Heavy metal pollution from industrial wastewater is an important source. A method for heavy metals determination in industrial wastewater based on laser-induced breakdown spectroscopy (LIBS) technique was studied and the on-line monitoring system that used automatic graphite enrichment and spatial plasma confinement detection was developed and field demonstrated. The limits of detection (LOD) of heavy metal elements (Cd, Cr, Cu, Ni, Pb, Zn) could reach several μg/L. In Tongling, the on-line heavy metal monitor was field demonstrated. The calibration curves of copper and zinc were built on site, and then on-line monitoring was conducted. The measurement results of this monitor were compared with ICP-OES and had a good correlation. The results showed that the heavy metal monitor could be used for on-line detection of heavy metals in wastewater and had a good reliability.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Heavy metals in water will influence the environment in many ways, it’s also a great threat to human health. There are many sources of heavy metals in water, among them, the industrial wastewater plays an important role . Compared to other sources such as domestic sewage and garbage penetration, the industrial wastewater is more polluting and much easier to monitor. Understanding of the emission status timely and control of pollution sources are the keys to current governance. So it has great significance to conduct on-line monitoring of heavy metals from industrial discharges. Different industries emit different heavy metals. Petrochemicals produce large amount of nickel, chromium and vanadium , while electroplating wastewater contains a lot of copper, zinc and cadmium . Technology and equipment which can monitor a variety of heavy metals simultaneously are required urgently.
At present, the most commonly used methods for online monitoring of heavy metals in water include colorimetry and anodic stripping voltammetry [4,5]. Colorimetry is usually used for specific heavy metal elements in water, which faces great difficult in multiple elements detection. Anodic stripping voltammetry has achieved industrialization and has been widely used in industrial wastewater monitoring. But it may use electrodes containing mercury which could produce secondary pollution, and the heavy metals will interfere with each other when multiple element detection is needed . It is still necessary to develop a fast technology which can achieve multi-elements simultaneous on-line measurement.
Laser induced breakdown spectroscopy(LIBS) has been widely used to detect element content in various environmental medium [7,8]. It is a potential tool to perform on-line detection of heavy metals in water . In the past three decades, LIBS has made great progress. The detection field of LIBS has successfully covered soil, rock, coal, and steel samples [10–13]. The most significant advantages of LIBS is rapid detection speed, no need of sample preparation and simultaneous detection of multiple elements . But it still faces many challenges in quantitative analysis of heavy metals in liquid samples, including splashing, short plasma life-time and weak spectrum. Many methods have been developed to solve these problems, such as water jets and matrix conversion of water to solid, which is widely researched and used [15,16]. Zhijiang Chen used wood slice to absorb the heavy metals in liquid samples and then analyzed the sample with LIBS . Dockery used ion exchange polymer membranes to enrich chromium from solutions . Lianbo Guo and Xiao Cheng used LIBS for on-stream analysis of slurries [19,20]. However, some of the matrix conversion methods needed longer sample pretreatment time which reduced the advantage of LIBS, and most of them were unable to process on-line. So, more new enrichment method which is fast and easy to achieve automatic control are needed.
In the last few years, LIBS equipment for metal sorting, ore detection and production monitoring kept appearing [21,22], but LIBS instruments for heavy metal detection in water have not been developed and reported. In this paper graphite enrichment LIBS was introduced. Graphite flakes with round groove were used to enrich heavy metals in liquid samples. Hemispherical confinement device of laser plasma is designed for further enhance the spectral emission intensity and duration of laser plasma. The LOD of different elements (e.g. Pb, Ni, Cd, Cr, Zn and Cu) was reduced with the increase of liquid quantity and could be reached several ppb. The automatic enrichment method suitable for continuous online work based on graphite enrichment was also studied. The heavy metal on-line monitor for heavy metals in industrial wastewater was designed and outfield demonstrated in Tongling, Anhui province. The research results provided an important application supports for on-line monitoring of heavy metals in industrial wastewater.
2.1. The LIBS setup
The LIBS setup was shown in Fig. 1. A Nd:YAG pulsed laser(Quantel, bigsky) operated at 1064nm wavelength was used as the excitation source. The pulse energy of the laser was 50mJ, the frequency is 2Hz and the pulse width is 5ns. A fiber optic spectrometer (AVS-DESKTOP-USB2, Avantes, Netherlands) is chosen to record the spectra. The spectrometer uses a CCD with the resolution of 512*2048 as the detector and contains 3 channels in order to cover the spectral range of 200-500nm, the spectral resolution is 0.8nm. The spectrometer has the delay time function. The delay time used in experiments is 1.5μs. The laser is reflected by a total mirror and focused by a convex lens to the graphite flakes which have been enriched with heavy metals.
