Monolithic silica aerogel (aerogel) is a highly porous material with pore diameters in the range of 10 — 100 nm. The porosity is above 90%, which combined with the nanometre pore size makes the aerogel a highly insulating material with a thermal conductivity lower than of still air [1]. Further decrease in thermal conductivity can be achieved if evacuated to a rough vacuum, i. e. below approximately 50 hPa in which case the thermal conductivity in the pore gas is almost eliminated [2].

Beside the low thermal conductivity a high solar energy and daylight transmittance is achieved, which makes aerogel a very interesting material for use in highly energy efficient windows [3]. The compression strength of aerogel is sufficient to take up the atmospheric pressure if evacuated but the tensile strength is very low, which makes the material fragile, i. e. if in contact with liquid water the surface tension in the pores would demolish the aerogel structure. So, the application of aerogel for window glazing requires the aerogel to be protected against water and tensile stress. This can be achieved by placing the aerogel between two layers of glass and apply a gas and vapour tight rim seal. When evacuated to a rough vacuum only compression stresses will be present in the aerogel due to the external atmospheric pressure.

Figure 1 shows the advantage of aerogel windows relative to commercial available low energy glazing for which the reduction in U-value is achieved by multiple layers of glass and low emissive coatings — measures that all reduces the solar energy and daylight transmittance. But aerogel glazing has a solar energy transmittance equal to plain double glazing and at the same time has a heat loss coefficient equal to the best triple layered gas filled glazing units. Monolithic silica aerogel is the only known material that has this excellent combination of high solar and light transmittance and low thermal conductivity — material parameters that makes it possible to achieve a net energy gain during the heating season for north facing windows in a northern European climate as the Danish climate.

The utilization of the passive solar energy passing through the windows is an important factor in reducing the annual energy consumption for space heating in northern European countries and has been the background for the research and development projects HILIT

[4] and HILIT+ [5] financed in part by the European Commission. The objectives for both projects were to improve the aerogel elaboration process with respect to material properties (both thermal and optical) and process parameters (drying, duration, safety, etc.) and to develop final aerogel glazing prototypes with a total U-value lower than 0.6 W/m2K and a total solar energy transmittance above 75%.

For the optimization of silicon cell manufacturing processes with respect to a reduction of breakage losses an i ntegral concept was developed and applied in industrial manufactu­ring lines. The concept consists on a combination of different approaches.

Measurements of the cell strength after specific manufacturing processes reveal, whether these processes induce or reduce damaging. The damage and crack formation and evolution history is investigated by fractographic and microscopic inspections of original cells broken during manufacturing. Fracture relevant processes are analysed by experi­ments and by numerical modelling in order to determine the acting mechanical or thermo­mechanical loadings. Numerical simulations by means of finite element analyses help to optimise the processes, because the influencing process parameters can be varied in the computer. Testing methods such as acoustic techniques or proof-tests are applied for the early detection and elimination of cells which are not able to survive further processing.

By a combination of the results of these approaches the causes of breakage losses can be localised, analysed and deleted.

[1] Beinert, J., Kordisch, H., Kleer G.: Studies on Fracture and Strength of Photovoltaic Silicon Wafers, World Renewable Energy Congress VII (WREC 2002), 2002, Cologne, Germany, Pergamon, 2002

The described system differs from existing installations using hydrogen because it employs components that are not yet widely applied. Additionally, it should be sufficiently safe and fault forgiving in order to be operated by "normal people”.

Due to its physical properties hydrogen differs from other gases. The main relevant points are the low density, that it is tasteless, odourless and invisible, its wide explosion range and that is tends to detonate.

The safety analysis for hydrogen systems [6] suggests that it can be handled without an unacceptable risk. However, the design of the installation has to take into account the peculiarities of hydrogen. Additional requirements stem from the fault forgiving properties mentioned above.

