Quantitative Determination of Manganese in Aqueous Solutions and Seawater by Laser-Induced Breakdown Spectroscopy (LIBS) Using Paper Substrates

Abstract

The detection of manganese (Mn) in industrial wastewater and seawater plays an important role in pollution monitoring and the investigation of geochemical and biological processes in the ocean. An approach has been introduced in this work to improve the detection sensitivity of Mn in liquids by laser-induced breakdown spectroscopy with a filter paper as solid substrate. The calibration curves of Mn in aqueous solutions were obtained with the detection of a Czerny–Turner spectrometer and an echelle spectrometer, respectively. The results showed that the Czerny–Turner spectrometer equipped with an intensified charge-coupled device (ICCD) had a more sensitive detection of Mn in aqueous solution with this approach. The limit of detection (LOD) for Mn was down to 0.11 mg/L with laser pulse energy of 90 mJ. With the same approach, the compact echelle spectrometer equipped with an ICCD was used to verify the feasibility for rapid onsite detection. The calibration curves for Mn in simulated industrial wastewater and seawater were constructed to calculate relevant LODs. The LODs of Mn were 2.78 mg/L in mixed solutions and 2.73 mg/L in seawater by calculation. Both the calibration curves and LODs were affected slightly by the matrix effect in the experiment. In order to assess the accuracy, a mixed solution including Mn, Cr, Cd, and Cu with known concentrations was determined, and good agreement between the measured and real values were achieved. It demonstrated that this approach has significant potential for rapid onsite detection of Mn and other metal elements in industrial wastewater and seawater.

Keywords

Laser-induced breakdown spectroscopy LIBS Paper substrate Mn in aqueous solution and seawater Limit of detection

INTRODUCTION

Manganese (Mn) as a heavy metal element could be extremely harmful to the aquatic environment when exceedingly dissolved in both industrial wastewater and seawater.¹ The minimum concentration of Mn tolerable in industrial wastewater discharged is usually no higher than 10 mg/L depending on the type of discharge (domestic, industrial).² Additionally, the geochemistry of Mn in the ocean attracts the attention of investigators to determine its key role in geochemical and biological processes.^1,3 Consequently, the determination of manganese levels in industrial wastewater and seawater is greatly important and necessary. Conventional analytical techniques for Mn in water are accomplished mainly in the laboratory, such as inductively coupled plasma mass spectrometry (ICP-MS) and atomic absorption spectroscopy (AAS). However, due to the presence of a complex matrix, these approaches are time consuming and demanding in sample preparation for precise results.^4,5 Therefore, a technique is imperative for onsite element detection of industrial water and seawater with rapid quantitative determination.

Laser-induced breakdown spectroscopy (LIBS) has drawn growing attention in the past tens of years as a rapid, online, sensitive, and multi-elements analysis method with little or no sample preparation.^6,7 It has been extensively used for a wide range of scientific and industrial applications with different degrees of success.^8–11 However, due to the complex process of the ablation in bulk, the lifetime of plasma is relatively short, which results in poor detection sensitivity compared with that in air.^7,12

In order to improve the detection sensitivity of liquid samples, a variety of methods have been developed with different degrees of success. Converting a liquid sample into a solid one has been proved as an efficient way to increase the LIBS sensitivity, and typical conversion methods correspond to freezing of a liquid sample,¹³ adding liquid drop-wise or collecting its residual onto a filter paper,^14–17 incorporating a liquid in a powder before making a solid pellet,¹⁸ using ion exchange polymer membrane,¹⁹ or obtaining a dried deposition of a liquid on a solid surface.^20,21 The last method has been recently revisited as a surface-enhanced LIBS technique, which takes the advantages of plasma induced on a metallic surface.²² In this work, a filter paper substrate was used to absorb aqueous solution samples. Having compared the paper substrates used in Refs. 15–17, it could be seen that the wet paper substrate did not have to be dried using a hot air blower and then cooled down to room temperature before detection directly. The limit of detection (LOD) of lead in aqueous solution (3.87 mg/L) had been obtained,²³ lower than those obtained with the similar methods above.

