QMT Features: July 2011
Lighting the way with metrology
Using high resolution confocal light microscopy,  Dr Eckard Wefringhaus, a scientist at the International Solar Energy Research Centre in Konstanz, Germany, is investigating how alterations in the surface of solar panels may improve performance.

The energy consumption requirements of the human race are expected to increase around 50% by the year 2035, while human energy use is a major cause of global warming. For these reasons, a shift towards renewable energy sources and away from rapidly dwindling fossil fuels is long overdue.

In order to make renewable energy the most attractive economic proposition for energy companies, governments and individuals, significant effort is being made to increase the efficiency of such systems. One approach is to use solar cells, which convert light energy irradiated onto the planet’s surface by the sun into electrical energy that can be stored and utilised when required. One of the main challenges facing this technology is to maximise the efficiency of light capture; at present up to around 22% of the energy striking the cell is converted to electrical energy, with the rest either reflected or lost as heat.

Dr Eckard Wefringhaus, a scientist at the International Solar Energy Research Centre in Konstanz, Germany, is investigating how alterations in the surface of solar panels may improve performance. The surface of a solar cell is analysed in high resolution using a confocal microscope, and potentially beneficial surface modifications are characterised in great detail, facilitating the optimisation of solar cell efficiency.

Solar challenges
Solar cells convert light energy into electricity via the photovoltaic effect, which occurs when electronic semiconductor materials are struck by photons, causing them to release electrons that can be harnessed to generate an electric circuit. Cells are often composed of mono-crystalline or multi-crystalline silicon, cut into wafers approximately 200 µm thick, the former of which is more expensive but can be used to make more effective solar cells.

This is due to the fact that, unlike multi-crystalline silicon, mono-crystalline silicon is amenable to surface modification using alkaline etching techniques. The etching step creates a complex texture on the surface of the wafer, composed of thousands of small pyramidal structures. The pyramids help to maximise photon absorbance by changing the angle at which incoming light waves are reflected, ensuring that more light energy is captured and less is reflected away. Optimising the etching pattern on the surface of the cells has the potential to maximise the capture of light energy, thereby increasing the efficiency of the system.

During solar cell production, the structure and surface of the cells must be carefully analysed to ensure that strict manufacturing standards are being adhered to. This involves verifying accuracy at each stage of production, including checking the surface of the silicon for mistakes and imperfections. Without such a high level of quality control, each cell would perform significantly below optimal efficiency. In addition, research scientists are using similar methodological approaches to production engineers, although in this case their goal is to optimise the design, function and efficiency of each component. Of particular interest is the identification of new ways to improve light capture via modifications in the surface of the silicon wafer. This frequently involves cantilevers or microscopes to analyse the surface topology of the cell surface, so that the modifications or texture patterns that maximise solar cell efficiency can be identified.

Everything’s ok on the surface
Dr Eckard Wefringhaus and his research group at the International Solar Energy Research Centre are currently working on ways to optimise the surface etching of solar panels to provide best performance. Amongst other things, the study aims to find a way of accurately describing the pyramidal pattern etched into the silicon. By precisely defining variables such as pyramid distribution, height and number it is much easier to carefully define the surface configurations that provide optimal performance. This information can then be utilised in a mass production setting to ensure that all solar cells will operate at maximum efficiency.

Several approaches have been taken to characterise the surface of a solar cell. A form of cantilever, known as a profilometer, can be dragged across the exterior of the silicon, with small movements in the height of the device recording the crevices and protrusions present. This method is only as accurate as the diameter of the probe tip and is time-consuming as it can only work in two dimensions. In order to measure a square panel, the profilometer must ‘scan’ the wafer a large number of times, across a slightly different trajectory each time. Perhaps the greatest weakness of this method is the possibility of causing damage to the surface of the solar cell. This leads to confusion between real measurements and artefacts generated by the analytical process, and could render the solar cell unusable.

