App Note 52 - SERDS - Newport Corporation
Contents
1. Figure 6 Comparison of R2 for identification of ethanol in rum vodka mixtures obtained by SERDS at 785 nm conventional Raman with baseline subtraction method at 785 nm and 1064 nm Raman with baseline subtraction In the case of 785 nm measurements the maximum exposure time was selected to bring the signal close to the CCD saturation Different number of averages was taken with the rule that the total measurement time remained the same as for pure vodka while the fluorescence increased and the exposure time decreased GO Newport 5 Experience Solutions compound Note that the R is sensitive to any component of the residual be it the shot noise or other spectral features Figure 6 compares the R values for the ethanol in the dark rum mixtures obtained via SERDS baseline fitting and long wavelength Raman with baseline fitting The R values are shown as a function of relative concentration of rum in the mixture with vodka As is clearly obvious from the figure the R deteriorates quickly for the baseline fitting method and gets noticeably worse even when measured on the 1064 nm system even though baseline fitting was applied in that case as well Due to the fact that SERDS spectra are automatically free from any background have zero integral over the spectral window and nearly perfect derivative shape they are extremely convenient for performing automated Raman analysis as they turn out to be nearly orthogonal for majori2. 4th power of laser wavelength which means that it is 3 4 times weaker with 1064 nm excitation as compared with commonly used 785 nm excitation Generally it means that significantly longer collection times are necessary to collect as much Raman signal when using 1064 nm laser Furthermore the use of 1064 nm excitation laser requires using InGaAs detectors instead of silicon CCDs In case of dispersive Raman systems these are InGaAs arrays which typically have much higher thermal noise than that of a silicon CCD at the same temperature As a result InGaAs detector arrays for dispersive Raman instruments must always be cooled to lowest practical temperature to achieve satisfactory noise levels Although cooling the detector is not a technical obstacle for laboratory instruments it presents an obvious issue for portable battery operated Raman instrumentation as it means more power consumption larger battery and shorter battery life in addition to increase in cost and complexity of the instrument For that reason a lot of attention has been paid over the years to other methods of combating fluorescence in Raman measurements which can be separated into two classes 1 the digital filtering methods and 2 the physical methods of subtracting fluorescence contribution from the Raman signal of which the Shifted Excitation Raman Difference Spectroscopy or SERDS is the prime example Shifted Excitation Raman Difference Spectroscopy The u
3. IR is another convenient feature which allows the end user the freedom of choice of the excitation wavelength Since these laser sources are compact they are very compatible with field portable Raman instruments which presents an important advantage as the signal collection in the field is often not free from background light and the samples are often not pure thus increasing the likelihood of encountering interference from the fluorescence of the contaminants Being able to analyze fluorescent samples while using the preferred i e shorter excitation wavelength has the advantage of being able to use silicon CCD detectors which have high sensitivity lower thermal noise and low price tag associated with them Furthermore SERDS can be implemented without any changes to the Raman probe and spectrometer hardware making it simple and versatile for equipment manufacturers When equipped with silicon CCD even a portable Raman instrument will have higher resolution and faster signal processing times then even laboratory size long wavelength Raman instruments Because the Raman signal is so much stronger at shorter wavelengths the signal collection times are often many times faster for 785 nm systems for example as compared with the 1064 nm system at similar laser illumination intensities As the signals for the two excitation wavelengths are collected in rapid succession the SERDS method does not require collecting the background si
4. that are convenient to test with the head pointed horizontally The collection fiber delivers the signal to a bench top system F 3 spectrometer with 30 cm focal length and 600 groves per millimeter diffraction grating The spectrometer has adjustable slit and variable resolution The detector was a cooled CCD with 1024x256 pixel array from Princeton Instruments For SERDS analysis the spectra are collected sequentially with laser 1 and then with laser 2 In LS 2 laser source both lasers run continuously during the course of the experiment to assure best stability of the output power whereas fiber optic switch is used to cut off the light to the sample The wavelengths of the lasers are stable to lt 5 pm over the course of the day Measurements In order to compare the performance of SERDS with conventional methods such as baseline fitting in a situation where high amount of fluorescence obscures the Raman signal it is convenient to use a simple substance with strong and variable fluorescent background For that purpose we have analyzed mixtures of dark rum whose Raman spectrum Raman spectra of rum taken at 785 nm 60000 50000 0 10 40 70 100 40000 30000 Light intensity a u 20000 10000 0 300 500 700 900 1100 1300 1500 1700 1900 Raman shift Figure 3 Raman spectra of rum an Vodka mixtures taken at 785 nm The plots are shown for di