As Fig. 1 shown, the sample preparation module can be divided into 3 functional modules, including sample adding module, heating module and graphite flakes switch module. By sample adding module, the wastewater is injected into the graphite flake accurately. The graphite flake is round and has a diameter of 30mm. There is a annular groove in the flake which is used to contain wastewater. The single sample addition amount is 0.5mL. After a certain time, the rotating platform rotate a certain angle and the graphite flake is transported to heating module. The heating temperature is set to 95°C in order to avoid liquid boiling and sample loss. After all the liquid has evaporated, the salt in the wastewater is evenly distributed in the lower surface of the groove. In this way, the sample matrix was transferred from liquid to solid successfully. When the temperature sensor judges that the wastewater has been completely evaporated, the graphite flake rotates a certain angle again to the laser excitation module. After detected by LIBS, the graphite flakes rotates to the graphite flakes switch module and unloads, a detection process is ended.
The spectral data acquisition module is shown in Fig. 2. Spatial confinement of the plasma can significantly increase the intensity and stability of the spectrum [23,24]. So a reflector, a focusing lens and a hemispherical cavity are integrated together. The laser pulse is focused by a lens with 50mm focal length and upright incidents to the graphite flake. Plasma is generated in the hemispherical cavity which has a diameter of 10mm. The whole module is compact and stable in structure, allowing high spectral intensity and stability to be obtained.
The heavy metal on-line monitor for heavy metals in industrial wastewater was designed as shown in Fig. 3. The left part of the instrument was sample adding module, heating module, graphite flakes switch module and spectral data acquisition module. The right part included the laser power supply, spectrometer and control system. The wastewater could be extracted by a peristaltic pump from sewage outlet. The length, width and height of the heavy metal monitor was 70, 40 and 30 cm respectively.
2.2. Sample preparation
The calibration solution was prepared by mixing specific amount of nitrate of copper, lead, chromium, zinc, cadmium and nickel with known amount of deionized water to obtain samples with different heavy metal concentrations. All reagents used were analytical grade. These calibration solution was used to set up the calibration curves of heavy metals and calibrate the heavy metal monitor. In the graphite flake, carbon element accounted for 99.99%. In addition to carbon, a small quantity of calcium, magnesium and sodium were distributed uniformly.
The calibration solution or wastewater was injected steadily in the groove of the graphite flake. The injection process usually lasted one minute, in order to guarantee the stability between samples. The liquid distributed on the bottom of the graphite flake uniformly. After a few minutes of heating(less than 5 minutes), solute precipitated on the bottom uniformly. After the above steps, the liquid sample transferred to solid state. When on-site monitoring of industrial wastewater was carried on, the monitor extracted samples directly from the wastewater outlet using plastic tubes and peristaltic pumps. The whole sample preparation time was usually less than 6 minutes, with the addition time of spectra detection, concentration calculation, the entire detection process for one liquid sample could be controlled within 10 minutes, which was close to anodic stripping voltammetry. The monitoring frequency could be up to 6 times one hour, which could meet the needs of on-line quantitative analysis of heavy metals in industrial wastewater.
3. Results and discussion
3.1. Detection ability of on-line heavy metal monitor
One drawback of LIBS is the high limit of detection(LOD), for heavy metals in soil, the LODs were usually more than 10 mg/kg. For different enrichment method of heavy metals in water, the LODs varied from tens parts per billion (ppb) to several parts per million(ppm). In China, the sewage discharge standards specified the highest contents of heavy metal elements in industrial wastewater, which were below 1 ppm. In fact, after treatment, the contents of heavy metals in industrial wastewater were tens parts per billion generally. This required that the detection sensitivity of the on-line heavy metal monitor should high enough, in other words, the LODs of the monitor should be tens μg/L.
The LODs of copper, chromium, cadmium, lead, nickel, zinc were tested and calculated in laboratory. The diameter of one graphite flake was 3cm. In one detection process, 80 laser shots would ablate in different sites of the flake. After singular value removing, all remaining spectrum would remove background and average. The characteristic lines of copper, chromium, cadmium, lead, nickel, zinc were 324.75, 425.44, 214.44, 405.78, 352.45 and 202.54nm respectively, and listed in Table. 1. Calibration curves of these elements were built by intensity of these characteristic lines and corresponding contents.
