TECHNOLOGY RamaN SpECTROSCOpY ➤
several wavelengths, then they need several lasers, but OPSL lasers are entirely interchangeable,’ says Pfeufer. ‘If you know how to work with one laser, then you know how to work with every other wavelength product.’ The alternative, he says, would be to mix and match between diode and DPSS technologies, taking certain wavelengths from each selection as are available. ‘Mixing technology also means mixing suppliers, mixing interfacing, and mixing geometries of the housing and laser beam; this is ugly, and leads to difficulties.’
Why choose one wavelength over another?
Although Raman spectroscopy can work at any wavelength from IR to UV, being able to select a wavelength for a given application is important to many customers. UV Raman in particular, Pfeufer says, is important. ‘The Raman signal is a weak signal, and its intensity scales with 1/ λ4
. As such, halving the excitation wavelength increases the signal by four times.’
The sample, or the environment it is in, will often dictate that a certain wavelength be used, particularly in biological applications. ‘When imaging biological tissue, we have to fight with other fluorescence,’ says Pfeufer. ‘Putting short or visible wavelength light into a medium generally means a higher chance of causing fluorescence, to the extent that we can no longer see the Raman signal. We try to balance the need for strong signals against avoiding fluorescence in the sample.’ These biological applications include performing Raman spectroscopy on living tissue, which could one day be used to screen for cancer marker compounds. ‘At the moment, this is still at an R&D stage, but there are dozens of institutes working heavily on this, so it will come in time,’ says Pfeufer. Raman spectroscopy will ultimately allow doctors to detect degeneration of the skin instantaneously, or to check tumours for cancerous tissue more quickly than is currently possible. ‘Imagine that a patient is in an
18 ELECTRO OpTiCS l june 2011
The laser linewidths required vary greatly between various applications Image courtesy of WiTec
operating theatre, opened up. The surgeon removes tissue from a tumour to see if it is dangerous or not. If it takes tens of minutes to analyse that tissue, that can be a problem.’
a bigger picture
Raman is not confined to analysis of a single point within a sample, and indeed many of its applications rely on building up a Raman image of the sample, to look at the distribution of a drug within a single pill, for example. WiTec (Ulm, Germany) specialises in producing Raman imaging systems integrated with confocal microscopy, able to produce a 3D image of a sample with information about the distribution of chemicals within it. ‘Our systems can achieve spatial resolution of diffraction limited 200-250nm,’ says Harald Fischer, marketing director at the company. He reiterates the importance of being able to select an appropriate excitation wavelength: ‘The systems are also flexible with respect to the wavelength used, in the 325-785nm range. If samples show fluorescence, especially biological materials, when excited with a visible wavelength such as 532nm, we can shift to 683 or 785nm to get rid of the fluorescence. Also, when looking at larger and more complex molecules, Raman spectroscopy can reveal a lot of information about their properties; different wavelengths of excitation light affect different vibrational modes. Although Raman imaging is
still largely a research-driven field, industrial applications are being
developed. According to Fischer, the technique is particularly useful wherever surface properties are to be analysed. ‘When producing specialised coatings, for example, users need higher and higher resolution imaging tools as the coatings get thinner. Such coatings exist in polymer applications, and also in some novel drug delivery applications,’ he says.
Point-by-point mapping of a sample is not a new technique, Fischer explains, but what makes
200,000 spectra
can be recorded in a single, high- resolution image
WiTec’s solution stand out is the speed at which the image can be taken. ‘200,000 spectra can be recorded in a single, high-resolution image. Also, one system is capable of large area imaging on millimetre scales, as well as high resolution imaging at sub-micron scales. An older point-by- point approach might have needed five to 10 seconds per acquisition, meaning a 200 x 200px image could take three or four hours to acquire. Acquisition time using our current system is approximately 0.1ms per pixel. We achieve this by optimising optical throughput at every stage, choosing optimal components,’ he says, adding that electron-multiplied CCD cameras help reduce acquisition times further.
Stranger wavelengths, larger spectrometers WiTec’s Raman microscope is used in semiconductor and photovoltaic applications in order to build up a stress map of a component on a small scale. ‘A Raman peak shifts depending on how stressed the material is,’ explains Fischer. Elsewhere in the same industry, the same process of stress monitoring is already being used on a larger scale. Erik Schoeffel, VP marketing and sales at US-based spectroscopy specialist McPherson, explains the company’s high-performance Raman systems are used in silicon processing to check the effects of coatings on individual wafers, and to check for relative stresses. ‘Based on this information, the customers can make more accurate predictions about how best to start etching away the silicon to produce the final circuits. Raman spectroscopy is able to give information showing local lattice strain and crystal quality,’ he says. One of the key developments
Schoeffel picks out as having allowed McPherson to address these markets is that the company’s spectrometers are able to work at a long focal distance of up to two metres.
As well as allowing the system to be placed a comfortable distance away, Schoeffel explains that a larger spectrometer can lead to higher performance: ‘Fundamentally, if we think of a diffraction grating of being like a prism – the light goes in and comes out as a rainbow – the longer the distance between the sample and the grating, then the further apart the red and blue are separated in space. By making the spectrometer larger, with a longer stand-off distance, we increase the capability to resolve discrete points in between.’ Schoeffel says the drive for stand-
off detection of perceived security threats such as anthrax can be met in this way – by spectrometers with telescopes in front of them. Having already made the jump
from lab to fab, it seems that Raman spectrometers could soon make a further move into the public domain. l
www.electrooptics.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36