In this work, Mn in aqueous solution and seawater are detected using the paper substrate. The calibration curves of Mn in aqueous solutions are obtained with a Czerny–Turner spectrometer and an echelle spectrometer. The LODs obtained with this method will be further compared to those previously reported using other similar methods. In order to verify the feasibility of rapid onsite detection with this approach, the compact detection system (an echelle spectrometer equipped with an intensified charge-coupled device (ICCD)) was used, and the calibration curves and LODs of Mn in simulated industrial wastewater and seawater could be obtained. These investigations are used to achieve a more sensitive detection of Mn and other heavy metals in laboratory and develop a feasible and sensitive method for rapid onsite detection of Mn and other heavy metals in industrial wastewater and seawater by LIBS.

EXPERMENTAL

Experimental Setup. The experimental setup used in this work is shown in Fig. 1. The fundamental output (1064 nm) of a Q-switch Nd:YAG laser (Quantel, Brilliant) with a 10 ns pulse duration operated at 10 Hz was focused onto the surface of a filter paper using a plano-convex lens of 50.1 mm focal length (L1). The filter paper was adhered to a metallic sheet which was mounted on a rotating motor with a speed of 60 rpm. Plasma emissions were collected by a combination of two lenses (L2 and L3), both with a focal length of 38.1 mm, and then coupled into an optical fiber. The output of the fiber was connected to the entrance of a spectrometer equipped with an ICCD (Andor, DH734i-18F-03) which was synchronized to the Q-switch of the laser. In this experiment, two spectrometers were employed. The first one was a Czerny–Turner spectrometer (Acton, SP-2500i) with a grating of 1200 lines/mm, which was used to assure a high throughput but a narrower bandwidth of about 20 nm (the insert in Fig. 1a). The second one was an echelle spectrometer (Andor, Mechelle 5000) that was used to obtain the spectra in a wide spectral range of 220–850 nm. The focal lengths of the lens installed in the Czerny–Turner spectrometer and the echelle spectrometer were 500 and 198 mm, respectively. Each spectrum was obtained using a similar process using 100 laser pulses on average.

Fig. 1.

Schematic diagram of the LIBS experimental setup.

Sample Preparation. Commercially available filter papers (Whatman 42, England) were used as substrate. A metallic sheet (2.5 × 2.5 cm), on which two-layer filter papers were adhered, was fixed on a rotating motor (60 rpm). In this work, the commercial analytical reagent MnCl₂·4H₂O is dissolved with deionized water and seawater (1 km depth of ocean) with a Mn concentration of 500 mg/L, which was used to obtain a set of reference samples by dilution for Mn concentration between 500 and 10 mg/L (500, 300, 200, 100, 50, and 10 mg/L). Prepared samples were dipped gradually, and about 10 mL of the solutions were filtered to avoid the influence of soak time on the measurement repeatability when the paper was dipped into aqueous solutions as in Refs. 16 and 17. Smooth forceps were used to keep the surface of the wet paper surface flat, and samples were left for 3 min prior to detection, giving them time to develop homogeneous surfaces automatically. Our method did not have to be dried and then cooled to room temperature, which was less time consuming and more convenient than those applied in Refs. 15–17.

For a better performance in quantitative analysis, repeatability of measurement is highly valued. Notice that the metallic sheet contained Al mostly and some traces such as Cu. With only one-layer paper adhered, the metallic sheet is at high risk of ablation from the laser cutting through. The spectral lines of elements from the sample were thus interfered by the emissions evaporated from the metallic sheet. In this work, such repeatability was evaluated by observing the variation of the Cu I 324.75 nm line with a Cu concentration of 200 mg/L in solutions as different measurement replications were performed. The obtained results were presented in Fig. 2. In Fig. 2a, the Cu I 324.75 nm line intensities obtained from one-layer paper were higher than those of two-layer because one-layer paper was broken slightly to ablate the metallic sheet, which enhanced Cu I 324.75 nm line intensities but decreased the data reliability and reproducibility. The line intensities obtained with the two-layer paper were similar with those obtained when one-layer paper was adhered on the glass sheet, shown in Fig. 2b. However, the relative standard deviation was improved from about 10 to 5.1% in comparison with one-layer paper. Therefore, two-layer paper was a better choice for quantitative analysis in this experiment. As can be seen in Fig. 3, two photos were taken showing two-layer paper on the metallic sheet before and after laser ablation. The surface morphology of the filter paper exhibited good uniformity after ablation and made it possible to obtain good data reliability and reproducibility.

Fig. 2.

The repeatability assessment carried out under the same experimental conditions: (a) one- and two-layer papers were pasted on the metallic sheet; (b) one-layer paper was pasted on the glass sheet. LIBS signal of Cu 324.7 nm line was used for this assessment, and the horizontal lines represented the mean values of seven measurements.