Taking copper and zinc as examples, calibration curves of copper and zinc were shown in Fig. 4. From the figure above, the intensity of copper spectral line was much stronger than that of zinc. When concentration of zinc was below, the spectral line stability is poor. The LODs of copper and zinc had great difference.
Their LODs were calculated by the known relation,
In Eq. (1), S was the slop of the calibration curve, Sb the standard deviation of the background, and the statistical factor k could be equal to 3. By Eq. (1), the LODs of copper, zinc were calculated the results were 4.8 and 36.1μg/L. As reflected in Fig. 4, the LOD of zinc was nearly 8 times copper. Through this method, the LODs of chromium, cadmium, nickel and lead were also calculated, which were 12.2, 3.1, 34.3 and 32.6μg/L. The results showed that this heavy metal monitor had a good detection ability for these heavy metals we cared about.
3.2. Field operation and calibration
An on-site campaign of LIBS measurement was performed in Tongling, which is in the south of Yangtze river. Tongling was a small city dominated by mineral smelting. There were many industrial and mining enterprises which generated large amount of sewage with lots of heavy metals during the production process. All related enterprises had supporting sewage treatment equipment and corresponding heavy metal monitoring equipment.
The heavy metal monitor based on LIBS was placed nearby a wastewater outlet of a copper smelter. So the key monitoring element was copper. A plastic tube was inserted into the wastewater, and the other end of the tube was directly connected to the heavy metal monitor.
A typical spectrum of wastewater was shown in Fig. 5. From the figure, trace amounts of zinc, strontium could be detected. Large amount of aluminum, calcium, and magnesium could also be detected. It was difficult to judge the concentrations of elements in wastewater only through the intensity of the characteristic spectral lines. To obtain the content of the elements which we concerned in the wastewater, 500 milliliter wastewater sample was collected. Part of the sample was measured by inductively coupled plasma mass spectrometry (ICP-MS). The results of measurement were taken as the true value of heavy metal concentrations in wastewater. The other part of the sample was used to calibrate the monitor and build proper calibration curves. As a result, the concentrations of copper and zinc were 16.6 and 18.7ppb, and the concentrations of other heavy metals were all below 5ppb, which were significantly below the LODs of the monitor. In the subsequent continuous measurement process, copper and zinc were taken as the measurement target.
The wastewater after treatment contained large amount of sodium, magnesium, and calcium salt, which was significantly different with the standard solutions prepared in laboratory. The calibration curves built in laboratory were not suitable for on-site detecting. The first thing needed to do before field experiment was built proper calibration curves. 3 kinds of calibration methods were proposed, including multiple enrichment, multiple dilutions and adding extra copper salt.
For the first calibration method, multiple enrichment, it was obvious that multiple enrichment of wastewater on the same graphite flakes could get samples with different concentrations of copper and zinc. But in the actual experiment, the sample with higher copper concentration did not have stronger spectrum, contrary to expectations, the sample had much weaker spectrum than that of samples with lower copper concentration. After observing, there was a layer of white salt crystals laid flat on the graphite. When laser shot on the graphite, the salt crystals absorbed the laser energy and melt. So less laser energy was used to ablate the sample and generate plasma. It was final confirmed that the multiple enrichment was not suitable for monitor calibration.
For the second calibration method, multiple dilutions, it was contrary to the first method. Samples with different copper concentrations were obtained by different dilution ratios. The calibration results also showed that the light metal salt in wastewater had a great influence on spectrum of heavy metals. After dilution, the concentrations of copper in samples did not have a linear relationship with their intensity of spectral lines.
For the third calibration method, adding extra copper salt, the operation was more complicated than the first two methods. By adding different qualities of copper nitrate and zinc nitrate, different samples with gradient concentrations of copper and zinc were obtained. The calibration curve of copper was built, as Fig. 6 shown.
When conducting the field experiment, 4 gradient samples were prepared. The initial sample was the wastewater collected directly, the gradient concentration was 25μg/L. Based on standard addition method, the initial concentration of copper could be obtained, which was 16.6μg/L. From the above figure, the linearity of the curve was acceptable, but the standard deviation of spectral line intensity was huge which would cause large deviation of detection results. The relative standard deviation (RSD) of spectral line intensity was more than 25%.