5. 4 nm excitation Raman Intensity SERDS Raman Intensity Raman shift wavenumbers Emerald 785nm Emerald 1064nm Emerald SERDS Figure 10 Raman spectra of emerald taken at 785 nm and 1064 nm SERDS spectrum of emerald is also shown to demonstrate the clarity of spectral features resolved by this method example Fig 10 shows Raman spectra of emerald collected at 785 nm and 1064 nm As is evident from this result the amount of fluorescence is not reduced by moving to the 1064 nm excitation The SERDS method on the other hand produces very clear and identifiable spectrum as is shown in the same figure Summary and conclusions As is evident from these results the SERDS method offers distinct advantages for Raman analysis in most situations Its advantage in comparison with the digital filtering methods is primarily in accurate elimination of the fluorescence whereas the latter methods allow only approximate removal of the fluorescent background and only in the situations where it is varying very slowly Even in the examples considered here where the fluorescence has no identifiable spectral features the residual left after the baseline removal deteriorates the R2 significantly In comparison with 785 nm excitation the 1064 nm Raman system usually produces much smaller amount of fluorescence as is the case for the dark rum For that reason it has traditionally been used a
6. APPLICATION NOTE 2 Background Free Raman Spectroscopy using SERDS Technique with LS 2 Dual Wavelength Laser Source Newport Corporation DANGER GO Newport Experience Solutions Introduction Raman spectroscopy has been experiencing a period of growing interest for qualitative and quantitative analysis in a vast scope of applications pertinent to various industries including pharmaceuticals petrochemical and law enforcement This growth is enabled by increase in the offering of affordable Raman instrumentation some are portable and ruggedized for field use Such increase in availability and portability is powered in turn by compact high performance wavelength stabilized laser diodes However with widening of the scope of applications for Raman spectroscopy the challenge of sample fluorescence becomes more and more evident Besides the natural fluorescence that many substances possess there are issues of sample contamination with fluorescent compounds in the field that one must deal with in real life situations One example of that is fluorescent agents such as caffeine and flour regularly used in cutting illicit street drugs Although the long wavelength typically 1064 nm Raman instruments have been considered the benchmark in dealing with fluorescent substances working with that excitation wavelength presents a number of issues First of all as is well known the Raman signal diminishes as the
7. Vodka 25000 25 20000 30000 55 15000 85 Raman signal a u 100 10000 5000 5000 500 10082man shift cr 500 2000 2500 Figure 4 Raman spectra of the rum vodka mixtures taken at 1064 nm Spectra are shown prior to any corrections instrument As expected the long wavelength excitation produces much smaller amount of fluorescence although fluorescence is still prominent even in that wavelength region which makes the baseline removal a must even in this case in order to be able to perform substance recognition To keep the measurement conditions similar all the experiments were performed at the same power on the sample 450 mW for both the 785 nm and the 1064 nm systems For better comparison the spectral resolution of the Spex spectrometer was also adjusted by opening of the slit to such width as to match the resolution of the 1064 nm Agility system approximately 18 cm l The exposure time was always selected to keep the maximum of the signal just below the saturation level of the detector array Therefore when large amount of fluorescence was collected the exposure time was shortened accordingly When performing analysis of highly fluorescent samples signal averaging should be employed to increase of the signal SERDS spectra vs baseline fitting 6000 5000 4000 3000 2000 1000 1000 2000 3000 4000 Raman Intensit
