Research  activity  3rd  year  &  final  report  

Research  Unit  (RU)  4  :  Nanotechnologies  and  Nanoelectronic  Materials  

New scientific instrumentation

a) Low-power continuous-wave generation of visible harmonics in silicon photonic crystal nanocavities

One of the most relevant results obtained by Unit 4 during the third year of the project is the experimental demonstration of a silicon nano-device capable of efficient light emission in the visible spectral range by exploiting low-power continuous-wave harmonic generation. The device is based on a high quality factor (high-Q) photonic crystal (PhC) nanocavity realized on a silicon membrane, with optimized in- and out-coupling efficiency (SEM image shown in Fig. 1a). By optically pumping the PhC nanocavity from the top with an infrared laser tuned to resonance with the cavity mode (Fig.1b), we observe a bright dual wavelength emission in the blue/green (450-525 nm) and red (675-790 nm) visible windows (Fig. 1c,d,e). The pump powers necessary to observe continuous-wave harmonic generation are as low as few microwatts, yielding the record conversion efficiencies of ~10−5/W and ~10/W2 for the second- and third - harmonic processes, respectively. Scaling behaviours as a function of pump power and cavity quality-factor are then demonstrated for both second- and third- order processes. Furthermore, successful comparison of measured and calculated emission patterns indicates that third-harmonic is a bulk effect while second-harmonic is a surface-related effect at the sidewall holes. This result demonstrates the very appealing possibility to extend the coherent emission of telecom infra-red lasers to shorter wavelengths in the visible and near ultraviolet. Thus, our results are promising for obtaining practical low-power, continuous-wave and widely tunable multiple harmonic generation on a silicon chip.

References: “Low-power continuous-wave generation of visible harmonics in silicon photonic crystal nanocavities”, M. Galli et al., Optics Express 18, 26613 (2010).

Figure 1

b) Optical bistability in a silicon photonic crystal nanocavity The interest in integrated all-optical logic devices has been triggered by the idea that the elimination of optical-to-electrical signal conversion will lead to the fabrication of very low-power signal processor systems. Therefore, it is reasonable to require an all-optical logic gate to be very small and to operate at very low power. Optical bistability is a fundamental physical phenomenon that makes it possible to realize all-optical logic gates. It implies that the optical response of the component is nonlinear, thus the resonant wavelength and absorption depend on the optical power. However, in traditional devices, the nonlinear effects needed to achieve optical bistability require a high optical power density, which have made optical bistable operation difficult to demonstrate at a reasonably low input power. In this work we demonstrate optical bistable action in a silicon nanocavity with input power as low as few tens of µW. The device is based on a high-Q (30000) low-volume (~λ3) PhC nanocavity designed for operation in the telecom window around 1.55 micron. Figure 2a shows the cavity resonance spectra, measured by means of resonant scattering of laser light, as a function of the incident power. A clear signature of optical bistable behavior, i.e. a sharp decrease in the cavity response giving rise to a typical asymmetric lineshape, is evident as the probing wavelength is scanned upwards across resonance for increasing powers. Optical bistability is also demonstrated by measuring the hysteresis loop for different detuning conditions with respect to cavity resonance. This is shown in Fig. 2b, where the output power is plotted as a function of the input power for cavity detuning in the range 100-200 pm. The opening of a well defined hysteresis loop and bistable operation is demonstrated for a detuned wavelength by only 100 pm, at a power threshold as low as 35 µW. These results are important for the realization of ultra-low power integrated logic gates on an all-silicon platform. References: “H. Gibbs, Optical Bistability: Controlling Light with Light (Academic Press, Orlando, 1985). “Opticl bistable third-harmonic generation in silicon photonic crystal nanocavities”, P. Andrich et al, to be submitted to Applied Physics Letters.