In order to improve the stability of the spectrum, one kind of internal calibration method was taken. Because the content of the elements in the wastewater was uncertain. It was uncertain to use a single element as the internal standard. So the sum of the spectral intensities was taken as the internal standard. In the wastewater, there are large amount of sodium, magnesium, and calcium. So the sum intensity of spectral lines of Na (330.23nm), Mg (279.55, 280.27,285.21 and 383.83nm), Al (308.22, 309.27, 394.40 and 396.15nm), Ca (422.67, 442.54, 443.5 and 445.5nm) and C (247.85nm) is calculated as the internal standard. The stability of the sum of the spectral intensities was much better than that of one single element. The RSD of spectral sum was reached 10.24%, which was much lower than one single element. In field experiment, the experimental conditions changed varied, it was hard to guarantee the stability of the measurement spectrum. To some content, the poor spectral stability limited the development of LIBS technology. After internal calibration, the calibration curve of copper was shown in Fig. 7.
In the above figure, the x axis was the concentration of copper, the y axis was the value which was obtained through intensity of copper divided by the spectral sum. In order to make the value easier to identify, the value was multiplied by ten thousand. Obviously, the internal calibration by spectral sum significantly improved the reliability of the calibration curve. Although it was still unable to compare with ICP-MS and AAS, it was effective for excessive warning detection of heavy metals in wastewater.
Besides copper, the calibration curve of zinc was also built, which was shown in Fig. 8. For this monitor, the detection ability for copper was much stronger than zinc. The calculated LOD of zinc was 36.1 μg/L which was much higher than copper. But when the concentration of zinc was lower than 36.1 μg/L, and higher than 20μg/L, the spectral line of zinc was still clear to be seen. There were several reasons for this situation. First, in the LOD computational formula, the confidence coefficient could be 2, so the LOD of zinc could reduce to 24μg/L. Second, the spectral line of zinc was in the deep ultraviolet region and without interference. But the absolute spectral intensity of zinc was low and had a poorer stability when the concentration of zinc was below.
3.3. Data analysis of continuous operation
After the monitor was calibrated, a continuous operation lasted 5 days was conducted. During the experiment, the monitor detected one wastewater sample every 15 minutes in daytime which had a higher frequency than anodic stripping voltammetry. In the evening, one sample was detected unattended every other hour. The whole continuous operation was conducted without any human interference. In order to verify the accuracy of the measurement results, one wastewater sample was collected every half hour, and synchronized with monitor detection. After the field experiment, all samples collected on site were detected by ICP-OES, and the results were taken as the true content in the wastewater sample. There was another prerequisite for continuous operation. The matrix of wastewater should keep stead in order to make the calibration curve built in the previous section suitable for the following continuous test. Figure 9 showed the detection results of copper in the first day, the result detected by ICP-OES was also shown in the figure. The results measured by the monitor had good consistency with ICP-OES. In a single day, the concentration of copper constantly changed. According to the introduction of a factory staff, during the entire wastewater process, the raw materials for copper precipitation scheduled delivery. At the beginning of the raw materials entering the reaction pool, the copper content in the produced wastewater was significantly lower.
Figure 9 clearly showed that the concentration of copper in wastewater was very low after 12 o’clock, and stayed stable throughout the night. The detection results between ICP-OES and LIBS had a very good correlation. For those samples with a higher concentration of copper, the measurement errors could be lower than 10%, but when the copper concentration was near the LOD, the measurement error could reach 20% and the stability significantly decresed.
The continuous monitoring results of zinc was also obtained which was shown in Fig. 10. Unlike copper, the concentration of zinc in all wastewater samples were lower than the LOD, it is difficult to accurately quantify zinc in wastewater. In the continuous monitoring experiment, the average measurement error for zinc was 23.2%, which could meet the needs of semi-quantitative measurement of zinc.
From Fig. 9 and Fig. 10 we could see that, in the daytime the concentrations of zinc and copper were usually lower. Because in China power fare was much lower at night than in daytime. The enterprise would produce more products at night, so the concentrations of copper and zinc in wastewater changed at night and day. 12 o’clock was the lunch time, all workers in the enterprise would have a short time for rest, which resulted in the lowest copper content in wastewater.
The continuous monitoring results clearly showed the change of copper and zinc in wastewater and had good correlation with ICP-OES, in addition, the results also reflected the factory production status.
Currently the most widely used heavy metal monitor method for wastewater are spectrophotometry and anodic stripping voltammetry. The LOD of anodic stripping voltammetry is generally less than 1μg/L . The detection time of anodic stripping voltammetry in laboratory is less than 1 minute. the LIBS device performance is still weaker than other techniques. But it has the advantage of simultaneous detection of multiple elements, it can be used as a complement to other methods. On the other hand, the LIBS device does not consume reagents during operation, it is much more economical and environmentally friendly.
