Confocal microscopy provides an optical method for measuring surface roughness, one that`s now much more accepted in industry as an alternative to tactile measurements, as Greg Blackman discovers
The surface roughness of a manufactured part can be a critical parameter in industrial production, especially if it’s a moving part like a car engine cylinder or a ball bearing – too rough and it will stick, too smooth and it won’t retain oil. In industry, tactile profilometry, whereby a stylus is dragged over the surface, used to be the method of choice for measuring surface roughness. Now, however, optical profilometry such as confocal microscopy has become more widely used in industry.
Miriam Schwentker, product manager at Olympus, says: ‘There is much more confidence in optical profilometry as a measurement tool.’
A confocal microscope uses a laser to scan the sample, with the detector only receiving light from one plane – as opposed to standard optical microscopes that focus on one plane but capture everything, even areas out of focus. By scanning with a laser in the z-direction, a confocal microscope will generate a 2D image while collecting height information of what was in focus. This makes a confocal microscope very much like a profilometer, says Schwentker: ‘The stylus-based instruments generate a height map by running a stylus over a sample’s surface, whereas a confocal microscope does that with a laser.’
Various confocal microscopes are available: Olympus supplies its LEXT OLS4000 system for optical profilometry and roughness measurements, while Spanish firm Sensofar offers its PLu Neox confocal microscope among other models.
Roger Artigas, R&D manager at Sensofar, cites semiconductor production as a big market for the PLu Neox, with applications including measuring surface features on semiconductor chips, and inspection of surface patterning on solar cells and micropatterns on LED glass substrate layers.
The advantages of confocal microscopy are firstly that the optical method is non-contact, so there’s no risk of damaging the part under inspection, and secondly that it’s much faster than a stylus-based approach. Schwentker says a confocal microscope can generate an image of the surface of a standard mechanical part in a few seconds, faster than tactile profilometry. The LEXT OLS4000 microscope also has a motorised stage, allowing it to scan several images in a row and digitally stitch them together for a panoramic view.
Imaging particularly large samples or those with a steep incline over the surface like a rock sample, for instance, can prove difficult with conventional confocal microscopes. German company WITec has developed its TrueSurface microscope for Raman imaging over large or irregular surfaces.
‘A simple optical microscope or a confocal microscope could only focus on a small section of an inclined surface, or a lot of sample preparation is required,’ says Harald Fischer, marketing director at WITec. The TrueSurface microscope can first acquire the surface topography of a sample and then collect Raman spectra across the surface area. The Raman image will give the chemical distribution of the compound across the surface.
To acquire surface topography, the device uses a confocal chromatic sensor with a lens array. The constituent wavelengths of white light are refracted to different extents to give different focal points, which in turn can be used to generate a height profile of the surface and acquire the topography. The instrument follows the surface profile and acquires a confocal Raman spectrum at each image pixel. At 200 x 200 pixels the device will capture 40,000 spectra, although larger sample sizes are possible. Chemical distribution of a compound can then be mapped across the sample surface.
Applications for this technique include those in geology. ‘Rough or inclined rock samples would be difficult to study with confocal microscopy, whereas they can be analysed very easily with this technique,’ says Fischer. Another application is determining the chemical composition of a pharmaceutical tablet, which would otherwise have to be flattened for confocal or Raman microscopy. WITec’s Raman imaging system can also image very small sample volumes going down to the diffraction limit, around 200nm.
Ease of use
There are different requirements for different industrial processes; in automated systems for electronics inspection, for example, manufacturers want microscopes with large fields of view and the ability to match the field of view to a production process, says Scott Elderkin, business development manager at Qioptiq. His company has developed Optem Fusion, a modular micro-imaging lens platform for a number of micro-imaging and automated microscopy applications including microelectronics processing and packaging inspection. It incorporates a range of lenses of different magnifications. ‘For electronics packaging applications, for example,’ says Elderkin, ‘Qioptiq’s Fusion system would allow a packaging machine to navigate a PCB in low magnification, as well as providing higher magnification for alignment and bonding imaging to enable the machine to place a package on the board.’
Automated electronics inspections require setting up correctly, but are then left to run. In terms of manual inspection, though, there is the additional cost of training staff to use the instrument effectively. ‘In industry, training staff to use microscopes in production can be very time-consuming,’ says Schwentker at Olympus. Olympus’ opto-digital microscopes, of which the LEXT OLS4000 confocal instrument was the first, are designed to be more simple to use. ‘We wanted to implement all our knowledge at Olympus into software tools to make microscopy more accessible,’ says Schwentker.
Olympus has also launched its DSX series of opto-digital microscopes for inspection, capturing an image by standard microscopy with a camera. They provide high optical quality combined with excellent digital processing. The user interacts with the microscope through a touch screen interface and the software is designed to make the microscope easy to use.
‘There are many techniques to bring out certain features in the image, but generally you have to be an expert in microscopy to do that,’ says Schwentker. ‘The DSX series is designed to be able to be used by those without any particular expertise in microscopy.’ As an example, he says the system has a function called ‘best image’, whereby it takes nine images with different settings for the user to select which one brings out the desired feature. The microscopes are also motorised, so set-up is automatic to get the best image without having to be an expert in microscopy.
A lot of life science imaging concerns fluorescence, with its ability to denote the expression of genes or the movement of proteins in living cells. Hyperspectral imaging, collecting information across a number of wavebands, to brightfield or fluorescence microscopy allows the system to distinguish between emissions that are close together.
Gooch and Housego has developed its HSi-440C hyperspectral imaging system, to perform unmixing or classification of the image in near real time. ‘Most hyperspectral systems collect the data first and then apply unmixing algorithms after the event. It usually requires several steps to do that,’ says Alex Fong, senior VP at Gooch and Housego. ‘The aim is to analyse the image and classify all the pixels according to spectral profile. It’s a computationally intensive process and normally it’s carried out sequentially, with a spectral image of a scene captured and then processed.’
The HSi-440C can process the data as it’s being acquired, due to the spectral imager’s ability to switch between wavelengths incredibly quickly. Switching to different wavelengths occurs at 100µs, around 1,000 times faster than competing technology, says Fong. Combine that with a high-performance graphics processing card and advanced algorithms, means the system can capture and process the spectra in near real time – the frame rate is around 7fps and each band-pass is corrected at 27.6fps.
‘Being able to separate out the various wavelengths in real time is useful to a biologist or a life science researcher in that they can set up the parameters to unmix one portion of a slide and then scan around the rest of the slide for whatever morphological phenomenon is of interest,’ says Fong.