Figure 2

c) Enhanced room-temperature emission at telecom wavelengths in silicon photonic crystal nanocavities The realization of efficient silicon-based light sources is of key importance for the integration of devices with optical and electrical functions on a single chip. In this research activity, Unit 4 demonstrates an all-silicon nanometer-sized light source in a wide spectral range covering the two strategic telecom bands from 1.3 µm to 1.6 µm. The device is based on a modified PhC nanocavity fabricated on a 220 nm tick silicon membrane from SOI (Fig3a inset). A careful theoretical design is applied to engineer the radiation in the far-field such as to achieve efficient out-coupling efficiency in the vertical direction. By optically pumping the nanocavity with a focused laser beam at λ=532 nm, strongly enhanced light emission at wavelengths between 1.3 µm and 1.6 µm is reported at room-temperature is observed (Fig. 3a). Sharp and intense photoluminescence peak corresponding to the resonant modes of L3-type PhC nanocavities are observed to dominate the broad sub-bandgap emission from optically active defects in the crystalline Si membrane. We measure a 300-fold enhancement of the emission from the PhC nanocavity compared to the background due to a combination of far-field enhancement and the Purcell effect. One of the most striking features is that the cavity-enhanced emission has a very weak temperature dependence, namely less than a x2 reduction between 10 K and room temperature, as shown in Fig. 3b. This makes this approach suitable for the realization of efficient room-temperature light sources at telecom wavelengths as well as providing a quick and easy tool for the broadband optical characterization of SOI-based nanostructures. Moreover, our findings demonstrate the potential of PhC nanocavities in enhancing even very weak background emission from residual (unintentional) deep-centers in Si. A much more intense light emission could then be expected by the intentional introduction of a small amount of optically-active recombination centers such as dislocation-loops or point-defects in crystalline silicon. “Room-temperature emission at telecom wavelengths from silicon photonic crystal nanocavities”, R. Lo Savio et al, Applied Physics letters 98, 201106 (2011).

Figure 3

d) Absorption enhancement and light trapping in thin-film silicon solar cells with a periodic pattern

Photovoltaic conversion of solar radiant energy in electrical energy is a hot topic of the last decades, and for many years mono- or polycrystalline silicon solar cells have been the most popular devices. Usually the thickness of the absorbing layer is over 100 microns, and the cost of the silicon is more than a half of the total cost of the device; a significant reduction of the fabrication   costs could be achieved reducing the absorber thickness or changing the material. While traditional “thick” cells employ micrometer-scale patterns that base their action on geometrical optics, the limited thickness of thin-film cells requires the development of nanometer-scale, photonic based trapping structures. These structures can couple the incoming light into the absorbing material with different mechanisms, giving rise to separate coupling regimes that we have investigated by varying the structure parameters. The structure we considered in our work is shown in Fig. 4a. The silicon absorbing layer, of thickness d, lies on a silver back reflector. The front surface of silicon is patterned with a one-dimensional periodic pattern with period a, fill factor f = b/a, and depth h. A filling of silica is assumed to reduce surface recombination losses. The top of the structure is coated with an antireflection (AR) layer, with thickness l and refractive index n. Since our aim is to explore the role of the photonic components on the absorption, we devoted our attention to the functions of the pattern and the coating. The efficiency of the cell is estimated by means of the short circuit current, which, in the approximation that all the electron-hole pairs generated are collected by the electrodes, is simply the integral of the photon flux spectra weighted with the absorbance of the device. The spectral contribution to the short-circuit current calculated for most of the structures previously mentioned are shown in Fig. 4b. Comparing curves (1) and (2), it can be seen that the application of an AR coating on top of an unpatterned structure plays an important role above 2 eV, while the spectral region between 1.12 and 2.75 eV is strongly affected by the patterning [curves (3) and (4)]); above 2.75 eV the patterned structure behaves similarly to the unpatterned coated structure. The difference between shallow etching (3) and deep etching (4) is limited to the energies below 2 eV. In the spectral region above 2.5 eV the patterned cells behave almost like an ideal absorber, therefore, a further increase in the short-circuit current can be achieved only by acting on the spectral region [1.12, 2.5] eV.

Reference: “Light trapping regimes in thin-film silicon solar cells with a photonic pattern” S. Zanotto et al. Optics Express 18, 4260 (2010).