In the present work, an automatic monitor based on LIBS technology for heavy metals in wastewater was developed. The graphite enrichment method was studied to solve the problems of water splashing, shorter life of plasma and weaker spectrum signal. A hemispherical confinement device of laser plasma was also designed for further enhance the spectral emission intensity and duration of laser plasma. The experimental results showed that the stability and repeatability of the measurement spectrum were improved, the LOD of different elements (e.g. Pb, Ni, Cd, Cr, Zn and Cu) was reduced and could reach several ppb. An automatic enrichment method suitable for continuous online work based on graphite substrate was also studied, and corresponding automatic enrichment device was designed. An on-line monitor of heavy metals determined for industrial wastewater with automatic sample enrichment and fast spectral measurement was designed, and realized on-line automatic measurement of heavy metals in industrial wastewater. A demonstration operation was carried out in a copper smelting enterprise in Tongling, China.The continuous results showed that the on-line heavy metal monitor had good correlation with ICP-OES for those wastewater samples with high heavy metal content. Compared with anodic stripping voltammetry, the online monitor still had some shortcomings, but it already could be used for excessive warning of heavy metals.
National Natural Science Foundation of China (NSFC) (21735005); National Research and Development Program of China (2016YFD0800902-2); Science and Technology Major Project of Anhui Province (16030801117).
The authors are also grateful for the technology and platform support from the key laboratory of optics for environmental monitoring technology of Anhui, the key laboratory of optics for environmental monitoring technology of ministry of environment protection of China.
1. S. Khan, Q. Cao, Y. M. Zheng, Y. Z. Huang, and Y. G. Zhu, “Health risks of heavy metals in contaminated soils and food crops irrigated with wastewater in Beijing, China,” Environ. Pollut. 152(3), 686–692 (2008). [CrossRef] [PubMed]
3. N. Adhoum, L. Monser, N. Bellakhal, and J. E. Belgaied, “Treatment of electroplating wastewater containing Cu2+, Zn2+ and Cr(VI) by electrocoagulation,” J. Hazard. Mater. 112(3), 207–213 (2004). [CrossRef] [PubMed]
4. Y. L. Hung, T. M. Hsiung, Y. Y. Chen, Y. F. Huang, and C. C. Huang, “Colorimetric Detection of Heavy Metal Ions Using Label-Free Gold Nanoparticles and Alkanethiols,” J. Phys. Chem. 114(39), 16329–16334 (2010).
5. K. Zinoubi, H. Majdoub, H. Barhoumi, S. Boufi, and N. Jaffrezic-Renault, “Determination of trace heavy metal ions by anodic stripping voltammetry using nanofibrillated cellulose modified electrode,” J. Electroanal. Chem. (Lausanne Switz.) 799, 70–77 (2017). [CrossRef]
6. V. K. Unnikrishnan, R. Nayak, K. Aithal, V. B. Kartha, C. Santhosh, G. P. Gupta, and B. M. Suri, “Analysis of trace elements in complex matrices (soil) by Laser Induced Breakdown Spectroscopy (LIBS),” Anal. Methods 5(5), 1294–1300 (2013). [CrossRef]
7. G. Niu, Q. Shi, M. Xu, H. Lai, Q. Lin, K. Liu, and Y. Duan, “Dehydrated carbon coupled with laser-induced breakdown spectrometry (LIBS) for the determination of heavy metals in solutions,” Appl. Spectrosc. 69(10), 1190–1198 (2015). [CrossRef] [PubMed]
9. S. A. Davari, S. Hu, and D. Mukherjee, “Calibration-free quantitative analysis of elemental ratios in intermetallic nanoalloys and nanocomposites using Laser Induced Breakdown Spectroscopy (LIBS),” Talanta 164, 330–340 (2017). [CrossRef] [PubMed]
10. D. Meng, N. Zhao, Y. Wang, M. Ma, L. Fang, Y. Gu, Y. Jia, and J. Liu, “On-line/on-site analysis of heavy metals in water and soils by laser induced breakdown spectroscopy,” Spectrochim. Acta B At. Spectrosc. 137, 39–45 (2017). [CrossRef]
