F. Kraemer

Constitutive behaviour of copper ribbons used in solar cell assembly processes

One of the driving factors for a steady reduction in wafer and cell thickness is the present shortage of polysilicon feedstock combined with the need to reduce manufacturing costs in photovoltaic module production. Therefore materials and manufacturing processes must adapt to maintain acceptable mechanical yields and module reliability. The soldering of solar cell strings is a critical step in the production of photovoltaic modules. Mechanical load and temperature induced stresses can cause cracking in the cells. During the soldering operation, the cell and the wires heat up and expand and then later contract when the heat is removed below the melting point of the solder. The differential contraction between the Cu and the Si combined with thermal gradients, cause stress to build up in the system. Since the solder thickness (5 … 20 µm) is relatively small compared to thickness of the copper ribbon (100 … 200 µm) and the thickness of the silicon solar cell (160 … 200 µm), the constitutive behaviour of the copper ribbons is one of the key factors to reduce breakage after soldering of solar cells into strings.

Mechanical behaviour and fatigue of copper ribbons used as solar cell interconnectors

The soldering of solar cell strings is a critical step in the production of photovoltaic modules. Temperature induced stresses can cause cracking in the cells. During the soldering operation, the cell and the wires are heat up and expand. During the subsequent cooling phase they contract. The differential contraction between the Cu and the Si combined with thermal gradients, causes mechanical stress in the assembly. Moreover the lifetime of solar modules can be limited by the fatigue of the copper-ribbons. Since the front glass and the silicon cells have a significant difference in their coefficients of thermal expansion, temperature fluctuations are able to induce thermo-mechanical stresses in the photovoltaic module. The paper will present the results of the mechanical tests on copper materials. In order to give an explanation for the fatigue behaviour of the material, a correlation to the relevant microstructures will be given. Based on the investigated mechanical behaviour practical implications for proper handling of copper ribbons during solar module production processes will be concluded.

Interconnection technologies for photovoltaic modules - analysis of technological and mechanical problems

The paper describes the problems of interconnecting single solar cells with each other to create a photovoltaic module. High power und low voltages demand the transport of high currents through the interconnection wires. The resistance of the wiring is crucial, because it significantly influences the total module efficiency. However, increasing the width and height of the rectangular wires leads to significant mechanical problems, because the brittle silicon solar cells are prone to breakage. Therefore innovations in interconnection technologies are needed to release mechanical stresses during the assembly process.

BGA lifetime prediction in JEDEC drop tests accounting for copper trace routing effects

Experimental drop test results of 2nd-level assemblies can be influenced by numerous impact factors. The explicit definition of drop testing conditions by the JEDEC standard JESD22-B111 was intended to create a highly repeatable, and thus comparable, experimental setup. Recent developments showed, however, shifting failure modes from component to PCB side.

Mechanical problems of novel back contact solar modules

In this paper, the mechanical integrity of interconnect structures of different types of photovoltaic modules is analyzed by FEM-simulations. The goal is to develop a theoretical framework for the systematic understanding of mechanical behavior of photovoltaic modules. The results of this investigation figure out, where are critical locations in the photovoltaic module assembly that limit the module lifetime. Another goal is to show, how changes in the assembly technology (e.g. geometries, processes, materials) will have an effect on the overall reliability of the photovoltaic module. FEM-Simulations were carried out, in order to analyze the mechanical stresses, which were built up during the operation of the photovoltaic modules. Thermo cycling test between -40°C and +80°C were simulated. These tests are used in the qualification procedure for photovoltaic modules according to TEC/EN 61215. The results of the study show that stresses during the assembly and the operation occur in different locations in dependence of the assembly design. Tn the traditional modules the silicon underneath the busbar and the copper ribbon crimp between two adjunct cells are the locations where most of the damage is likely to accumulate. In contrast to this the solder joints in the outermost row of the back contact cells are the critical locations in the novel back contact design. The paper will explain the differences of thermo-mechanical behavior between the designs.

The effect of copper trace routing on the drop test reliability of BGA modules

The Jedec drop test has become a popular method for the assessment of the dynamic mechanical reliability of 2nd level assemblies. It delivers repeatable results and is thus well suited for the development of a virtual lifetime model based on FEM simulations. Detailed experimental studies showed PCB copper trace fractures as the dominating failure mode. The virtual risk assessment applied a two steps approach (sub-modeling technique). The overall PCB motion was computed by a global model of the entire Jedec board. In the second step, the copper trace load was investigated with a single component. The resultant plastic strain showed a clear dependency on the relation between copper trace routing and dominating PCB deformation. A strain map was derived which indicates the strain level as a pre-factor to each trace routing condition. The validity of the strain map concept was proven by comparison with experimental results. The strain map was able to precisely identify the failing interconnections at each component position. The combination of interconnection strain energy from the global model and the strain map was able to match the experimental sequence of failures across the Jedec board exactly. A lifetime model was derived which is able to predict the cycles-to-failure of three different package types with less than 25% deviation to the tests. Hence, this lifetime model sets the ground for virtual prototyping that also includes the BGA drop test endurance.

A detailed investigation of the failure formation of copper trace cracks during drop tests

The development cycle of new products can be dramatically reduced if exact lifetime models are at hand. This requires the precise knowledge of the failure modes and the failure position under all test and service conditions. In case of dynamic mechanical loads like drops of BGA modules, broken copper traces at the PCB side are more and more often observed to be the ultimate failure effect. However, straightforward FEM simulations have shown unrealistic high stress and strain results not matching experimental observations which prove that a realistic representation of this failure is not trivial.

