拉曼光谱基线校正解读

2 Spectroscopy 29(2) February https://www.360docs.net/doc/0913929558.html,Baseline Correction for Raman Spectra Based on Piecewise

Linear Fitting

The correction of baseline drift is an import part for data preprocessing. An interval linear fitting method based on automatic critical-point-seeking was improved, which made it possible for the baseline to drift automatically. Experimental data were acquired from the sulfamic acid catalytic

reaction of the aspirin system, which consisted of different proportions of aspirin. A simulated base-line with different interval values of moving average smoothing determined setting parameters in this method. After baseline drifts caused by fluorescence are removed, the differences of character-istic aspirin peaks proved the efficiency of this method.

Kuo Sun, Hui Su, Zhixiang Yao, and Peixian Huang

Rcharacterization for its ability to obtain information on vibrations from samples. It can also be used for on-line monitoring using a fiber-optic Raman probe (1,2). The Raman spectra show the characteristics for species in sharp and dense peaks. However, during the application of Raman spectroscopy, fluorescence of organic compounds in the samples, which are sometimes several orders of magnitude more intense than the weak Raman scatter, can interfere with the Raman signals (3). A phenomenon of baseline drift shows up, making the resolution and analysis of Raman spectra impractical.Both instrumental (4) and mathematical methods have been developed to reduce the drifted baseline caused by fluorescence. The use of an excitation wavelength such as 785–1064 nm lasers, which does not eliminate fluorescence (5), is the most traditional instrumental method. Raman scattering is directly proportional to the fourth power of frequency; as the excitation wavelength increases, the sen-sitivity of the Raman becomes severely reduced. The use of anti-Stokes Raman spectroscopy is another method, based on theory (6). Mathematical methods (7–10) include the first and second order derivatives, wavelet transform, me-dian filter, and manual polynomial fitting. These methods are useful in certain situations, but still have some limita-tions. For example, derivatives are effective, but as a result the shape of the Raman spectrum is changed; wavelet trans-form can be differentiable in the high- and low-frequency components of the signals; however, it is difficult to choose a decomposition method. Manual polynomial fittings re-quire the user to identify the “non-Raman” locations manu-ally (11), and afterwards the baseline curve is formed by fitting these locations. Consequently, the result involves the inevitable subjective factors and, in addition, the

workload is always heavy. Therefore, it is important to choose an op-timal decomposition method.Piecewise linear fitting based on critical-point-seeking was proposed in this study. The method determines an op-timum corrected spectrum by correlation analysis, which can conquer these limitations. A Raman spectrum from the sulfamic acid catalytic reaction of an aspirin system was used as a study subject. By using this method, the Raman spectrum drifted baseline was automatically eliminated, leaving only the corrected spectrum. Basis of Qualitative and Quantitative Raman AnalysisA Raman spectrum is a plot of the intensity of Raman scattered radiation as a function of its frequency differ-ence from the incident radiation (usually in units of wave-numbers, cm-1). This difference is called the Raman shift, which is the basis of qualitative analysis (12). The intensity or power of a normal Raman peak depends in a complex way upon the polarizability of the molecule, the intensity Theory and Method

Figure 1: The process of interval linear fitting baseline correction.tively lower the frequency and sensi-tivity as shown in equation 2: xk,smoothxk+i

(=2+1)

[2]

where ω is the interval number of the

moving average smoothing window, which must be an odd number.Step 2: Locating Local Extrema

Suppose that the function f(x) has an extreme value at point x = x0 in a certain neighborhood (x0 –δ, x0 + δ)where the derivative of the function is defined and is not 0. If x ? (x0 –δ, x0) the derivative is positive, whereas if x ?(x0 + δ, x0) the derivative is nega-tive, f(x0) is maximum, otherwise f(x0) is minimum. After finding the local minimum, we c an get a set of minimal critical points λi (i = 1, 2, . . ., n). The Raman spectrum was divided into n - 1 intervals by λi.

Step 3: Fitting Baseline

Every interval (λ1,λ2), (λ2,λ3), . . .,

Figure 2: The sampling Raman spectrum device.