8. bert Adv Drug Deliv Rev 57 1144 2005 3 L S Taylor and FW Langkilde J Pharm Sci 89 1342 2000 4 W P Findlay and D E Bugay J Pharm Biomed Anal 16 921 1998 5 D Pratiwi J B Fawcett and K C Gordon T Rades Eur J Pharm Biopharm 54 337 2002 6 W M Chung Q Wang U Sezerman and R H Clarke Appl Spectrosc 45 1527 1991 7 J B Cooper P E Flecher T M Vess and W T Welch Appl Spectrosc 49 586 1995 8 M Ku and H Chung Appl Spectrosc 53 557 1999 9 R J Stokes K Faulds and W E Smith Proc SPIE 6741 67410Q1 2007 10 R E Littleford P Matousek M Towrie A W Parker G Dent R J Lacey and W E Smith Analyst 129 505 2004 11 E Horvath J Mink and J Kristof Mikrochim Acta 14 745 1997 12 K Faulds W E Smith D Graham and R J Lacey Analyst 127 282 2002 13 B Saegmuller G Brehm and S Schneider Appl Spectrosc 54 1849 2000 QAD Newport 7 Experience Solutions Newport Corporation Worldwide Headquarters 1791 Deere Avenue Irvine CA 92606 In U S 800 222 6440 Tel 949 863 3144 Fax 949 253 1680 Email sales newport com Newvyport Experience Solutions Visit Newport Online at www newport com This Application Note has been prepared based on development activities and experiments conducted in Newport s Technology and Applications Center and the results associated therewith Actual
9. ccurate analysis For these reasons we offer a comparatively much more simple and practical approach to performing SERDS analysis using two affordable wavelength stabilized laser diodes operating at slightly offset wavelengths The lasers are stabilized by use of the Volume Bragg Grating VBG technology and are very compact efficient and inexpensive making them well suited for portable battery operated Raman instrumentation Advantages of SERDS The most important advantage of the SERDS method over the conventional methods of digital removal of the fluorescence background is accurate elimination of the fluorescence contribution as opposed to approximation of such using a polynomial function as is the case in baseline fitting for example This advantage is particularly obvious when the background fluorescence has some spectral features but even in the situations when the fluorescence has a slowly varying smooth shape the advantages of SERDS are very apparent Using two fixed wavelength VBG stabilized laser sources is a particularly simple and efficient method of implementation of SERDS Such sources are now available in high brightness and high power at a variety of wavelengths and are very reliable and stable in wavelength over the lifetime of a device These lasers do not require constant wavelength monitoring as their wavelength is known and stable Availability of such sources in a wide range of spectrum from the 400 nm to the near
10. e which is then collimated by an aspheric lens After the lens there is a laser clean up filter a 3 nm bandpass filter which serves to cut down the amplified spontaneous emission from the laser source as well as the Raman signal from the delivery fiber Since the system is using larger aperture components and longer distances lens relays are employed in order to properly guide the laser and most importantly signal light throughout the system The first lens relay is comprised of lenses RLI and RL2 which project the front focal plane of the laser collimating lens onto the front focal plane of the sample focusing collection lens The laser light is folded by a fold mirror and the first notch filter and then one more time by a pentaprism installed in a rotating AQAD Newport 3 Experience Solutions swivel mount This is done in order to have the flexibility of pointing the laser down up or horizontally depending on the type of a sample being studied without affecting the alignment of the system and therefore its collection efficiency The Raman signal is collected by the focusing collection lens in the back scattering geometry and then relayed by the lens pair RL2 and RL3 onto the front focal plane of the focusing lens that couples the Raman signal into the collection fiber The signal passes through two notch filters that effectively eliminate the Rayleigh scattering from the sample In order to have an ability to view the sample a
11. e of the 1064 nm system This is due to smaller Raman scattering cross section at longer wavelength as well as to the difference between Si and InGaAs detectors that the measurement times for the 1064 nm system are typically longer than those for the 785 nm systems approximately by factor of 5 for the systems employed in this study Therefore an interesting comparison between the 785 nm and thel064 nm excitation wavelength would be the coefficient of determination commonly referred to as R2 achieved within the comparable total collection times It is calculated according to the following formula os ye Oi y 2 1 where y are the measured spectral intensities at pixel or the spectral channel number i N is the total number of pixels in the detector or the number of the spectral channels fi is the spectral intensity of the library spectrum at the same point in spectrum and y is the average of the measured spectrum k The R coefficient allows to evaluate the relative quality of the Raman analysis under the conditions of the experiment It provides a measure of the magnitude of the residual between the measured spectrum and the library spectrum of a Comparison of methods R coefficient Coefficient of determination of ethanol in rum vodka mixtures for different methods 1 2 0 8 0 4 t 1064 nm 0 2 0 20 40 60 80 100 120 Rum percentage SERDS Baseline fitting
12. ecautions This Application Note shall not be copied reproduced distributed or published in whole or in part without the prior written consent of Newport Corporation Copyright 2013 Newport Corporation All Rights Reserved The Newport N logo is a registered trademarks of Newport Corporation Newport Corporation Irvine California has been certified compliant with ISO 9001 by the British Standards Institution DS 111301 11 13 AD Newport Experience Solutions
