Array Of Cylindrical Cavities Research Articles

Photonic Crystal Emitters for Thermophotovoltaic Energy Conversion

This paper reports the design, fabrication, and characterization of 2D photonic crystal (PhC) thermal emitters for a millimeter-scale hydrocarbon TPV microgenerator as a possible replacement for batteries in portable microelectronics, robotics, etc. In our TPV system, combustion heats a PhC emitter to incandescence and the resulting radiation is converted by a low-bandgap TPV cell. The PhC tailors the photonic density of states to produce spectrally confined thermal emission that matches the bandgap of the TPV cell, enabling high heat-to-electricity conversion efficiency. The work builds on a previously developed fabrication process to produce a square array of cylindrical cavities in a metal substrate. We will present ongoing incremental improvements in the optical and thermo-mechanical properties, the fabrication process, and the system integration, as recently combined with fabrication using novel materials, such as sputtered coatings, to enable a monolithic system.

Solar thermophotovoltaic energy conversion systems with two-dimensional tantalum photonic crystal absorbers and emitters

Solar thermophotovoltaic (STPV) systems convert solar energy into electricity via thermally radiated photons at tailored wavelengths to increase energy conversion efficiency. In this work, we report the design and analysis of a STPV system with 2D photonic crystals (PhCs) using a high-fidelity thermal-electrical hybrid model that includes the thermal coupling between the absorber/emitter/PV cell and accounts for non-idealities such as temperature gradients and parasitic thermal losses. The desired radiative spectra of the absorber and emitter were achieved by utilizing an optimized two-dimensional periodic square array of cylindrical cavities on a tantalum (Ta) substrate. Various energy loss mechanisms including re-emission at the absorber, low energy emission at the emitter, and a decrease in the emittance due to the angular dependence of PhCs were investigated with varying irradiation flux onto the absorber and resulting operating temperature. The modeling results suggest that the absorber-to-electrical efficiency of a realistic planar STPV consisting of a 2D Ta PhC absorber/emitter and current state of the art InGaAsSb PV cell (whose efficiency is only ~50% of the thermodynamic limit) with a tandem filter can be as high as ~10% at an irradiation flux of ~130kW/m2 and emitter temperature ~1400K. The absorber-to-electrical STPV efficiency can be improved up to ~16% by eliminating optical and electrical non-idealities in the PV cell. The high spectral performance of the optimized 2D Ta PhCs allows a compact system design and operation of STPVs at a significantly lower optical concentration level compared with previous STPVs using macro-scale metallic cavity receivers. This work demonstrates the importance of photon engineering for the development of high efficiency STPVs and offers a framework to improve the performance of both PhC absorbers/emitters and overall STPV systems.

An Investigation of Pool Boiling Heat Transfer on Single Crystal Surfaces and a Dense Array of Cylindrical Cavities

The pool boiling heat transfer characteristics of smooth single crystal and densely packed cylindrical cavity surfaces were investigated using two highly wetting fluids, perfluoro-n-hexane (FC-72) and n-hexane. Three single crystal copper surfaces and five undoped single crystal silicon surfaces with different plane orientations were considered. In addition, silicon surfaces with densely packed cylindrical cavities with diameters ranging from 9 to 75 μm, depth ranging from 9 to 20 μm, and spacing ranging from 75 to 600 μm were tested for comparison. It is observed that the copper single crystal surfaces show increasing heat transfer coefficient with decreasing atomic planar density. The single crystal silicon surfaces show increasing heat transfer coefficient with increasing atomic planar density. Plausible molecular scale mechanisms are discussed. In contrast, the silicon surfaces seeded with cylindrical cavities having diameters of 27 μm or less generally yield higher heat transfer coefficients than the single crystal silicon surfaces. A decrease in the cavity spacing results in a larger number of cavities on the surface, and the heat transfer coefficient increases as a result. Cavity depths of 6 and 20 μm result in the same heat transfer coefficient irrespective of cavity diameter. The nucleation site density for the cylindrical cavity surfaces is measured and reported at low superheat using a novel imaging technique.

Demonstration of silver micro-joints for flip-chip interconnect

► Silver micro-joints for flip-chip interconnect are formed between silicon chips and copper substrates by solid state bonding. ► No flux is used during bonding. ► Bonding temperature is compatible with typical reflow temperature of lead-free solders used in industries. ► Solid state bonding occurs when silver and copper atoms are brought within atomic distance so as to share electrons. ► Silver joints do not contain intermetallic compound, avoiding reliability issues associated with its growth. We report formation of silver (Ag) micro-joints for flip-chip interconnect. For demonstration, the micro-joints were made between silicon (Si) chips and copper (Cu) substrates. Two different Ag joint sizes, 40 and 15 μm, respectively, were produced. In experiment, arrays of cylindrical cavities were fabricated in thick photoresist on 2-in. Si wafers that were coated with chromium (Cr) and gold (Au) dual layer. The Si wafers are diced into 6 mm × 6 mm chips. The cavities were filled up with pure Ag by electroplating process. The photoresist was then stripped off. The Si chip with Ag columns was bonded to Cu substrate at 260 °C in 80 mTorr vacuum. During bonding, Ag columns deform and their surfaces conform to and mate with the surface of Cu substrate. Solid state bonding can incur when the Ag atoms and the Cu atoms are brought within atomic distance so that they share electrons. The Ag columns were well bonded to Cu. The ductile Ag joints are able to accommodate the thermal expansion mismatch between Si and Cu. The Ag joints do not contain any intermetallic compound (IMC). Accordingly, this interconnect technology avoids all reliability issues associated with IMC growth in solder-based flip-chip joints. Compared to solders, Ag joints have superior electrical and thermal properties. Pull test results show that the strength of 40 μm Ag joints passed MIL-STD-883H by wide margin.