Fig. 3.

The surface morphology of filter paper before and after laser ablation.

Spectral Line Selection. In order to obtain high performance for quantitative analysis, selection of a sensitive and suitable spectral line for an analytical element to be detected is very important. In this work, Mn in different aqueous solutions was detected with two spectrometers. Figure 4 shows the typical LIBS spectra of Mn in different aqueous solutions detected with the echelle spectrometer: (a) Mn in MnCl₂ solutions, (b) Mn in mixed solutions of Mn-Cr-Cd-Cu with a ratio of 1:1:1:1, (c) Mn in seawater, and (d) Mn in MnCl₂ solution detected with the Czerny–Turner spectrometer. The concentrations of Mn in them were 100, 500, 500, and 5 mg/L, respectively. Each spectrum was presented in the most interesting spectral ranges. Inserts were also used in Fig. 4 to get the details around 260 and 656 nm with enlarged intensity scales. It could be seen that six ionic lines of Mn were detected obviously, and some spectral lines of elements contained in the paper substrate were also observed in the Figs. 4a–4c. As for the criteria of selection, the selected line was first relatively intense for the element in the given conditions of ablation and detection, and it was also required to be as free as possible from any interference with other spectral features. According to the National Institute of Standards and Technology atomic spectra database,²⁴ the relative intensities at 257.61, 259.37, and 260.57 nm were, respectively, 12 000, 6200, and 4300, all three of which were more intense than those around 294 nm. The three ionic lines were observed obviously at a low concentration of 5 mg/L, shown in Fig. 4d. The line emission of 257.61 nm exhibited better the ratio of signal to noise, and it was not interfered and free of self-absorption effect. In this work, the intensities of three atomic lines of Mn around 403 nm were weaker than those of the ionic lines, probably due to self-absorption, as shown in Fig. 4c. The line with the highest intensity of Mn II at 257.61 nm was thus selected as the analytical line for quantitative analysis.

Fig. 4.

The typical LIBS spectra of Mn in different aqueous solutions detected with the echelle spectrometer: (a) Mn in MnCl₂ solutions, (b) Mn in mixed solutions of Mn-Cr-Cd-Cu with a ratio of 1:1:1:1, (c) Mn in seawater, and (d) Mn in MnCl₂ solution detected with the Czerny–Turner spectrometer as a comparison.

RESULTS AND DISCUSSION

Manganese Detected with Czerny–Turner Spectrometer. In this section, a Czerny–Turner spectrometer with high throughput was used to enhance detection of Mn in aqueous solutions. For LIBS measurements, it is essential to select a suitable detection window by time-resolved spectroscopy, which can improve the signal-to-background ratio (SBR) of the data and detection sensitivity. In this work, in order to obtain the optimal delay time and width, the signal intensity and SBR for the Mn II 257.61 nm line were observed with delay time ranging from 600 to 4400 ns in Fig. 5. It could be seen that the background emission was strong in the early stages shown in Fig. 5a, and the intensity of spectral line was at high level. However, SBR was at low level shown in Fig. 5b. Notice that the relative standard deviations of the intensity and SBR were estimated to be 7 and 8.5%, respectively. At the delay time of 2000 ns, SBR was at maximum as shown in Fig. 5b. In this work, the optimal delay time was selected at 1200 ns when SBR was half its maximum value and the gate width was set at 2000 ns, as shown in the shadow area of Fig. 5b.

Fig. 5.

Temporal evolution of LIBS emission of Mn in MnCl₂ solution taken with a gate width of 200 ns and the step of 200 ns: (a) spectral lines of Mn, and (b) signal intensity and signal-to-background ratio of the line Mn 257.61 nm.

For quantitative analysis, the calibration curves would be constructed with the optimized experimental parameters. Figure 6 shows the calibration curves of Mn under different concentrations for different laser pulse energies (30, 60, and 90 mJ). The calibration curves are fitted with a linear function

Fig. 6.