13. fferent concentration of rum relative to vodka The exposure time remained constant for all different concentrations is essentially that of pure ethanol with vodka which has about the same concentration of the analyte i e ethanol in this case Figure 3 shows raw Raman spectra of the dark rum collected with 785 nm The spectra are shown before any baseline correction has been performed As can be observed the amount of fluorescence in the dark rum obscures the Raman signal nearly completely when the 785 nm excitation is used Pentaprism A Swivel In addition to the comparison between SERDS and numerical filtering methods mentioned above we have performed long wavelength Raman analysis with 1064 nm laser excitation as a benchmark comparison Long wavelength Raman has long been considered the preferred method for studying highly fluorescent compounds so it was rather interesting to evaluate it side by side with SERDS performed at 785 nm We have used a commercial 1064 nm laboratory dispersive Raman system for these experiments The data were collected on the same samples which were then processed by the same methodology including baseline removal to compare the results Figure 4 shows the spectra collected on the 1064 nm Raman AD Newport 4 Experience Solutions Raman spectra of rum taken at 1064 nm Raman spectra of rum added to vodka mixtures spectra taken at 1064 nm 45000 40000 35000
14. gnal independently as it is automatically corrected for it during the subtraction of the two collected spectra And finally for majority of the practical applications the differential shape of the SERDS spectra is very convenient as the spectra are naturally centered i e have zero integral and are nearly orthogonal for dissimilar substances This feature makes them very fitting for all automatic signal processing algorithms be it quantitative Raman impurity detection or unknown substance recognition Experimental The SERDS measurements are performed using a dual laser SERDS laser source LS 2 produced by Newport The lasers operate at two closely spaced wavelengths which in the case of the measurements shown here are 784 5 nm and 785 5 nm The wavelength separation is selected to correspond to the approximate line width of the Raman lines of the substances under study Note that the exact wavelength separation is not significant for the accuracy and practicality of SERDS however wavelength separation that is much smaller than the width of the Raman bands would result in increase in noise in SERDS spectra The LS 2 laser source delivers the output of either one of the two lasers to the output port via a fiber optic switch A fiber optic cable is then attached to the output port of the LS 2 and coupled into a Raman probe or a bench top Raman system Fig 2 The system is designed to accept a fiber coupled input from a laser sourc
15. nd the tested area we have introduced a viewing camera which creates an image of the sample using a small portion of collected light reflected by the beamsplitter The image of the sample is created by the camera focusing lens that is positioned after the relay lens RL4 The system also employs a reticle that allows centering the laser light on the sample not shown in Fig 2 Experimental setup RL3 Camera lens Viewing camera Collection fiber To spectrometer Focusing lens Beamsplitter Notch filters ere Fold mirror 4 Ds lens Collimating RL1 yy Cleanup filter Laser y i Source fiber Sample i Focusing collection lens Figure 2 Schematic of the bench top Raman system on which the measurements were taken The lenses RL1 RL4 are optical relay lenses projecting the front focal plane of the laser collimating lens onto that of the sample focusing collection lens RLI and RL2 pair and then to the front focal plane of the collection fiber focusing lens RL2 and RL3 pair as well as the observation camera arm RL2 and RL4 pair The pentaprism mounted in a swivel mount allows to point the focusing head of the system in any direction from facing down to horizontally to straight up which is convenient when dealing with different type of samples liquids in vials viewed from below solid samples viewed from above and liquids in flat wall vials as well as large objects
16. nderlying principle of the SERDS is the fact that Stokes and Anti stokes components of the Raman scattering are frequency shifted relative to the excitation wavelength and therefore are tied to it Laser induced fluorescence on the other hand is independent from the excitation wavelength and retains practically identical spectral shape when the excitation wavelength experiences a relatively small change lt 10 nm for example For that reason if two spectra are collected with a small shift in the excitation wavelength and subsequently subtracted the resultant spectrum will be free from the fluorescence contribution while retaining the Raman signal albeit in a form of a difference spectrum Fig 1 Fluorescence Dmecccanes Seenens taaan seseanT we e gt 2a oP D D lt D a k 2 X a Ss D o eS a Se Spectrum 1 Spectrum 2 Raman difference spectrum Figure 1 Illustration of the principle of SERDS For this ability of accurate subtraction of the fluorescence background the SERDS approach has long been considered attractive effectively expanding the use of the conventional dispersive Raman instrumentation equipped with inexpensive and efficient CCD detectors to the classes of samples that exhibit strong fluorescence SERDS method was proposed and tested over time in various application areas generally with good success with respect to eliminating the fl