11. I. Rehan, M. Z. Khan, K. Rehan, A. Mateen, M. Aamar Farooque, S. Sultana, and Z. Farooq, “Determination of toxic and essential metals in rock and sea salts using pulsed nanosecond laser-induced breakdown spectroscopy,” Appl. Opt. 57(2), 295–301 (2018). [CrossRef] [PubMed]
12. M. Gaft, E. Dvir, H. Modiano, and U. Schone, “Laser induced breakdown spectroscopy machine for online ash analyses in coal,” Spectrochim. Acta B At. Spectrosc. 63(10), 1177–1182 (2008). [CrossRef]
13. G. Hubmer, R. Kitzberger, and K. Mörwald, “Application of LIBS to the in-line process control of liquid high-alloy steel under pressure,” Anal. Bioanal. Chem. 385(2), 219–224 (2006). [CrossRef] [PubMed]
14. L. J. Radziemski, “From LASER to LIBS, the path of technology development,” Spectrochim. Acta B At. Spectrosc. 57(7), 1109–1113 (2002). [CrossRef]
15. M. S. Cheri and S. H. Tavassoli, “Quantitative analysis of toxic metals lead and cadmium in water jet by laser-induced breakdown spectroscopy,” Appl. Opt. 50(9), 1227–1233 (2011). [CrossRef] [PubMed]
16. D. M. Díaz Pace, C. A. D’Angelo, D. Bertuccelli, and G. Bertuccelli, “Analysis of heavy metals in liquids using laser induced breakdown spectroscopy by liquid-to-solid matrix conversion,” Spectrochim. Acta B At. Spectrosc. 61(8), 929–933 (2006). [CrossRef]
17. Z. Chen, H. Li, M. Liu, and R. Li, “Fast and sensitive trace metal analysis in aqueous solutions by laser-induced breakdown spectroscopy using wood slice substrates,” Spectrochim. Acta B At. Spectrosc. 63(1), 64–68 (2008). [CrossRef]
18. C. R. Dockery, J. E. Pender, and S. R. Goode, “Speciation of chromium via laser-induced breakdown spectroscopy of ion exchange polymer membranes,” Appl. Spectrosc. 59(2), 252–257 (2005). [CrossRef] [PubMed]
19. L.-B. Guo, X. Cheng, Y. Tang, S.-S. Tang, Z.-Q. Hao, X.-Y. Li, Y.-F. Lu, and X.-Y. Zeng, “Improvement of spectral intensity and resolution with fiber laser for on-stream slurry analysis in laser-induced breakdown spectroscopy,” Spectrochim. Acta B At. Spectrosc. 152, 38–43 (2019). [CrossRef]
20. X. Cheng, X. Yang, Z. Zhu, L. Guo, X. Li, Y. Lu, and X. Zeng, “On-stream analysis of iron ore slurry using laser-induced breakdown spectroscopy,” Appl. Opt. 56(33), 9144–9149 (2017). [CrossRef] [PubMed]
21. G. S. Senesi, D. Manzini, and O. De Pascale, “Application of a laser-induced breakdown spectroscopy handheld instrument to the diagnostic analysis of stone monuments,” Appl. Geochem. 96, 87–91 (2018). [CrossRef]
22. B. Bousquet, G. Travaillé, A. Ismaël, L. Canioni, E. Michel-Le Pierrès, E. Brasseur, S. Roy, I. le Hecho, M. Larregieu, S. Tellier, M. Potin-Gautier, T. Boriachon, P. Wazen, A. Diard, and S. Belbèze, “Development of a mobile system based on laser-induced breakdown spectroscopy and dedicated to in situ analysis of polluted soils,” Spectrochim. Acta B At. Spectrosc. 63, 1085–1090 (2008). [CrossRef]
23. A. M. Popov, F. Colao, and R. Fantoni, “Spatial confinement of laser induced plasma to enhance LIBS sensitivity for trace elements determination in soils,” J. Anal. At. Spectrom. 25, 837–848 (2010). [CrossRef]
24. L. B. Guo, B. Y. Zhang, X. N. He, C. M. Li, Y. S. Zhou, T. Wu, J. B. Park, X. Y. Zeng, and Y. F. Lu, “Optimally enhanced optical emission in laser-induced breakdown spectroscopy by combining spatial confinement and dual-pulse irradiation,” Opt. Express 20(2), 1436–1443 (2012). [CrossRef] [PubMed]
25. S. M. Rosolina, J. Q. Chambers, C. W. Lee, and Z. L. Xue, “Direct determination of cadmium and lead in pharmaceutical ingredients using anodic stripping voltammetry in aqueous and DMSO/water solutions,” Anal. Chim. Acta 893, 25–33 (2015). [CrossRef] [PubMed]