Figure 4

e) Electrical conduction and optical properties of doped silicon-on-insulator photonic crystals

The realization of efficient light sources in Silicon-based systems is of fundamental importance for the integration of devices with photonic and optical functionalities in a single chip, to be adopted in telecom platforms. Unit 4 has realized, in collaboration with the University of St. Andrews - Scotland, a prototypal device consisting in a light-emitting diode, electrically pumped, based on a Silicon photonic-crystal (PhC) nanocavity. This device should allow obtaining a light source of micrometric dimensions, high integration level, and high emission efficiency. The device consists in a p+-p-p+ planar junction realized through B implantation in crystalline silicon. The PhC was realized in the p-region of the device using electron beam lithography and reactive ion etching, with a triangular pattern and a nanocavity optically resonant at wavelengths around 1.5 microns (Fig. 5a). The electrical injection was carried out by metal contacts with CMOS compatible processes. The electrical properties, i.e. the resistance R and the current-voltage I-V curves, have been measured as a function of the doping level and the PhC air- filling fraction. The resistance curves as a function of the above parameters (Fig. 5b) can be explained by the effect of the depletion region around the PhC holes, as due to the Fermi-level pinning of the Si electronic bands. Moreover, for a complete characterization of the trade-off between the electrical properties and the optical losses due to the free carriers, we measured the resonant cavities modes of L3 PhC cavities as a function of the B concentration levels. We found that surprisingly high doping levels, up to 1018 /cm3, are acceptable for practical devices with Q factors as high 4x104. We have thus provided useful guidelines for the design of electrically pumped photonic-crystal nanodevices, based on Silicon and electrically pumped with high emission efficiency in the telecom optical windows.

Figure 5

Prototypes

Luminescent multilayer solar concentrator

In the last year of the Project, Unit 4 started a collaboration with ENI for the development of cost-effective luminescent solar cell concentrators, through the initiative “Along with Petroleum - materiali fotoattivi”. Luminescent solar concentrators (LSCs) are hybrid photovoltaic (PV) devices in which PV cells are applied on the lateral edges of polymeric slabs, as shown in Fig. 6 for a multilayer structure. Inside each concentrator layer, one or more luminescent dyes are dispersed. These materials can absorb sunlight in the UV-vis range and re-emit photons by luminescence, which is guided towards the lateral PV cells. A schematic illustration of a 2-layers LSC device is illustrated in Fig. 6a. Each LSC layer is made of PMMA, in which one luminescent material is dispersed. In the configuration shown, sunlight in the range 350-550 nm is absorbed and emitted luminescence is collected by PV cells at each edge. In this way each device can act as concentrator for both direct and diffuse incident sunlight. In this work c-Silicon PV cells are used for photoconversion. LSCs used in this work are made of slabs of polymethyl methacrylate (PMMA), with front surface dimensions up to 10cm x 10cm and thickness 0.6 cm, doped with one or two organic, luminescent molecules: 9,10-diphenylanthracene (Aldrich, 97%), hereafter denoted as DPA, and 4,7-di-2-thienyl-2,1,3-benzothiadiazole (ENI, 99,9%), hereafter denoted as DTB. These and other dyes with large Stokes shift are synthesized [6] and dispersed in a solution containing PMMA (Altuglas VSUVT clear 100). The slabs are then obtained by a process of compounding and injection molding. Optionally, Ag lateral mirrors and back-reflectors can be applied, to reduce optical losses and improve solar energy conversion, as shown in Figure 7. Here we report the external quantum efficiencies for four LSC monolayer devices, with front dimensions 2.2 × 2.2 cm2, thickness 0.6 cm and one PV cell applied on one edge (in order of increasing conversion efficiency): bare LSC (black line), polished LSC (red line), polished LSC + 3 Ag lateral mirrors (blue line), polished LSC + 3 Ag lateral mirrors + back-reflector (green line).

The characterization of these structures is performed first for the selected materials (matrix, luminescent dyes and PV cells) and then for final devices (LSC + PV cells). Final devices’ characterization includes I-V curve measurements under AM1.5 standard conditions, and external quantum efficiency (EQE) measurements, with spectral and spatial resolution on the LSC’s front surface. From I-V curves, an energy conversion efficiency up to 1.5-2% is extracted. EQE, instead, is the key physical quantity in order to compare different devices and configurations, as shown in Figure 5b. Special care is taken to minimize propagation losses in the devices. In order to reach higher conversion efficiencies, we are harvesting a wider solar spectrum in the VIS–IR range (λ>550 nm) by synthesizing and exploiting different dyes.  

Figure 6

Figure 7