(λi–1,λi) can be fitt ed in a linear equa-tion as shown in equation 3:

φ(x)=))+

(f(λi)–f(λi-1))((λ+λ)

(x–)+

ii-1(f(λi)+f(λi-1))

x∈(λi-1,λi)i=(2,3,...,n)

[3]

where φ(x) is the fitting baseline.Step 4: Removing Baseline

A corrected spectrum F(x) was acquired after the fitting baseline was removed from the original spectrum.

[4]

Step 5: Performing Correlation AnalysisA correlation analysis method was conducted between the original spec-trum X and the corrected spectrum Y as shown in equation 5: F(x)=f(x)–φ(x)

rcoe [5]

The flowchart is shown in Figure 1.

https://www.360docs.net/doc/0913929558.html,

-1). With Figure 3: Original Raman spectra of samples.fitting, and output were integrated into a function based on Scilab 5.4.0 (https://www.360docs.net/doc/0913929558.html,/).

Parameter Settings In this approach, the interval number of moving average smoothing is the primary parameter. A suitable inter-val for a simulated baseline is chosen. If a strong baseline slope exists in the Raman spectrum, another regulating parameter must be set, the derivative start point. The Raman spectrum ob-tained at the 18-min mark, with the maximum baseline drift, was selected for setting parameters.The baseline (red) was eliminated without both smoothing and setting of the derivative start point. The cor-rected spectrum is shown in Figure 4. Many Raman peaks (from 300 cm-1 to 1100 cm-1) of components in the reac-tion system were removed as the base-line because of the frequency shift and high sensitivity. An inverse peak (0–100 cm-1) exists because of the high slope in the estimated baseline (0–240 cm-1). With the setting of the derivative start point at 60, the original spectrum of object was handled with the selected interval values of 5, 15, 25, and 35, re-spectively. They are shown in Figures 5a–d.The inverse peak disappears (0–100 cm-1) because of the derivative start point setting, which increased match-ing with the original spectrum. When the interval value was 5 (Figure 5a), useful information (250–1000 cm-1) Figure 4: Correction without setting parameters.Raman Platform SettingA laser with the wavelength of 785 nm was used as the excitation light source (Laser-785, Ocean Optics), and a Raman spectrometer (Scientific-grade QE65000, Ocean Optics) was used for the detector. The Raman information was obtained using a fiber-optic probe (BAC100-785-OEM, Ocean Optics), with Spectrasuite spectroscopy soft-ware (Ocean Optics); the x axis on the workstation menu was selected to be Raman shift, and the selected integral time was 1/s to obtain the Raman spec-trum for the 0–2000 cm-1 spectral range (1044 data points).

Experimental SectionExperimental DataWe saved the Raman spectrum of ace-tylsalicylic acid (AR, Tianjin Guangfu Fine Chemical Research Institute) first. According to the literature (15), we pre-cisely weighed acetic anhydride to 41.0 g (AR, ChengDu KeLong Chemical Co.,Ltd.), salicylic acid to 27.7 g (AR, Tianjin Guangfu Fine Chemical Re-search Institute), and sulfamic acid to 0.5 g (AR, ShanTou XiLong chemical factory). The acetic anhydride, salicylic acid, and sulfamic acid were transferred sequentially to a 100-mL, three-necked, round-bottomed flask that was main-tained at a temperature of 81 °C in a

inal spectrum, a group of correlation coefficients was used consisting of a hit list from the corrected spectrum in the range of selected interval val-ues from 3 to 35 (with the derivative start point setting at 60). The hit list is shown in Figure 6.

background, and the baseline slope was eliminated, all correlation coefficients between the original spectrum and corrected spectra were less than 1. The correlation coefficients were less than 0.7 in the range of interval value from 3 to 7 because the Raman scattering of the measured object was removed. When the interval number was in the range from 9 to 21, the corrected spec-tra matched favorably with the original spectra. Although the correlation coef-ficient increased in the range from 29 to 35, the corrected spectra were distorted as shown in Figure 5d.

