Differential scanning calorimetry, pressure calorimetry, and densitometry have been employed to study the relation between volume and enthalpy changes in the melting regime of lipid membranes. We demonstrate a rigid proportional relation between volume expansion coefficient and heat capacity. This result is first shown in densitometric experiments. It implies that calorimetric profiles obey a simple scaling law for the temperature axes in experiments with applied hydrostatic pressure. In a theoretical paper (Heimburg, T. Biochim. Biophys. Acta 1998, 1415, 147-162), we have argued that this relation has far-reaching consequences for the predictiblity of elastic constants from the heat capacity. The proportionality constant between volume and enthalpy changes is found to be independent of the lipid, which has interesting consequences for the calculation of the elastic constants of biological lipid mixtures with unknown composition. We demonstrate this for lung surfactant, which displays a similar relation between volume and enthalpy changes.

We investigated the relaxation behavior of lipid membranes close to the chain-melting transition using pressure jump calorimetry with a temperature accuracy of approximately 10(-3) K. We found relaxation times in the range from seconds up to about a minute, depending on vesicular state. The relaxation times are within error proportional to the heat capacity. We provide a statistical thermodynamics theory that rationalizes the close relation between heat capacity and relaxation times. It is based on our recent finding that enthalpy and volume changes close to the melting transition are proportional functions.

Ultrasonic velocity and heat capacity temperature profiles of various lipid mixtures have been recorded with high accuracy. This included mixtures of phophatidylcholines with different chain length as well as phosphatidylcholine mixtures with diacyl glycerides. Following previous studies relating the heat capacity to the isothermal compressibility of lipids close to the chain melting transition, we found that the measured ultrasonic velocities are very similar to those calculated from the heat capacity. This implies that we are able to determine the compressibility changes from the excess heat capacity and the heat capacity changes from ultrasonic velocity measurements. The sound velocity and heat capacity traces are discussed with respect to the phase diagrams of the lipid mixtures.

Spatial variations of the electronic properties of solar cells are simulated with the help of a multi-diode model. Increasing the local series resistance limits the degradation of the overall performance by reducing losses of the open circuit voltage originating from cell areas with minor electronic quality. At the same time, the fill factor of the device decreases such that an optimum local series resistance for maximum power conversion efficiency is found. The value of the optimum series resistance depends on the degree of electronic inhomogeneity. Thus, optimization rules for spatially uniform solar cells are different from those for nonuniform cells.

An elegant laser tailoring add-on process for silicon solar cells, leading to selectively doped emitters increases their efficiency h by Dh ¼ 0.5% absolute. Our patented, scanned laser doping add-on process locally increases the doping under the front side metallization, thus allowing for shallow doping and less Auger recombination between the contacts. The selective laser add-on process modifies the emitter profile from a shallow error-function type to Gaussian type and enables excellent contact formation by screen printing, normally difficult to achieve for shallow diffused emitters. The significantly deeper doping profile of the laser irradiated samples widens the process window for the firing of screen printed contacts and avoids metal spiking through the pn-junction.

Dedicated to Professor Horst P. Strunk on the occasion of his 65th birthday PACS 72.40.+w, 84.60.Jt, 85.30.De An equivalent circuit model consisting of parallel connected diodes with different electronic quality simulates the electronic properties of solar cells with spatially inhomogeneous material quality. Variations of the local saturation current density result in a degradation of the open circuit voltage, the fill factor and, in consequence, of the overall power conversion efficiency. However, a local series resistance introduced into this network limits this degradation by preventing areas with high saturation current density to dominate the electronic losses of the entire device. Analyzing the integral current/voltage-curves of the networks shows the diode ideality larger than unity to result from resistive limitations to the spatially inhomogeneous current flow.

Spatial variations of the local open circuit voltage in Cu(In,Ga)Se2 solar cells are analyzed by an electron beam induced voltage (EBIV) technique. The major pattern visualized by our EBIV measurements are spatial inhomogeneities on a length scale of between 5 and 20μm. Quantitative evaluation of the EBIV signals shows that the loss of open circuit voltage due to the inhomogeneities is about 100mV. Additional analysis of our samples by energy dispersive x-ray analysis excludes fluctuations of the Ga or Cu content as the source of the inhomogeneities. Instead, the spatial inhomogeneous supply of Na from the glass substrate turns out as a possible origin of inhomogeneities. Spatially resolved secondary ion mass spectroscopy measurements show that the Na content of our Cu(In,Ga)Se2 samples varies between 0.03 and 0.15at.% on a length scale of tens of micrometers.

No abstract