1 Results

As a result of this research project, a procedure for identifying configurations of autonomous energy systems which

• are safe to handle,

• sufficiently reliable,

• optimized as to the number and type of equipments required,

• energetically efficient.

will be available. Additionally, the relationship between the degree of redundancy, component lifetimes and their repair times can be analyzed.

Since the research is ongoing, calculation results will be communicated in the oral presentation at the conference.

An effective way to protect a surface from a hot fluid stream is to inject a cooler fluid under the boundary layer, forming a protective and cooling layer or film along the surface [7]. This concept could be applied to the cooling of PV facades, particularly at the top of the facade, where the air temperature behind the panels can be 70°C and higher. In cooling problems such as that of gas turbine blades, an array of small holes is often drilled through the surface of the blade material. The holes are drilled normal to the surface or sometimes more effectively at different angles. The objective is to cool the surface downstream of the holes by injecting the blanket of cool fluid into the boundary layer, along with improving heat transfer at the hole location due to increased heat transfer contact area. Film cooling usually involves the use of one or two rows of holes followed by an impenetrable area in the material surface, similar to the situation in the PV problem. Full coverage film cooling refers to the situation in which the cooling target is the area between the cooling holes, and generally approaches the effects seen in transpiration cooling. One solution envisaged in this project is with cooling holes around the PV cell and is a combination of full coverage and regular film cooling, with both transpiration and film cooling modeling theory being relevant. The natural buoyancy effects in a vertical fagade will induce convective flow through the cooling duct at the rear of the panels, and cause the suction of the cooling film onto their back surface.

Film cooling has primarily been used to reduce heat transfer from a hot gas stream to an exposed wall. The introduction of the secondary fluid into the boundary layer may be considered to produce an insulating layer, but can also be considered as a heat sink that reduces the temperature in the downstream boundary layer and thus the temperature of the wall. The secondary fluid usually serves both functions. For the PV fagade it is envisaged that both of the effects of film injection on the boundary layer will be desired at different points along the height of the fagade. At the lower ends, as the ambient air is drawn through the perforations into the duct, the mixing in the boundary layer will allow it to act as a heat sink and increased heat transfer will result.

Further up the fagade, as the temperature of the air in the duct will have increased, a more protective role will be required from the film of air injected into the boundary layer [8].

As outlined in Goldstein’s Review of Film Cooling, 1971 [8] and subsequent papers, the hole geometries of secondary injection as well as the density and various other parameters can be manipulated to achieve the desired cooling effectiveness. The main parameters to influence the role and effect of the injected air are: Hole geometry; Injection angle of holes; Number of holes per row; Number of film cooling rows; Position of film cooling rows, and Pressure loss or blowing parameter.

The use of cooling slots as opposed to discrete holes was covered by Goldstein in his film cooling reviews, and researched by Golneshan and Hollands [9] in their work on transpired solar collectors. Both research groups found that the cooling slots could provide both increased film cooling effectiveness and increased heat transfer,

depending, again on various parameters, with the blowing parameter being the most influential.

A spread of injection angles of 15°, 30°, 45°, 60°, 90° will be investigated for the set hole diameter and pitch. As it is anticipated that the most effective streamwise angle for simple injection will be found to be around 30°, this streamwise angle will be used for the compound angle injections. Even a slight change of orientation angle will increase effectiveness, with his tests incorporating angles of 0°, 30°, 60°. Lateral orientation angles of 60° and 30°, with a low and high simple injection angle will be investigated for this study.

The partially description of the folded planar device with geometric optics is only valid for structure dimensions up from several tenths of microns [3]. The optical modeling of the system can be divided into two parts. As the thickness of thin film systems of sev­eral hundred nanometers is much smaller than the coherence length of the incident light, wave optics has to be considered. The calculations of the absorbtance in distinct lay­ers were performed by near field simulations [4], [5],[6]. Optical constants are taken for the MDMO-PPV/PCBM system [7] which is comparable to the P3HT/PCBM system with a slight shift in complex refractive index towards shorter wavelengths. The thickness of the photoactive layer and the PEDOT layer was 100nm. Both polarisations of the incident light — transversal electric (TE) and transversal magnetic (TM) have to be considered.

