plasmonic multilayer structure absorbs ∼15–20% of the total sunlight impinging on the surface, whereas the active layer only absorbs ∼50%. The absorption by the metal gives rise to further heat dissipation, which does not contribute to electron and hole (charge carrier) generation. In other words, the absorptance and hence the external quantum efficiency of an a-Si solar cell is usually overestimated because the measured quantities do not take into account the wasted absorption and heat dissipation of the plasmonic metal structure.
Still, the unique properties of plasmonic structures attract substantial interest among researchers. It has been shown that the spontaneous emission rates and internal quantum efficiencies of LEDs can be increased by using plasmonic structures.4 However, many important issues concerning superbright plasmonic LEDs remain. Here, we focus on the far-field distribution of a plasmonic LED.
We consider the angle-resolved photoluminescence (ARPL) spectrum of an indium gallium nitride (InGaN)/GaN multiple-quantum-well LED with a cascading plasmonic grating. The plasmonic grating consists of two cascaded silver (Ag) gratings with a half-pitch displacement. The pumping source is transverse magnetic (TM)-polarized and the collected photoluminescence (PL) signal is transverse-electric (TE)-polarized, referred to as the TM-TE state. The ARPL spectrum of the plasmonic LED shows a clear resonance peak redshift from 453.3 to 456.8nm as the collecting angle increases from –35 to –37°: see Figure 3(a). In general, the angular distribution of a conventional LED is Lambertian, that is, the radiant intensity is directly proportional to the cosine of the angle between detection and the surface normal. To compare with a Lambertian light source, the angular scale is usually shown up to 60°, where the radiant intensity becomes half that observed at the surface normal. Here the angular scale is shown up to 75°. In TE-TM mode, there is a dark fringe for incident angles θ=−21° to θ=−30° for emission at a wavelength of 430–470nm: see Figure 3(b). This fringe is in fair agreement with the SP-polariton (SPP) dispersion at the Ag/air interface: see the dashed line in Figure 3(b). This shows that some of the emitted light is coupled to SPPs at the Ag/air interface and dissipates as it propagates, giving rise to a dark fringe. If nothing is done to recover power lost to the SP modes, they then contribute to nonradiative loss.
In summary, the function of photonic devices must be considered when determining the role of heat dissipation from plasmonic structures. While heat dissipation contributes to losses in LEDs and can give rise to misleading performance metrics in solar cells, it can also be exploited to manipulate the emission properties of thermal emitters. We are currently investigating a plasmonic structure with an ultra-broadband absorption spectrum for broadband solar thermal energy applications.
The authors are grateful for the financial support of this research received from the National Science Council of Taiwan under grant NSC 99-2120-M-002-012-.
Chih-Ming Wang is an assistant professor. He received his PhD (2006) in optical science from National Central University, Taiwan. His research interests include plasmonics, nanophotonic devices, and micro-optical systems.