Calibration curves of Mn in aqueous solution for different laser pulse energy (30, 60, and 90 mJ).

y = a + s x

(1)

where y is the integrated intensity (or normalized intensity) of the emission line chosen to calculate the concentration of the element, x the concentration of the element, a and s are, respectively, the intercept and the slope of the calibration curve. The fitting parameters are given in Table I. The linear correlation coefficient R² of the calibration curve is also presented. In these curves, each data point corresponded to an average of eight measurements under the same experimental conditions, and the error bar was the standard deviation of these eight measurements. The slope of the fitting line increased as laser pulse energy strengthened, while the difference decreased. In this work, the concentrations of samples used ranged from 10 to 500 mg/L to avoid saturation effect on calibration curves with higher concentrations. In general, nonlinear behavior of calibration curves was observed under the concentration of above 500 mg/L. It was why we established the calibration curves for the concentration of below 500 mg/L in order to have a more precise determination of the slope for low concentrations in the ppm range, shown in Fig. 6.

TABLE I.

Fitting parameters used for the calibration curves presented in Fig. 5.

Energy (mJ)	Slope s	Intercept a	R ²
90	157.23	5890.09	0.992
60	129.36	5977.90	0.994
30	89.18	1833.21	0.995

The LOD could be determined by the definition

LOD = 3 σ / S

(2)

where σ is the standard deviation of background which represents the fluctuation of the detected signal at the wavelengths of the line emissions when elements to be detected are totally absent, and s is the slope of the calibration curve. In this work, σ was determined from ten measurements of background signals under the same experimental conditions in which the paper substrate absorbed pure deionized water. In this work, LODs of other common heavy metal elements in aqueous solution were also calculated with different laser pulse energies and listed in Table II. Limits of detection for Mn, Cu, Cd, Cr, Zn, and Pb obtained by this approach were one to two orders lower than those obtained by direct detection of liquid samples, as when laser was focused on a static liquid surface²⁵ or liquid jets,²⁶ generally better than those obtained by similar approaches. However, the LODs of certain elements were higher than those obtained by other approaches due to different experimental conditions, such as the ablation under inert gas circumstance.^17,20

TABLE II.

Limits of detection of Mn and several other heavy metal elements in aqueous solutions with different laser pulse energies.

		LOD (this work)/mg/L

Elements	λ (nm)	30 mJ	60 mJ	90 mJ	LOD (other works)/mg/L^a
Mn	257.61	0.27	0.18	0.13	4¹⁴; 0.036²¹; 6²²; 6²⁶
Cu	324.7	0.72	0.54	0.39	4¹⁴; 0.01²⁰; 0.029²¹; 7²⁵; 2.4²⁶
Cd	226.5	0.84	0.6	0.4	7¹⁴; 129¹⁸; 0.59²¹; 10²⁶
Cr	360.53	0.96	0.71	0.52	29¹⁴; 0.018¹⁷; 1.2¹⁸; 0.034²¹; 10²⁵; 43²⁶
Zn	202.55	0.78	0.62	0.51	5¹⁴; 21¹⁸; 1²⁰; 120²⁵; 11²⁶
Pb	405.78	3.87*	2.34	1.27	18¹⁴; 0.075¹⁷; 20¹⁸; 10²⁰; 0.074²¹; 100²⁵

Engine oil samples in paper substrate,¹⁴ filter paper substrate under inert gas circumstance,¹⁷ Ca(OH)₂ substrate,¹⁸ carbon substrate under inert gas circumstance,²⁰ wood slice substrate detected by photomultiplier tube,²¹ metallic surface,²² liquid surface,²⁵ liquid jets.²⁶

The Pb result has been published.²³

Manganese Detected with Echelle Spectrometer. In order to apply this method to rapid onsite detection, it was necessary to construct a compact detection system to obtain reliable data with high sensitivity. In this work, a compact echelle spectrometer equipped with an ICCD was used. In this section, Mn in pure solution was first used to be detected for much better results with this compact detection system. In order to assess the feasibility for rapid onsite detection, the samples of Mn in simulated industrial wastewater and seawater were used.

Manganese in Pure Solution. Figure 4a showed the typical spectra of Mn in pure solutions at a concentration of 100 mg/L with echelle spectrometer. The detection delay and gate width were used at 1200 and 4000 ns, respectively. A suitable laser pulse energy of 60 mJ was used in the following detection with echelle spectrometer. The advantage of echelle spectrometer was that the panoramic spectra allowed an easy access to a suitable internal reference line. In this work, the strong and well-resolved H_α line, shown in Figs. 4a–4c, was used to normalize the intensity of the Mn line by a simple division between both intensities, which had been verified in Charfi and Harith.²⁷

Figure 7 showed the calibration curves of Mn in pure solution detected by echelle spectrometer with and without normalization. The figure illustrated that the improvement was obtained as a result of the normalization procedure on the reliability and reproducibility of the plotted data in this figure. The correlation coefficient R² was improved from 0.984 to 0.996, and the relative standard deviation was decreased from about 10 to 5.2%. The detection precision was thus improved greatly with the normalization of the H_α line. The corresponding LOD of Mn was 2.33 mg/L by Eq. 2.