The absorptance in the distinct layers and the fraction of reflected light is calculated.

The propagation of the reflected light inside the prismstructure is described by geomet­ric optics. Calculations are presented for the simple setup of normal incidence on the substrate leading to a twofold reflection on the thin film system under 45°. The spec­tral absorptance for TE- and TM-polarisation is shown in comparison to the case of nor­mal incidence on a planar thin film system (figure 3). From convolution of the spectral absorption with the AM1.5 solar spectrum, an increase in absorptance of 55 % in TE — polarisation and of 33 % in TM-polarisation was calculated. This leads to an average in­crease of 44 % for unpolarised light. Optical losses due to shadowing by the microgrid are not taken into account.

Table 1 and Fig. 3 summarize the results of the tensile tests. For PET elastic modulus values of about 4000MPa were obtained. In contrast, the fluoropolymers PVF and PVDF exhibit significant lower elastic modulus values, ranging from 1500 to about 2000MPa. The lowest elastic modulus of about 600MPa was determined for the EVA-PVF-PET-PVF laminate, the elastic modulus is dominated by the EVA elastomer. The plastic modulus values are ranging from 10 to 76MPa. PVDF showed a negative slope of the stress-strain curve after the yield point; the plastic modulus E2 was set zero.

Fig. 4 depicts the force versus displacement plots of the three thermoplastic laminates PVDF/PET, PVDF/SiOx-PET and PVF/PET-PVF. The curve of the PVDF/PET laminate without the silicon oxide barrier layer shows two levels of peel force. The high level is interpreted as a cohesive fracture in the peel arm and the low level can be attributed to adhesive fracture between the two peel arms (Moore and Williams, 2001). The peel force is independent of the configuration; the mean value is 9.1±0.3N. As to the peel angle ф of the PVDF/PET laminate significant differences between the two configurations were obtained. Configuration A (PVDF peel arm top) yielded an angle of 13.3±0.4°, whereas in configuration B (PET peel arm top) an angle of 11.9±0.6° was measured. This difference can be explained by gravity effects in combination with the low flexural stiffness of the laminate. Fig. 5 shows the curvature of the PVDF/PET laminate.

For the PVDF/SiOx-PET laminate with an 80 nm thick silicon oxide barrier layer no effect of the configuration on the peel angle was observed. For both configurations mean values of 12.7±0.4° were obtained. In contrast to the PVDF/PET laminate the peel force is significant higher (11.1±0.3N).

The PVF/PET-PVF laminate did not show stable peeling, the PVF peel arm broke before peeling could be established at a maximum force of 8.6±1.4N. For the PVF/PET-PVF laminate it was not possible to do the adhesive fracture toughness evaluation.

Due to the differences in the peel angle ф the adhesive fracture energy (GA) values of the PVDF/PET laminate without silicon oxide barrier layer scattered significantly. GA values of 192±23J/m2 were obtained. Doing the evaluation for each configuration A and B, Ga values of 210±12J/m2 and 175±17J/m2 were calculated for configuration A (PVDF peel arm top) and B (PET peel arm top), respectively. For the PVDF/SiOx-PET laminate (with the silicon oxide barrier layer) an enhanced GA value of 284±20J/m2 was determined. In comparison to common thermoplastic laminates which are used for packaging purposes the tested laminates showed equal or more than 30% higher GA values (Moore and Williams, 2001).

The investigations on the two PVDF/PET laminates with and without SiOx barrier layer indicated that the energy based fracture mechanics approach is much more

sensitive to describe the delamination behaviour compared to the usually applied load based analysis.

Fig. 6 indicates that the adhesion between PET and EVA is much higher than between PET and the other materials (PVF, PVDF). A peel force P of 142±6N, a peel angle ф of 77.3±0.8° and an adhesive fracture energy of 7895±454J/m2 were determined. The obtained GA value is significant higher than any in the literature reported value of GA for plastics. Hence, for the investigated laminates a delamination within the backsheet material may be more likely than delamination between backsheet material and EVA solar cells encapsulation material.

