Solar so good

Most of the solar radiation energy absorbed by solar cells is converted into thermal energy. This may cause the cells' junction temperature to rise unless the heat is efficiently dissipated to the environment. Heat removal is therefore an important issue in solar cell operation. Encapsulation of solar cells affects the heat flow into and out of a PV module and thereby increases its temperature. This effect reduces the PV voltage and thus lowers the PV efficiency. In addition, this temperature increase can cause various malfunctions and degradation of the PV module. Elevated temperatures increase stresses associated with thermal expansion.

[caption id="attachment_26177" align="alignright" width="300" caption="Figure 3: DEA test result of the curing of EVA at 150°C at 1Hz"][/caption]

Investigation of the thermal conductivity and thermal diffusivity is important in overcoming thermal management problems during the PV module operation. The Laser Flash (LFA) technique is a fast and reliable absolute method for determining these thermophysical properties, including the specific heat. This data can then be used for:

• Prediction of the heat transfer and temperature profile as a starting point for the description of the processing behaviour of multi-sheet systems
• Thermal diffusivity and thermal conductivity as input data for numerical simulation.

Silicon-based technologies have always played a significant role within the energy industry (today’s market share >90%).  The drive in the solar industry is to improve the efficiency of the PV modules produced. Since higher efficiency is a direct function of increased thermal conductivity, it is important to be able to determine these values.

Figure 1 shows the measurement of the thermophysical properties of a silicon wafer. In the temperature range from -100°C to 500°C, the thermal conductivity and thermal diffusivity (LFA measurement) continuously decrease. The determination of the specific heat was carried out with a DSC system. The standard deviation of the data points is <1%.

The most promising cell types are thin-film solar cells based on direct semiconductors with high absorption coefficients. In fact, the highest laboratory efficiencies are reported for thin-film solar cells based on the chalcopyrite semiconductor system Cu (In,Ga) (S,Se). However, they do also have some disadvantage which lies in their physical-chemical properties and their stoichiometry. This makes crystal growth and/or the manufacturing of epitaxial layers relatively complicated. Here, TG-DSC coupled to a mass spectrometer (MS) can provide important information for studies of physical-chemical properties such as: Melting point, evaporation, characterisation of gas species during heating, determination of impurities and temperature of incongruent melting.

[caption id="attachment_26178" align="alignleft" width="300" caption="Figure 4: First and second heating of a commercially available EVA film (Mitsui Chemicals Fabro)"][/caption]

Figure 2 shows the measurement of CuGaSe₂ (band gap 1.68eV) that can act as the top cell in a PV tandem device with CuInSe₂ as the bottom cell. CuGaSe₂ was synthesised from Cu, Ga, and Se taken in stoichiometric amounts. At 450°C, the evaporation of Se₃ is detected by means of the isotope distribution between m/z 230 and m/z 245 what indicates a non-stoichiometric material. The presence of iodine shows that it was used as a mineraliser for synthesis. The presence of Se at temperatures higher than 900°C is due to the thermal degradation of CuGaSe₂. Regulation of the Se vapour pressure is required to control the stoichiometry.

One of the most critical steps in PV production is the encapsulation of solar cells. This encapsulation should allow for efficient power generation in any application while protecting the cells from damage or corrosion caused by mechanical shock or the environment (moisture, dust, etc.). Thermal treatment of the polymer resin during the encapsulation process is performed in the laminator in two steps, the lamination and the curing. This treatment effectuates a bonding of the multiple layers of materials with thermo-sensitive polymer films. The most widely used encapsulant is EVA (ethylene vinyl acetate copolymer). As the polymerisation reaction is irreversible, the thermal treatment of the PV cell encapsulation is crucial. The quality and lifetime of the PV modules/arrays depend on the calibre of this production process.

In Figure 3, a DEA test of an EVA sample was carried out in a lab furnace which allowed programming of time and temperature ramps (up to 40 K/min) and the use of disposable comb sensors. The multi-frequency measurement (frequencies between 1Hz and 10000Hz) was carried out, and the ion viscosity (W/cm) was monitored. Presented here is the behaviour of the ion viscosity at 1Hz.

DSC is well suited for performing quality control analyses on commercially available EVA films with respect to their glass transition temperature, melting behaviour and degree of curing. Competitive materials can easily be compared, and the cross-linking process can be investigated as a function of temperature (dynamic tests) or time (isothermal tests).

Figure 4 demonstrates the DSC measurements* on an EVA film sample (approx. 7mg) at heating rates of 10K/min. In the first heating (blue curve), the glass transition temperature at -28°C is followed by an endothermic double peak between 50°C and 100°C. This melting behaviour can be correlated to a lamellar thickness distribution. The exothermic peak at 158°C indicates the exothermic crosslinking reaction. Noticeable is the rather low reaction enthalpy (-14.15 J/g) in comparison to epoxy resins (typically between -400J/g and -500J/g). In the second heating (red curve), the glass transition occurred at almost the same temperature. The endothermic double peak between 40°C and 80°C has changed to a broad shoulder with its maximum at 63°C. The higher the crystal thickness, the higher the melting temperature; therefore, the change from a peak into a broad shoulder is an indication of a distribution of crystals with reduced thickness as a consequence of the thermal treatment in the first run. No exothermic reaction peak is present in the second run, indicating that the crosslinking process was finished after the first heating.

Characterisation, quality control and research and development of solar cells can be well done with the different Thermal Analysis and thermophysical properties testing methods. The tests results can be directly used for example to improve the production process (e.g. kinetic analysis of the crosslinking reaction), optimise crystal growth (e.g. stoichiometry control), investigate curing behaviour of encapsulants, and to study heat transfer mechanism (e.g. determination of the thermal conductivity). The few application examples shown here, already demonstrate that many of the thermal management tasks can be investigated with the help of Thermal Analysis. The clear advantage lies in the fact that these methods are standardised, less complicated in operation and sensitive enough to fulfil the demands of state-of-the-art research and development tasks.

The Author

Peter Davies of NETZSCH-Gerätebau GmbH

*Acknowledgement Our thanks to Dr. W. Stark and M. Jaunich from the Federal Institute for Materials Research and Testing ("BAM") in Berlin for the measurements DSC and discussion. The results and detailed discussion are published in Polymer Testing 30 (2011) 236-242.

•