Assessment of long term reliability of photovoltaic glass–glass modules vs. glass-back sheet modules subjected to temperature cycles by FE-analysis

Abstract Quantifying the reliability of photovoltaic (PV) modules is essential for consistent electrical performance and achieving long operational lifetimes. Optimisation of these parameters increases the profitability of photovoltaic electricity because such systems should only require an initial capital investment. There are several aspects in a PV module which compromise its profitability. One such important aspect is the thermo-mechanical stress that is induced by day to night temperature cycles during every day of operation. Since this stress obviously cannot be omitted the PV module set-up should reduce the resulting internal loads to a minimum. This paper analyses the effects of the thermally induced stresses in two different module constructions. The thermo-mechanical reliability of photovoltaic modules is tested by the IEC standard 61,215 which accelerates the day to night cycles. Detailed analysis of this experimental test method is done by FEM simulations. Results of those numerical analyses are able to directly analyse the internal stresses in a PV module. The investigation presented here applies a classic module assembly for H-patterned cells with a single front glass and a plastic back sheet which is the reference type. The second packaging type for H-patterned PV cells is the glass–glass module which replaces the back sheet by a second glass sheet. Both module types have the same base area including 60 solar cells and the same total thickness. Each of the module assemblies are transferred to 3-D FE-models and subjected to temperature cycles. The simulation results show no module deformation for the symmetrical glass–glass module while the glass-back sheet assembly deforms by several mm. The mechanical results show that the solar cells are displaced towards each other when temperatures decline and vice versa during temperature increase. This forced movement causes stresses and strains in the interconnection structures of the modules. The analyses reveal that inside the glass–glass module the copper ribbons and solder layers are subjected to higher mechanical loads compared to the reference type. In case of the glass–glass module the copper ribbons may fail which can result in a complete cut of the series-connected solar cell strings.

Modelling of the mechanical behaviour of copper in 2 nd level interconnection structures

The paper presents an approach to model the mechanical behaviour of copper in 2nd level interconnect structures in electronic assemblies. The discussed mechanical models were analysed in ANSYS and LS-DYNA FEM-Software in order to simulate the performance of typical structural elements in electronic assemblies, such as copper traces or solder pads. Loading conditions span a wide range from low rate deformation of thermal cycles to high rate deformation during drop testing. This study investigated the effect of different mechanical properties of copper, with respect to the stress-strain relationship in 2nd level interconnects. The effect of the anisotropy of Youngs modulus, in addition to the effect of isotropic cyclic hardening on the resulting deformation and stresses in the copper structures, were analysed. Furthermore, the resulting contact forces at the copper pad to the solder and to the PCB epoxy material, were investigated. This paper presents specific observations made during the three-dimensional finite element simulations of typical interconnect structures. Microstructural investigations were also carried out, such as to be able to correlate particular mechanical behaviour with established knowledge about copper as an FCC material. Grain sizes and texture of real copper traces are estimated. This study relates these particular features of real structures in electronic assemblies to published properties of copper mono- and polycrystalline materials. The importance of microstructural properties, such as grain size and orientation in terms of their respective influence on the results, are also discussed.

High strain rate behaviour of lead-free solders depending on alloy composition and thermal aging

This work focuses on the mechanical behaviour of lead free solder alloys under high strain rate and thermal aging conditions. Tensile experiments have been accomplished using miniature specimens with a diameter of 1 mm to stay close to the dimensions of real solder joints. The specimens have been manufactured by a casting process. A controlled cool down process was applied having a cooling rate of 50 K/min. This way the solder specimens solidified in a comparable manner to reflowed solder joints. Two lead-free solder alloys SnAg1.3Cu0.5 (SAC) and SnCu0.7 (SC) were used for specimen manufacturing. The specimen preparation also contained a anneal step and strain gauge assembly. The experiments were conducted using a high strain rate tensile tester providing a constant deformation speed up to 5 m/s. The stress and strain within the specimen was recorded with high resolution using strain gauge and laser triangulation sensors respectively. Stress data were used to evaluate the yield stress dependency on strain rate and aging conditions. Furthermore, fracture site appearance and ultimate strain were analysed for a better understanding of the deformation behaviour. Cross sections were used to enable an insight into the deformation mechanism taking place depending on the applied strain rate and aging condition. Experiments were done at room temperature for specimens in the as cast state and after a considerable isothermal storage at 150 °C for 1000 h. The applied strain rates covered a range of 25 s-1 to 870 s-1. The strain rate depending material behaviour considering alloy composition and aging condition was incorporated in a finite element model and used to analyse the solder joint stress during a standard drop test experiment. SAC revealed a high sensitivity on the applied strain rate. Its yield stress increased with the strain rate. The yield stress of SC showed a lower sensitivity and a decrease towards higher strain rates. The ultimate total strain decreases with strain rate but shows quite high values and therefore pointing at an overall ductile behaviour for both solders. However, fracture site features of both solder alloys indicate a change of the deformation and fracture behaviour within the tested strain rate interval. Cross sectional analysis of SAC specimens also verify a change in the deformation behaviour. If the SC specimens are exposed to a prior isothermal storage the yield stress level decreases. The measured yield stress behaviour was fitted using the Cowper-Symonds-Equation for both solders and ageing conditions. Utilizing the as cast SAC strain rate sensitive material description for modelling a standard JEDEC drop test experiment the stress distribution shows clearly potential solder joint failure sites.