Acknowledgements

The research work of this paper was performed within project S.9 at the Polymer Competence Center Leoben GmbH (PCCL, Austria) within the framework of the Kplus — Program of the Austrian Ministry of Traffic, Innovation and Technology with the contributions by the University of Leoben, Graz University of Technology, Johannes Kepler University Linz, Joanneum Research ForschungsgmbH and Upper Austrian Research GmbH. The PCCL is funded by the Austrian Government and the State Governments of Styria and Upper Austria.

The incoming radiation at the site of the PV system has to be known to simulate the expected energy yield. This information is derived currently by measurements of the geostationary weather satellite Meteosat-7 (It will be changed to Meteosat-8 in the future.). These data have a high temporal resolution and a better spatial coverage than meteorological ground stations. The Meteosat images are received half-hourly operational at the University of Oldenburg. The Heliosat method (Hammer, 1999) is used to determine ground irradiance for a pixel with a resolution of 3 km x 5 km above the site of a PV system. Validations with ground stations showed that the error of this method can reach up to 30% for hourly values and lies between 5 and 10 % for monthly irradiance values (Hammer, 1999). These errors are strongly dependent on the weather situation (i. e. variability) and on the sun elevation angle. Under cloudy sky conditions the error increases while only small errors occur on days with clear skies. Larger errors also occur for low solar elevation angles and result in more deviations during wintertime. The quality assessment on these derived irradiance data and its dependencies is rather complex. Further details and explanations on the quality assessment will be published soon.

One further development to improve the irradiance calculation and to reduce the error is the introduction of a new model to calculate diffuse irradiance (Mueller et al, 2004).

Another way for improving the accuracy is the combination of the satellite data with ground measurements of irradiance with the geostatistical method kriging-of-the-differences. The error can be reduced this way to 3-4% for monthly values and also for hourly data (Betcke and Beyer, 2004).

These errors within the derived irradiance data have to be considered in the error detection routine. The exact knowledge of these errors and its dependencies is an absolutely necessary prerequisite for a variable definition of allowed differences between the expected and the real energy yield and therefore, for the exact determination of the occured malfunction.

Before conducting the experiment, the knowledge of the product/process under investigation is of prime importance for identifying the factors likely to influence the outcome. In order to compile a comprehensive list of factors, the input to the experiment is generally obtained from all the people involved in the project with expert knowledge. In this case the main source of such knowledge was Dr Daniel Nuh of EETS, who designed and built the laminator used for module production.

The typical lamination cycle for a fast cure Solar EVA is illustrated in figure 4. First the Solar laminate is assembled from its individual layers and placed face down on the hot plate. The lid of the laminator is closed and the plate is held at a constant temperature. Under vacuum, the EVA is now subjected to a set time, TEVA. At point A, atmospheric air is allowed into the top of the vacuum divider sheet, which applies a pressure on the laminate pressing it down onto the hot plate. Here the laminate is left for a further set time, TPC. Finally the vacuum is switched off and the laminate removed for cooling.

The different variables or factors associated with the Lamination process are:

* Temperature, 0C — Operating Temperature

* Pressure, Torr — Vacuum pressure

* TEVA — Solar EVA Molten Time, mins.

* TPC — Post Curing Time, mins.

The pressure is a fixed factor throughout the cycle therefore it is excluded from the Design of Experiments. Therefore the Factors to be studied are:

* Temperature, 0C — Operating Temperature

* TEVA — Solar EVA Molten Time, mins.

*

TPC — Post Curing Time, mins.

In comparing mature and immature design of sustainable products, appearance will evidently play a role as is their user friendliness, their usability and last but not least the way of PV modules implementation. In practice it turns out there will be a compromise between the appearance, its functionality, usability and utility. Where to draw the precise line will be a matter that can change at each case.

