Photonic Crystals-based Hybrid Structures for Laser and Photovoltaic Applications

Peiris, Frank C.

In the last three years, in order to achieve our objective of fabricating structures suitable for laser and photovoltaic applications, we made several interesting discoveries, culminating in a manuscript which has been submitted to the journal of applied physics. Below we list the major landmarks we have accomplished during the past three years; Synthesizing and characterizing a library of constituents based on porous TiO₂ and SiO₂ where their indices of refraction can be tuned to suite a specific application. Fabricating a hybrid structure, based on two photonic crystals and a mesoporous layer, in order to achieve a cavity-based structure for post-growth tuning. Demonstrating a seven period dielectric stack, with alternate layers of TiO₂ and SiO₂, which shows a reflectivity of over 90%. While the details of the first two items have been covered in our first and second year projects, in the past year, we have focused mainly on the third item. As originally envisioned in this proposal, our goal was to create devices that would enhance light capture efficiency in photovoltaic devices. In order to accomplish this, our plan was to couple 1-D photonic crystals (PCs) into photovoltaic materials. Since 1D-PCs exhibit a high reflectivity band due to their forbidden band gap for photons, by tuning this band to be coincident with the absorption spectrum of the photovoltaic material, we planned to harvest more light. In addition to fabricating very high reflective 1-D PCs for this purpose, during the past three years, we have also demonstrated that TiO₂ can be used as both a photovoltaic material as well as one of the constituents of the 1-D PCs, producing structure that is less complex and more efficient. Upon successfully creating a library of TiO₂ and SiO₂ mesoporous films with various indices of refraction, our next task was to stack a sufficient number of TiO₂ and SiO₂ mesoporous layers in order to achieve high reflectivity dielectric mirrors. From our library of materials, we also had to select two mesoporous films of TiO₂ and SiO₂that had a very large contrast in their indices of refraction. In accomplishing this task, one of the major bottlenecks encountered was that the layers that were spin-coated subsequently infiltrated through to the bottom layers. This in turn altered the indices of refraction of the bottom layers. As the reflectivity of the structure is dictated by the contrast of the indices of refraction between the two constituents of the dielectric stack, the infiltration generally reduces the contrast, resulting in a lower reflectivity of the final structure. In order to circumvent this issue, the mesoporous films were partially calcined, leaving behind most of the surfactants in their pores. Once the required number of layers was deposited, we calcined the structure to 300 C in order to remove all of the surfactants and to achieve the required porosity. This procedure allowed us to deposit seven periods (i.e., fourteen layers) in our final composite structure. In Fig.1 we show the indices of refraction of the two constituent materials of the dielectric stack. The two specific films were selected as they yield a very high contrast in their indices of refraction, which in turn will produce a high reflectivity.

Figure 1: Indices of refraction of the two constituent materials of the dielectric stack. The higher index material is the titania mesoporous layer with a porosity of ~30% (left curve) while the lower index material is the silica mesoporous layer with a porosity of ~40% (right curve). By knowing the indices of refraction, next we determined the optical thickness of each layer of the dielectric stack suitable to yield a high reflectivity at a specific wavelength. In addition to selecting two constituents of high index contrast, in order to increase the reflectivity, we wanted to deposit a fairly large number of periods. The maximum number of periods we were able to deposit was seven, and in Fig. 2, we show the experimental and the simulated ellipsometric data for this particular structure. The ellipsometric spectra allowed us to confirm the structural properties of the stack along with their optical properties.

Figure 2: Experimental (symbols) and simulated (solid lines) ellipsometric data for the seven-period dielectric stack. The triangles display Ψ-spectrum while the circles display Δ-spectrum, both of which are related to the Fresnel reflection coefficients of the entire structure.

Both ellipsometry and reflectivity measurements have enabled us to address a series of questions related to the growth of dielectric stacks. By pursuing different combinations of TiO₂ and SiO₂ mesoporous layers, we were able to determine the maximum porosity levels suitable for the structure. Furthermore, post-growth calcine temperatures were optimized after observing the reflectivity of several structures that were calcined at varying temperatures. These studies allowed us to finally obtain structures with very high reflectivity. In Fig. 3, we show the reflectivity for a sample which has seven periods of SiO₂ and TiO₂ mesoporous layers (i.e., fourteen layers). Theoretical calculations have shown that this structure, which has reflectivity of ~90%, can be used in conjunction with photovoltaic materials in order to efficiently harvest light. Several other structures were fabricated in order to achieve a reflectivity band in the green region which may be even better suited for photovoltaic devices. These structures also show a high reflectivity-band and provide evidence that the reflectivity-band can be easily tuned to suite a particular application. Encouraged by our structures grown on silicon substrates, we grew TiO₂ and SiO₂ mesoporous layered structures on conducting glass substrates, and are in the process of integrating electrical connections to these dstructures. We plan to quantify the improvements achievable through this novel device strategy. In the future, we will investigate on how to improve the efficiency by using mesoporous TiO₂, which is one of the constituents of the dielectric mirror, as the active material in photovoltaic devices. This will provide further impetus for the proposed device strategy.

Figure 3: Reflectivity spectrum of a 7-period dielectric stack composed of mesoporous SiO₂ and TiO₂ films.

www.acsprf.org

Reports: B1048489-B10: Photonic Crystals-based Hybrid Structures for Laser and Photovoltaic Applications