Fig. 7.

Calibration curves of Mn in pure solutions with and without normalization.

Manganese in pure aqueous solutions was detected with the Czerny–Turner spectrometer and the echelle spectrometer. These results demonstrated that a Czerny–Turner spectrometer detected with higher sensitivity, and its LOD of Mn was down to several hundred ppb level (lg/L). It provided a more sensitive detection of Mn and other heavy metals in laboratory with the help of the Czerny–Turner spectrometer. However, the Czerny–Turner spectrometer usually had a longer focal distance (larger size) so that it was limited to onsite application. By contrast, the echelle spectrometer could be applied onsite due to its compact size and wide spectral detection range that could partly eliminate signal fluctuation to improve precision by normalization, as shown in Fig. 7. For field samples which usually contained many different metal elements, many factors such as the interference of different elements and the instability of LIBS system might affect signal stability, causing imprecise detection. In the following section, Mn in mixed solutions would be investigated to assess the feasibility of onsite detection by this approach.

Manganese in Mixed Solutions. In order to simulate field samples of industrial wastewater, common heavy metal elements (Cr, Cd, and Cu) were added into the pure solution of Mn with a ratio of 1:1:1:1. Figure 4b showed the typical LIBS spectra of mixed solutions (Mn-Cr-Cd-Cu) that contained almost all the emission lines of elements under study. The detection delay was at 1500 ns, and the gate width was 4000 ns.

Figure 8 showed the calibration curve for Mn in mixed solutions with different concentrations. The correlation coefficient R² of the calibration curve was 0.998 which showed good linearity. The LOD of Mn in mixed solutions was 2.87 mg/L according to Eq. 2. It showed that the matrix effect increased LOD by about 23% for Mn compared with the result obtained in pure solution. In this work, other elements (Cd, Cr, and Cu) in pure and mixed solutions were also detected, and all LODs are listed in Table III. The LODs obtained in mixed solutions were affected slightly by the matrix effect compared with those obtained in pure solutions. It was noticeable that no sensitive spectral lines for Zn and Pb were observed in the wavelength range of 220–850 nm with the echelle spectrometer, and the preliminary results suggested that the limits of detection of Zn and Pb were higher than 100 mg/L.

TABLE III.

Limits of detection of Mn, Cr, Cd, and Cu in pure solutions and mixed solutions using the echelle spectrometer.

	Limits of detection (mg/L)

Elements	Pure	Mixed
Mn	2.33	2.87
Cd	5.32	6.56
Cr	4.74	3.98
Cu	1.67	2.13
Zn	–	–
Pb	–	–

Fig. 8.

Calibration curve of Mn in mixed solutions (Mn- Cr-Cd-Cu) with normalization.

To assess the accuracy of this approach, Mn with known concentration in mixed solutions (100 mg/L) had been determined according to the experimental calibration curves obtained in the mixed aqueous solutions of Mn, Cr, Cd, and Cu. The concentrations of other elements (Cd, Cr, and Cu) were different with those used as standard samples for quantitative analysis, with 150, 150, and 100 mg/L, respectively. The LIBS result was 94 ± 5 mg/L under the average of seven measurements, and the relative standard deviation was 4.2% by calculation. Good agreement was obtained between the measured and known values, taking into consideration that the matrix effect was not exactly verified. It indicated that our method was suitable for rapid onsite detection with accurate results.

Manganese in Seawater. Seawater from 1 km in depth was used as a solvent to prepare the standard samples of Mn for quantitative analysis. As rich elements in the seawater (e.g. Na, Ca, Mg, and other elements) would influence the LIBS signal of Mn due to the inevitable matrix effect, the calibration curve obtained in mix solution was not suitable for seawater. We needed to plot appropriate calibration curves of Mn in seawater for quantification.