Determining the range of the inter-val values is the most important step. In the program, the range of interval values from 9 to 21 is favorable, as seen in Figure 6. The correction spectrum was obtained respectively for every in-terval value in the setting range after correlation analyses were conducted and between the original spectrum and every corrected spectrum we obtained a hit list of correlation coefficients. When Figure 5: Correction with the derivative start point setting at 60 and interval numbers at (a) 5, (b)15, (c) 25, and (d) 35.

Figure 6: Hit list in the range of interval numbers.

correlation coefficients on the list were served. The Raman spectrum of aspirin closest to 1, the corrected spectrum was is shown below the corrected spectra

of the samples. The Raman spectrum optimum.

of aspirin has characteristic bands in

Applicationthe region from 700 cm-1 to 800 cm-1, The six Raman spectra obtained di-which can be assigned to the aromatic rectly in the reaction system were cor-ring CH-deformation vibrations. The rected with this program.feature at 1045 cm-1 is attributed to the Figure 7 shows that the position of OH-bending vibration. Raman bands Raman peaks could be discerned in in the 1606– 1630 cm-1 spectral region the corrected spectra of samples, which are caused by both the CC-stretching means the information about compo-vibrations and CO-stretching vibra-nents in the reaction system is pre-tions of the carboxyl group (16,17).

https://www.360docs.net/doc/0913929558.html,

Spectrochim. Acta, are with

Figure 7: Hit list in the range of interval numbers.

Scatter intensity in Raman shift re-gions (700–800 cm-1, 1040–1050 cm-1, and 1600–1630 cm-1) increased with increasing reaction time.

(1) H. Torii, A. Ishikawa, and M. Tasumi, J. Mol.

Struct. 413, 73–79 (1997).

(2) S.K. Khijwania, V.S. Tiwari, F.-Y. Yueh, and

J.P. Singh, Sens. Actuators, B 125, 563–568 (2007).

Conclusions

(3) R.L McCreery, Raman Spectroscopy for

Chemical Analysis (Wiley-Interscience, New York, New York, 2000), pp. 25–26. baseline, a piecewise fitting method was developed that was able to automatically correct the base-line from acquired data, particularly for the fluorescence background in Raman spectra.

The interval value of a moving av-erage smoothing is the primary pa-rameter in the programming. Proper parameters were selected according to correlation coefficient to the highest value (closest to 1) in the hit list. This method makes characteristic peaks identifiable for further analysis, which could improve the quality of Raman spectra in other fields.

(4) E.A.J. Burke, Lithos. 55, 139–158 (2001). (5) J. Funfschilling and D.F. Williams, Appl. Spec-trosc. 30, 446 (1976).

(6) P.A. Mosier-Boss, S.H. Lieberman, and R.

Newbery, Appl. Spectrosc. 49, 683 (1995).

(7) A. O’Grady, A.C. Dennis, and D. Denvir, Anal.

Chem. 73, 2058–2065 (2001).

(8) A.V. Jagtiani, R. Sawant, and J. Carletta,

Meas. Sci. Technol. 19, 15 (2008).

(9) Y. Wang and M. JY, Comput. Appl. Chem. 30,

701–702 (2003).

(10) Y. Zhang, P. Zhong, and J.S. Wang, Comput.

Appl. Chem. 24, 465–468 (2007).

(11) C.A. Lieber and A. Mahadevan-Jansen, Appl.

Spectrosc. 57, 1360–1367 (2003).

Acknowledgments

(12) X. Chu, Molecular Spectroscopy Analytical

Technology Combined with Chemometrics and its Applications (Chemical Industry Press, Beijing, China, 2011), pp. 311–314.

The authors would like to acknowl-edge foundation support and research fellowships from the Biochemical Pro-cess Detection and Control Laboratory of the Department of Biology and Chemical Engineering, at Guangxi University of Science and Technology.

(13) Z. Wu, C. Zhang, and C. Peter, Catal. Today

113, 40–47 (2006).

(14) R. Sato-Berrú, Y. Medina-Valtierra, J. Medina-

Gutiérrez, and C. Frausto-Reyes, Spectro-chim. Acta, Part A 60, 2231–2234 (2004). References

(15) S. Yang, Journal of Kunming Teachers Col-

lege 29, 108–109 (2007).

For more information on this topic,

please visit our homepage at: https://www.360docs.net/doc/0913929558.html,