17. results may vary based on laboratory environment and setup conditions the type and condition of actual components and instruments used and user skills Nothing contained in this Application Note shall constitute any representation or warranty by Newport express or implied regarding the information contained herein or the products or software described herein Any and all representations warranties and obligations of Newport with respect to its products and software shall be as set forth in Newport s terms and conditions of sale in effect at the time of sale or license of such products or software Newport shall not be liable for any costs damages and expenses whatsoever including without limitation incidental special and consequential damages resulting from any use of or reliance on the information contained herein whether based on warranty contract tort or any other legal theory and whether or not Newport has been advised of the possibility of such damages Newport does not guarantee the availability of any products or software and reserves the right to discontinue or modify its products and software at any time Users of the products or software described herein should refer to the Users Manual and other documentation accompanying such products or software at the time of sale or license for more detailed information regarding the handling operation and use of such products or software including but not limited to important safety pr
18. s the benchmark in Raman analysis However as the side by side comparison with SERDS in similar experimental conditions shows SERDS can produce not only comparable but superior in quality analysis even when strong fluorescence interference is present In the case under study the advantages of stronger Raman signal and the better detector enjoyed by the 785 nm system resulted in nearly the same R for the highly fluorescent rum in about the same total acquisition time Furthermore as the example of Raman analysis of a specimen of emerald demonstrates the long wavelength excitation is not the ideal solution for all cases because some of the substances will exhibit fluorescence even with 1064 nm excitation Due to much stronger signal at shorter wavelengths and better sensitivity and noise of the CCD arrays the total analysis time in SERDS measurements is at worst comparable to that of a 1064 nm system And the SNR of SERDS measurements can be improved in a straightforward manner by simply increasing the number of averages for a strongly fluorescent sample Overall the drawbacks assigned to SERDS historically are almost entirely connected with the reconstruction of the difference spectra When no reconstruction is performed SERDS produces superior results in all circumstances as compared with conventional Raman References l C Gendrin Y Roggo and C Collet J Pharm Biomed Anal 48 533 2008 2 S Wartewig and R H H Neu
19. th R6G dye added to create fluorescence background The spectra were taken with the two lasers with different wavelength This amount of fluorescence was added to the mixtures for quantitative determination of the relative concentration of the components library spectra The Raman spectra were collected on a series of solutions with different concentration of methanol in mixture with ethanol All solutions contained large amount of R6G dye to simulate strong interference from fluorescence The raw Raman spectra of one of these solutions are shown in Comparison of SERDS spectra with numerical derivative SERDS spectrum vs derivative Signal a u p f Wavelength nm SERDS data Derivative Figure 8 SERDS spectra of the methanol ethanol mixture obtained using the spectra shown in Fig 8 blue line and the derivative spectrum obtained with the same date numerically magenta line The difference in SNR of the two spectra is obvious Fig 7 Large amount of fluorescence blocks the Raman features nearly completely and the peaks that are resolvable are actually those of the R6G dye Figure 8 shows SERDS Spectrum produced from these Raman spectra and also a spectrum obtained via numerical differentiation As can be plainly observed SERDS spectrum is much less noisy than the one obtained numerically Figure 9 shows the predicted concentrations of methanol in the mixt