Figure 4c showed the LIBS spectra of Mn in seawater with the detection delay of 1500 ns and gate width of 4000 ns. Common elements in seawater (Ca, Na, and Mg) were observed obviously. Figure 9 showed the calibration curves of Mn in seawater. The LOD of Mn in seawater was 2.73 mg/L according to Eq. 2. From Figs. 7 and 8, we could see that the fitting calibration curves of Mn in mixed solution (Mn-Cr-Cd-Cu) and seawater were not parallel, which proved the existence of the matrix effect in these solutions.

Fig. 9.

Calibration curve for Mn in seawater.

CONCLUSION

A paper substrate was used as a liquid absorber for quantitative analysis of Mn in aqueous solution and seawater by LIBS. The LOD of Mn in aqueous solution exhibited several hundred ppb level of 0.13 mg/L using the Czerny–Turner spectrometer. In order to apply it to onsite detection, the compact echelle spectrometer was used for two field samples of Mn solution simulating industrial wastewater and seawater. The results showed that the measured concentration of Mn in industrial wastewater had a good agreement to the real value even when the concentrations of other elements in the sample changed. However, due to the matrix effect in both solutions, relevant calibration curves of Mn were plotted and the LODs of two solutions were thus calculated to be 2.87 and 2.73 mg/L, respectively. Although our approach provided LODs no lower than those obtained with ICP-MS and AAS, its simplicity in sample preparation and its rapid detection capability make it especially suitable for application in complicated and different environmental conditions such as hazardous industrial wastewater and seawater. All results obtained suggest that this approach has great potential for the rapid onsite detection of Mn and other heavy metal elements in industrial wastewater and seawater, which would play an important role in pollution monitoring and the investigation of geochemical and biological processes in the ocean.

Footnotes

ACKNOWLEDGMENTS

Financial supports from National Natural Science Foundation of China (Grant Nos. 11104153 and 41376107) are highly acknowledged. One of authors (Junshan Xiu) would like to thank Zhen Shi for his language checking.

References

Yeats

P.A.

Dalziel

J.A.

Moran

S.B.

. “A Comparison of Dissolved and Particulate Mn and Al Distributions in the Western North Atlantic”. Oceanol. Acta. 1992. 15(6): 609–619.

“Wastewater Discharge Standards in Latin America”. http://en.wikipedia.org/wiki/Wastewater_discharge_standards_in_Latin_America [accessed Mar 29 2014].

Yeats

P.A.

Westerlund

Flegal

A.S.

. “Cadmium, Copper and Nickel Distributions at Four Stations in the Eastern Central and South Atlantic”. Mar. Chem. 1995. 49(4): 283–293.

Cabon

J.Y.

. “Improvement of Direct Determination of Cu and Mn in Seawater by GFAAS and Total Elimination of the Saline Matrix with the Use of Hydrofluoric Acid”. Talanta. 2005. 65(2): 402–407.

Milne

Landing

Bizimis

Morton

. “Determination of Mn, Fe, Co, Ni, Cu, Zn, Cd and Pb in Seawater Using Resolution Magnetic Sector Inductively Coupled Mass Spectrometry (HR-ICP-MS)”. Anal. Chim. Acta. 2010. 665(2): 200–207.

Pasquini

Cortez

Silva

Gonzaga

F.B.

. “Laser Induced Breakdown Spectroscopy”. J. Braz. Chem. Soc. 2007. 18(3): 463–512.

De Giacomo

Dell'Aglio

De Pascale

Capitelli

. “From Single Pulse to Double Pulse ns-Laser Induced Breakdown Spectroscopy Under Water: Elemental Analysis of Aqueous Solutions and Submerged Solid Samples”. Spectrochim. Acta, Part B. 2007. 62(8): 721–738.

Fichet

Mauchien

Wagner

J.F.

Moulin

. “Quantitative Elemental Determination in Water and Oil by Laser Induced Breakdown Spectroscopy”. Anal. Chim. Acta. 2001. 429(2): 269–278.

Burakov

V.S.

Tarasenko

N.V.

Nedelko

M.I.

Kononovb

V.N.

Vasilevb

N.N.

Isakov

S.N.

. “Analysis of Lead and Sulfur in Environmental Samples by Double Pulse Laser Induced Breakdown Spectroscopy”. Spectrochim. Acta, Part B. 2009. 64(2): 141–146.

10.

Boué-Bigne

. “Laser Induced Breakdown Spectroscopy Applications in the Steel Industry: Rapid Analysis of Segregation and Decarburization”. Spectrochim. Acta, Part B. 2008. 63(10): 1122–1129.