In an effort to obtain more data at a higher resolution, researchers have traditionally turned to scanning electron microscopy. The technique involves firing electrons at a sample housed in a vacuum-sealed specimen chamber. The electrons are reflected in different directions depending on the topology of the sample, allowing a highly detailed 3D representation of the surface to be constructed from the data collected. Although this approach provides a wealth of data, it can be slow to perform. Therefore, to speed up analysis, researchers often just image a few representative areas of the sample. This makes the application more amenable for use in the laboratory and on the production line, but misses much of the variation present on even a single silicon wafer.

Illuminating the way forward
To provide a faster means of generating a 3D topology map of a solar cell, Dr Wefringhaus and his colleagues have begun to harness the advantages provided by light microscopy. Using a laser confocal microscope, such as the Olympus LEXT OLS4000 (Figure 1), it is possible to scan in fine detail the surface of any sample, without the need for complex or time-consuming sample preparation, specialised stage conditions and without damaging the material being analysed. Confocal microscopy uses a highly focused laser beam to illuminate a sample at a specific position, while a pinhole in front of the signal detector eliminates out-of-focus light bouncing back off the sample thereby creating a crystal clear image (Figure 2). In combination, these features provide a high degree of spatial resolution, such that a sample area of the specimen can be scanned at multiple height positions and the data used to create a 3D representation. Scanning of each area can be completed within minutes and is only required once; the surface can then be analysed in great detail using software-based image analysis. The system can be automated using a programmable sample holder which, in combination with the autofocus capabilities of the LEXT OLS4000, allows multiple areas of a single sample to be scanned in a time-efficient manner with minimal user input. Up to 12 images are routinely captured, which is enough to reliably represent the whole surface of the wafer.

Taking in the pyramids
Dr Wefringhaus and his team are currently using the LEXT OLS4000 to investigate the relationship between the pyramid pattern found on the surface of a solar cell after the alkaline etching process, and the subsequent changes in efficiency provided by these modifications. However, before this can be achieved, it is necessary to establish a set of reliable and intuitive measurement standards with which to classify the pyramid pattern.

To do this, several silicon wafers were scanned and the accompanying Olympus STREAM analysis software was used to characterise each in terms of height variation. This data was then used to infer pyramid number and position using the MountainsMap software package produced by Digital Surf, providing a reduced, more manageable data set that only describes the co-ordinate positions of the pyramids (Figure 3). As can be seen in Figure 4, the heights of the individual pyramids followed a Burr distribution, providing an example of how this method can be used to generate quantitative, descriptive data. In the future, the group hopes to continue this research by developing statistical methods to describe the spatial distribution of pyramids (e.g. by dispersion indices).

These combined analysis parameters can then be applied to wafers produced using different etching conditions, allowing the researchers to study the relationships between pyramid pattern, surface roughness, light reflection and solar cell performance.

Global warming and the depletion of the global stock of fossil fuels are two major challenges facing the world’s population that can be met by investing in renewable energy sources. One such approach is to use conductive silicon solar panels to capture and store energy from the world’s most abundant energy source – the sun. In order to become the most effective renewable energy solution available, manufacturing processes that maximise solar cell efficiency will be required, as will the development of standards that are robust and simple enough to maintain in a production line environment.

Much of the efficiency of a solar cell unit can be attributed to the surface characteristics of the silicon wafer, as this is the element responsible for trapping the sun’s energy. Therefore, the development of rapid, accurate and reliable methods to analyse wafer topology are of upmost importance.  Although traditionally this has relied upon the use of cantilevers or electron microscopes, confocal light microscopy is rapidly becoming a useful tool to achieve this, as the approach can provide accurate data quickly and easily.

Confocal microscopes may soon become the metrology tool of choice in production factories and research institutes alike, due to their ease-of-use, flexibility and ability to produce accurate data without directly affecting the sample being investigated. l
Dr Eckard Wefringhaus email:
You can now view all QMT Magazine issues on your favourite tablet or smart phone.
Download the free Quality Manufacturing Today App from the Apple iTunes App Store or from QMT Magazine on Google Play.

Rob Tremain Photographer
Click above to see full page display and links to QMT articles.
Hexagon Logo 2
Mitutoyo logo
Control logo