Electrochemical desorption of self-assembled monolayers for engineering cellular tissues

Adherent cells, cell sheets, and spheroids were harvested noninvasively from a culture surface by means of electrochemical desorption of a self-assembled monolayer (SAM) of alkanethiol. The SAM surface was made adhesive by the covalent bonding of Arg-Gly-Asp (RGD)-peptides to the alkanethiol molecules. The application of a negative electrical potential caused the reductive desorption of the SAM, resulting in the detachment of the cells. Using this approach greater than 90% of adherent cells detached within 5 min. Furthermore, this approach was used to obtain two-dimensional (2D) cell sheets. The detached cell sheets consisted of viable cells, which could be easily attached to other cell sheets in succession to form a multilayered cell sheet. Moreover, spheroids of hepatocytes of a uniform diameter were formed in an array of cylindrical cavities at a density of 280 spheroids/cm 2 and were harvested by applying a negative electrical potential. This cell manipulation technology could potentially be a useful tool for the fabrication and assembly of building blocks such as cell sheets and spheroids for regenerative medicine and tissue engineering applications.

Glass-blown spherical microcells for chip-scale atomic devices

This paper demonstrates spherical vapor cells intended to be used in chip-scale atomic devices. A micro glass blowing process is introduced, in which multiple glass spheres are simultaneously shaped on the top of a silicon wafer and subsequently filled with rubidium. In the presented fabrication process, an array of cylindrical cavities is first etched in silicon. Next, a thin glass wafer is anodically bonded to the silicon wafer. The bonded wafers are then placed inside a furnace set to 850 ° C. At this elevated temperature, the viscosity of the glass is decreased and the heated trapped gas in the cavities expands, thus causing the glass to be blown into spherical cells. Microscopic alkali vapor cells are achieved by evaporation of Rb 87 through a small glass nozzle into the cell cavities. The cells are then sealed by anodic bonding. The fabricated cells are characterized and the presence of rubidium vapor inside the cells is verified by observing an absorption spectrum.

Scattering characteristics of finite arrays of cylindrical cavities in an infinite ground plane

In this paper, we investigate the scattering characteristics of finite one-dimensional (1D) and two-dimensional arrays of cylindrical cavities embedded within an infinite perfect electric conducting ground plane. Fundamental radar cross-section (RCS) features of said arrays are illustrated, with the RCS computed with respect to the scattered (diffracted) field component, obtained directly from the aperture currents. The aperture currents on the finite structure are computed using a hybrid space/spectral-domain MoM formulation. The complete set of (TE,TM) cylindrical eigenmodes are used to represent the unknowns. The dependence of the current distribution on the source polarization is shown for the canonical cavity. Three-dimensional plots of the RCS signature of a 1D finite array illustrate the unique radiation characteristics of such structures, e.g., the constant-directivity surfaces form cones about the array axis in 3D space. The effects of a superstrate layer on the radiation behavior of a 2D finite array are also shown. We find that the layer can in fact either enhance or diminish the radiation, depending on the source polarization. It also tends to reduce endfire radiation in all cases. The scattering signature of large finite arrays is computed approximately by the introduction of an active element factor defined specifically for scattering problems. Using the MoM solution of a reduced window array, we obtain the far-field patterns of larger structures by a simple computation of the array factor. An improvement to this approximation makes use of an exterior equivalent source, which models the localized edge effects. Finally, the use of nonuniform arrays for sidelobe suppression and 2D "masked" structures is illustrated.

A hybrid MoM solution of scattering from finite arrays of cylindrical cavities in a ground plane

In this paper, we propose a novel hybrid space/spectral-domain method-of-moments (MoM) calculation of canonical plane-wave scattering from finite periodic structures comprised of dissimilar cylindrical cavities in an infinite ground plane, covered by a dielectric superstrate. We take full advantage of analyzing this particular type of problem (comprised of a canonical geometry) by employing a set of entire-domain basis functions (waveguide modes), which truly span the Hilbert space in which the desired solution resides. The convergence behavior of a solution using said functions is quite robust. Results show that typically only a few basis functions are required to adequately represent the aperture fields with fidelity. We have also reformulated the conventional magnetic field integral equation in a hybrid fashion, whereby the exterior (unbounded half space) MoM interaction terms have been cast in the spectral domain and the interior (cavity) interactions cast in the space domain directly. The spectral representation allows us to trivially include a superstrate layer in the geometry and additionally leads to an efficient evaluation of the interaction integrals due to an overall reduction in the spatial dimensionality. The interior interactions are readily handled in the space domain, where we take advantage of the orthogonality property of waveguide modes. Most importantly, the technique is significantly more accurate than purely numerical methods, i.e., those in which a discretization of the geometry is necessary.