Real-time process monitoring (RTPM) is a method for semiconductor manufacturing monitoring and tuning using a physical prediction model. It is a fast and nondestructive process excursion measurement method which takes inputs from diffraction-based overlay measurements from YieldStar. The prediction model is created by a physical model which receives standard manufacturing information as input. The prediction capability has been validated in a manufacturing environment experiment with thin film thickness prediction difference less than 3%.

We have developed a microcantilever-based technique for measurement of heat conduction through individual nanowires. We fabricated silicon nitride cantilevers with nominal dimensions of length 100 μm, width 2–6 μm, and thickness 130 nm. Cantilever chips are designed with multiple cantilevers spaced at varying distances. With a reflective aluminum coating of optimized thickness, these bimaterial cantilevers can be used as ultrasensitive thermal sensors capable of measuring very small heat flux through a nanostructure fixed between two cantilevers. The ultrasensitive bimaterial cantilevers designed in this work are not limited to heat conduction measurements, but will also be useful for measuring near-field radiative heat transfer between a sphere, attached to the tip of the cantilever, and a flat plate.

Classical theory of thermal radiation can be used to describe radiative transfer between two objects when the characteristic length scales, such as linear dimensions of the objects and inter-object separation, are much longer than the characteristic thermal wave length given by Wien’s displacement law. In situations where the separation is comparable or smaller than the characteristic wave length, near field effects need to be taken into account. Near field radiative transfer can be enhanced by several orders of magnitude if the materials of the objects support surface phonon polaritons. Although near field radiation between two parallel surfaces have been well studied theoretically, the difficulties in probing radiative transfer between parallel surfaces with nanoscale gaps have prevented meaningful experimental validation. In recent years several experiments have measured near field radiation beyond blackbody limitation between a microsphere and a substrate. In those experiments employing optical beam deflection technique, the sphere has to be placed near the edge of the substrate in order to prevent unintentional chopping of the laser beam by the edge of the substrate. The actual measurements are performed between a sphere and a semi-infinite plane instead of an infinite plane. We report here an improved optical beam deflection system to measure the near field radiative heat transfer between a sphere and a substrate. With the new setup, the sphere can be placed sufficiently far away from the substrate edge, rendering it a better approximation of a sphere and an infinite plane. The experimental results and the numerical prediction using modified proximity approximation will be discussed.

Using an improved experimental setup, we measured near field radiative heat transfer that may benefit TPV up to a value as small as 0.5 nW/K. The experimental data is compared with modified proximity approximation prediction

Bimaterial atomic force microscope cantilevers have been used extensively over the last 15 years as physical, chemical, and biological sensors. As a thermal sensor, the static deflection of bimaterial cantilevers, due to the mismatch of the coefficient of thermal expansion between the two materials, has been used to measure temperature changes as small as 10−6 K, heat transfer rate as small as 40 pW, and energy changes as small as 10 fJ. Bimaterial cantilevers have also been used to measure “heat transfer-distance” curves—a heat transfer analogy of the force-distance curves obtained using atomic force microscopes. In this work, we concentrate on the characterization of heat transfer from the microcantilever. The thermomechanical response of a bimaterial cantilever is used to determine the (1) thermal conductance of a bimaterial cantilever, and (2) overall thermal conductance from the cantilever to the ambient. The thermal conductance of a rectangular gold coated silicon nitride cantilever is Gc=4.09±0.04 μW K−1. The overall thermal conductance from the cantilever to the ambient (at atmospheric pressure) is Ga=55.05±0.69 μW K−1. The effective heat transfer coefficient from the cantilever to the ambient (at atmospheric pressure) is determined to be ≈3400 W m−2 K−1.

Bi–material atomic force microscope cantilevers have been used extensively over the last 15 years as physical, chemical, and biological sensors. As a thermal sensor, the static deflection of bi–material cantilevers due to the mismatch of the coefficient of thermal expansion between the two materials has been used to measure temperature changes as small as 10−5 K, heat transfer rate as small as 40 pW, and energy changes as small as 10 fJ. Bi–material cantilevers have also been use to measure “heat transfer - distance” curves a heat transfer analogy of the force–distance curves obtained using atomic force microscopes. In this work, we concentrate on characterization of heat transfer from the microcantilever. The two quantities that we focus on are the thermal conductance of the cantilever, Gcant (units WK−1), and the thermal conductance due to microscale convection from the cantilever to the ambient fluid, Gconv (units WK−1). The deflection of the cantilever to changes in its thermal environment is measured using the shift in position, on a position sensitive detector, of a laser beam focused at the tip of the cantilever. By determining the response of the microcantilever to (1) uniform temperature rise of the ambient, and (2) change in power absorbed at the tip, the thermal conductance of heat transfer from the cantilever can be determined. When the experiment is performed at low enough ambient pressure so that convection is unimportant (¡ 0.1 Pa), Gcant can be measured. When the experiment is performed at atmospheric pressure the heat transfer coefficient due to convection from the cantilever can be determined.

Analysis of the composition, strain-relaxation, layer-tilt, and the crystalline quality of InyGa1-yAs/ InP1-xAsx thermophotovoltaic (TPV) diodes grown by metal-organic vapor phase epitaxy (MOVPE) is demonstrated using triple-axis X-ray reciprocal space mapping techniques. [n(0.53)Ga(0.47)s (E-gap = 0.74 eV) n/p junction diodes are grown lattice matched (LM) to InP substrates and lattice-mismatched (LMM) In0.67Ga0.33As (E-gap = 0.6 eV) TP%/ diodes are grown on three-step InP1-xAsx (0 <x<0.32) buffer layers on InP substrates. X-ray reciprocal space maps about the symmetric (400) and asymmetric (5 3 3) reciprocal lattice points (RELPs) determine the in-plane and out-of-plane lattice parameters and strain of the InyGa1-yAs TPV active layer and underlying InP1-xAsx. buffers. Triple-axis X-ray rocking Curves about the LMM In0.67Ga0.33As RELPshow an order of magnitude increase of its full-width at half-maximum (FWHM) compared to that from the LM In0.53Ga0.47As (250 vs. 30 arcsec). Despite the significant RELP broadening, the photovoltaic figure of merits show that the electronic quality of the LMM In0.67Ga0.33As approaches that of the LM diode material. This indicates that misfit-related crystalline imperfections are not dominating the photovoltaic response of the optimized LMM In0.67Ga0.33As material compared with the intrinsic recombination processes and/or recombination through native point defects, which would be present in both LMM and LM diode material. However, additional RELP broadening in non-optimized LMM In0.67Ga0.33As n/p junction diodes does correspond to significant degradation of TPV diode open-circuit voltage and minority carrier lifetime demonstrating that there is correlation between X-ray FWHM and the electronic performance of the LMM TPV diodes. (C) 2008 Elsevier B.V. All rights reserved.