11.

Kaiser

Galiová

Novotný

Červenka

Reale

Novotný

Liška

Samek

Kanický

Hrdlička

Stejskal

Adam

Kizek

. “Mapping of Lead, Magnesium and Copper Accumulation in Plant Tissues by Laser Induced Breakdown Spectroscopy and Laser Ablation Inductively Coupled Plasma Mass Spectrometry”. Spectrochim. Acta, Part B. 2009. 64(1): 67–73.

12.

Yueh

F.Y.

Sharma

R.C.

Singh

J.P.

Zhang

H.S.

Spencer

W.A.

. “Evaluation of the Potential of Laser-Induced Breakdown Spectroscopy for Detection of Trace Element in Liquid”. J. Air Waste Manage. 2002. 52(11): 1307–1315.

13.

Cáceres1

J.O.

Tornero López

Telle

H.H.

González Ureńa

. “Quantitative Analysis of Trace Metal Ions in Ice Using Laser-Induced Breakdown Spectroscopy”. Spectrochim. Acta, Part B. 2001. 56(6): 831–838.

14.

Yaroshchyk

Morrison

R.J.S.

Body

Chadwick

B.L.

. “Quantitative Determination of Wear Metals in Engine Oils Using LIBS: The Use of Paper Substrates and a Comparison Between Single- and Double-Pulse LIBS”. Spectrochim. Acta, Part B. 2005. 60(11): 1482–1485.

15.

Alamelu

Sarkar

Aggarwal

S.K.

. “Laser-Induced Breakdown Spectroscopy for Simultaneous Determination of Sm, Eu and Gd in Aqueous Solution”. Talanta. 2008. 77(1): 256–261.

16.

Zhu

D.H.

L.Z.

Wang

Chen

J.P.

X.W.

. “Determination of Ca and Mg in Aqueous Solution by Laser-Induced Breakdown Spectroscopy Using Absorbent Paper Substrates”. Appl. Opt. 2011. 50(29): 5695–5699.

17.

Lee

S.W.

Han

S.H.

. “Laser-Induced Breakdown Spectroscopy (LIBS) of Heavy Metal Ions at the Sub-Parts per Million Level in Water”. Appl. Spectrosc. 2012. 66(12): 1385–1396.

18.

Díaz Pace

D.M.

D'Angelo

C.A.

Bertuccelli

. “Analysis of Heavy Metals in Liquids Using Laser Induced Breakdown Spectroscopy by Liquid-to-Solid Matrix Conversion”. Spectrochim. Acta, Part B. 2006. 61(8): 929–933.

19.

Dockey

C.R.

Pender

J.E.

Goode

S.R.

. “Speciation of Chromium via Laser-Induced Breakdown Spectroscopy of Ion Exchange Polymer Membranes”. Appl. Spectrosc. 2005. 59(2): 252–257.

20.

Vander Wal

R.L.

Ticich

T.M.

West

J.R.

Householder

J.R.P.A.

. “Trace Metal Detection by Laser-Induced Breakdown Spectroscopy”. Appl. Spectrosc. 1999. 53(10): 1226–1236.

21.

Chen

Z.J.

H.K.

Liu

R.H.

. “Fast and Sensitive Trace Metal Analysis in Aqueous Solutions by Laser-Induced Breakdown Spectroscopy Using Wood Slice Substrates”. Spectrochim. Acta, Part B. 2008. 63(1): 64–68.

22.

Aguirre

M.A.

Legnaioli

Almodóvar

Hidalgo

Palleschi

Canals

. “Elemental Analysis by Surface-Enhanced Laser-Induced Breakdown Spectroscopy Combined with Liquid-Liquid Microextraction”. Spectrochim. Acta, Part B. 2013. 79-80: 88–93.

23.

Xiu

J.S.

Hou

H.M.

Zhong

S.L.

Wang

Z.N.

Zheng

R.E.

. “Quantitative Determination of Heavy Metal Element Pb in Aqueous Solutions by Laser-Induced Breakdown Spectroscopy Using Paper Slice Substrates”. Chin. J. Lasers. 2011. 38(8): 0815003.

24.

“NIST Atomic Spectra Database Lines Form”. http://physics.nist.gov/PhysRefData/ASD/lines_form.html [accessed Mar 12 2014].

25.