Up to now in PV powered products the main concern has just been the PV performance. However there is also a growing preference for additional comfort, convenience of use in other words improved user friendliness, or good ergonomically designed products and appealing designs are in demand. Depending on the user context as a result, this will have a direct impact on the shape and colour of the PV cells to be used.

In sustainable product design, the actual colour of the PV cells will become a less important factor since a dominant colour will on the long term limit the time a product is in vogue. This is for example demonstrated in the new PV roof projects in which the PV modules on the roof are built without being conspicuous, just resembling normal tiles. The main restriction is that the overall performance of the PV cells is not frustrated by its colour (see section 3.3)

1.1.3 Convenience of use, and appearance

Convenience of use is demonstrated for instance for a product which has to fit nicely into the palm of your hand and therefore should not have sharp edges. In figure 3a [Solar — Home,2004] it can be seen that the rectangular and flat shaped PV module on the PV torch hampers the proper use mainly due to its sharp edges. In addition it is not aesthetically designed. This design ‘craves’ for PV panels with curved surfaces, which follow the cylindrical shape of the battery holder as can be seen in Figure 3b.

1.1.4 PV implementation and appearance

Figure 4a: PV powered lamp

Comparing the PV powered lamp presented in Figure 4a with another lamp presented in Figure 4b namely a PV (streetlight) lamp designed in the framework of a master thesis project [Verkuijl 2000] one can observe the following differences:

In the first case, the PV module is just an add-on unit not well integrated into the overall design. In the second case the curved PV panels will function as Solar Cell on one side and simultaneously as light reflector of the generated light on the other side. A sound integration of the PV panel into the overall design is demonstrated. As a result of this integrated design, synergy between the energy conversion function and the energy use (light) is obtained. The energy storage was done at lamp location, introducing a mobile design concept. In addition the use of curved PV panels enhanced the aesthetics since this street lamp resembles a flying object. The latter design is more mature than the former one.

In both torch and lamp examples above the need of curved PV panels is evident. However one should be cautious to generalize such need to all PV powered products. A design rule could be that curved PV cells are to be used if and only if this feature introduces a new added value.

1.1 Evaluation strategy

Any attempt to measure slight differences in efficiencies (in the percent range) on complete modules is likely to be successful only if a large number of modules are monitored over long period of times. This is because it is far from trivial to prepare modules with similar initial efficiency (current mismatch between the cells, serial connections) and because it is not easy to measure complete modules before and after lamination. Ab-initio simulations of the optical properties of complete systems are also highly difficult, since several unknown parameters would have to be taken into account (surface structure of the cells, internal reflection at the backside of the cells, internal reflection inside the glass). Therefore we adopted the following experimental strategy to be able to predict as accurately as possible the efficiency gain in STC conditions (which will give the nominal gain in Wp of the module) and the total yearly energy yield (which is the quantity of interest for the operator of a PV system):

Selection of 24 standard multi-crystalline silicon solar cells with SiN antireflection layer, front screen-printed Ag and back Al contacts. Prior to measuring, all cells are tabbed with tinned-copper strings (normally used for cells interconnection)

• Measurement of the efficiency and of the short-circuit current Isc under STC (AM1.5g, 25°C), measurement of the spectral response and of the angular response of the cells. On this base, two similar groups of 11 cells are prepared. 2 cells are kept for calibration purpose.

• Individual encapsulation of each solar cell in mini-modules, in a standard Glass/EVA/Cell/EVA/Tedlar structure. The EVA foils are 1 mm thick and a curing temperature of 157°C was applied for 11 minutes. 11 Cells are prepared with the AR glass, 11 without.

• Repetition of the measurements with the mini-modules

• Preparation and outdoor monitoring of 2 modules (18 W each) with and without AR layer to evaluate temperature effects given by the aR layer

• Simulation of the expected yearly energy yields using a specially developed software (Real Reporting Conditions Rating [10]) using a meteorological database and taking into accounts all the parameters measured on the mini-modules.