20. ty of the substances of interest to Raman analysis A case in point is quantitative Raman analysis which can be illustrated on relative concentration measurements in binary mixtures e g of methanol and ethanol performed in the presence of a strong fluorescence background In this experiment a SERDS spectrum of an unknown mixture was compared with SERDS spectra from a library of a few dozen of pure components the two most likely components contained in the mixtures were identified and then relative concentration of the two was determined Note that no multivariate calibration methods e g partial least square regression PLSR were employed i e the system was not trained on the known mixtures rather all samples were treated as completely unknown In order to compare the results obtained by SERDS with what conventional Raman analysis would predict the fluorescence background was removed via numerical differentiation and then similar analysis was carried out as in the case of SERDS i e the spectra of the mixtures were compared with the Raman spectra of methanol ethanol mixture with large amount of R6G dye added 12000000 10000000 8000000 6000000 Amplitude a u 4000000 2000000 780 800 820 840 860 880 900 wavelength nm Figure 7 Uncorrected Raman spectra of a methanol ethanol mixture wi
21. uorescence background and extracting the Raman signal from it In majority of the studies conducted on SERDS large emphasis was put on the reconstruction of the Raman spectrum of the substance under study from the pseudo derivative one obtained as a product of SERDS process As a result the often quoted drawbacks of SERDS method are in fact the deficiencies of the various reconstruction algorithms At the same time it is a well understood fact that reconstruction is not required for any of the numerical algorithms involved in substance identification detection or quantitative measurements by means of Raman spectroscopy When the need of spectrum reconstruction is removed SERDS approach will deliver clearly superior performance to any of the numerical processing algorithms designed to eliminate the slowly varying fluorescence background QAD Newport 2 Experience Solutions Historically the SERDS method has been used in laboratories employing tunable wavelength lasers However the use of this type of laser sources presents a significant challenge for portable Raman systems Generally these lasers are much more costly than simple and efficient wavelength stabilized laser diodes But in addition to this there is an issue of the exact wavelength control over these lasers required to perform accurate qualitative and especially quantitative Raman analysis as it is not fixed or stable which requires constant active wavelength monitoring for a
22. ures obtained via SERDS and the numerical differentiation As one can see even though both methods give reasonably accurate prediction of the methanol concentration when it is present in large proportions the picture is quite different when lower proportion of the Determination of methanol concentration ina mixture with ethanol and R6G dye Alcohol concentration prediction with R6G in solution Methanol SERDS Perfect prediction Methanol Drvtv Predicted fractional concentration Actual fractional concentration Figure 9 Comparison of quantitative determination of concentration of methanol relative to ethanol in a binary mixture with R6G dye added to create fluorescence background as shown in Fig 8 The concentration was determined using SERDS and numerical differentiation to remove slowly varying fluorescence background The straight line shows the ideal values methanol is present in the mixture The conventional method starts loosing accuracy at about 20 25 methanol content and then fails completely at concentrations lower than 10 SERDS on the other hand predicts accurate concentrations QAD Newport 6 Experience Solutions even at 5 methanol content in the mixture Besides the advantages of SERDS illustrated above there are some substances which produce strong fluorescence when measured on a long wavelength Raman instrument For Raman spectra of emerald taken at 785 nm and 106
23. y se 400 600 800 1000 1200 1400 1600 1800 2000 5000 4000 3000 2000 1000 Raman Intensity Oo 1000 2000 3000 400 600 800 1000 1200 1400 1600 1800 2000 Raman shift cm Rum 1 average Rum 10 averages Rum 100 averages Ethanol Figure 5 SERDS spectra of rum top and spectra of rum obtained by the conventional method of baseline fitting bottom compared with that of pure ethanol The spectra were obtained with different number of averages and have been normalized and offset for clarity Different order polynomial functions were tested for the baseline removal and the best results are shown to noise ratio Note that due to the shortening of the exposure time when fluorescence is present there can be a proportionately larger number of averages performed within the same total measurement time Improvement in SNR with averaging is evidenced in the results shown in Fig 5 where spectra of the dark rum collected with 785 nm excitation are shown for different number of averages and compared with library spectra of pure ethanol The spectra are vertically offset for clarity As is clearly obvious from the figure the amount of noise is decreased significantly with averaging and becomes nearly negligible at 100 averages Note that the total acquisition time for 100 averages with 785 nm system was still under the time required for a single exposur
Download Pdf Manuals
Related Search