Open Access
Add to BibSonomy
Abstract
For monitoring the extent of eutrophication in water, phosphorus (P) was detected by laser-induced breakdown spectroscopy (LIBS). A plasma amplification method was proposed and the filtered aerosol was guided to interact with the collinear laser in conjunction with a nebulizer, cyclonic spray chamber, and quartz tube. With this method, the length of the plasma was amplified from 5.27∼8.73 to 17.58 mm. Moreover, the limit of detection (LoD) values of P in water improved from 6.13∼17.75 to 3.60 ppm. Furthermore, the average relative error (REAV) values reduced from 10.23∼23.84 to 6.17%. The root mean square error of cross-validation (RMSECV) values decreased from 16.68∼64.29 to 3.24 ppm. This demonstrated that plasma amplification LIBS could improve the quantitative analysis performance of LIBS detection of trace phosphorus in water.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Excessive concentrations of phosphorus (P) will eutrophicate the water, leading to red tides, water wars, and other phenomena in lakes, rivers, and the coastal ocean [1]. Therefore, the detection of P in water is especially important to monitor the extent of eutrophication [2]. According to the national standard of environmental quality standards for surface water in China (GB 3838-2002), ammonium molybdate spectrophotometry is the basic method for the detection of P in water. However, this method is time-consuming and takes more than 45 minutes (water quality-determination of total phosphorus-ammonium molybdate spectrophotometry, GB 11893-89).
Laser-induced breakdown spectroscopy (LIBS) is a technique for real-time elemental analysis of gaseous [3], solid [4], and liquid samples [5]. In particular, researchers found that LIBS also has potential for detection of P in geology [6], soil [7], food [8], and other fields. C. Pereirs de Morais et al. improved the limit of detection (LoD) of P in river sediment particles from 709 to 349 ppm by double-pulse LIBS [9]. S. Sanchez-Esteva et al. found that the best LoD of P in soil was 10 ppm, which deteriorated with soil sand content [10]. Y. Tian et al. combined LIBS with machine learning algorithms to determination P in seafood, and the LoD was about 370 ppm [11]. W. Sha et al. applied LIBS for the analysis of P in fertilizer and the LoD of phosphorus pentoxide was 0.28% [12]. C. M. Li et al. quantified P in pig iron and low alloy steel by time-resolved LIBS, the obtained LoDs of P were 12 and 9 ppm, respectively [13]. X. K. Shen et al. [14] and H. Kondo et al. [15] combined LIBS with laser-induced fluorescence for trace analysis of P in steel and the LoDs were around 0.7 and 5.4 ppm. M. A. Gondal et al. used LIBS for the quantitative analysis of slag samples in vacuum condition and the LoD of P was 13 ppm [16]. G. G. Arantes de Carvalho et al. introduced LIBS for analysis of plant materials and the LoD of P was 1 ppm [17]. G. Kim et al. quantitatively analyzed of 216 ppm P in water using the aerosol-assisted LIBS [18]. Recently, Y. Shao et al. transformed water from liquid to solid using iron hydroxide and dried it on the substrate surface; the LoD of P in water was 0.1 and 8.31 ppm [19]. However, the main research is on the solid matrix and there is few research on quantitative analysis of P in water without sample pretreatment.
For direct analysis of liquid by LIBS, the liquid was changed from bulk to flow as jet [20], droplet [21,22], and aerosol [23]. Among them, liquid aerosol had higher sensitivity and better spectral stability than the liquid jet and droplet [22–24]. Moreover, the liquid aerosol was produced through a nebulizer by the inert gas [25], which would inhibit the absorption of the ultraviolet emission spectrum of P by air. Therefore, liquid aerosol is suited for LIBS precise detection of trace detection of P in liquid phase matrices. However, the sensitivity of the quantitative analysis needs to be further improved by reducing the aerosol diameter [26]. A. A. Bol'shakov et al. introduced the cyclonic spray chamber to filter out the large diameter aerosols, plasma was produced in the exit of the chamber, and trace elements in oil were detected [27].
In this paper, a plasma amplification LIBS is proposed by combining a quartz tube with the entrance of a cyclonic spray chamber. Filtered aerosols are diverted through quartz tube and are fully interacted with a collinear focused laser. An amplification plasma is produced by interacted inside of the tube and is imaged in the incident slit of the spectroscope. The feasibility of plasma amplification LIBS (nebulizer combined with cyclonic spray chamber and quartz tube, Method 3) for analysis of P in water is verified against the liquid aerosol systems of nebulizer (Method 1) and nebulizer combined with cyclonic spray chamber (Method 2).
2. Experimental section
The experimental setup and three sampling methods used in this study is shown in Fig. 1. The setup consists of an aerosol system and single-pulse LIBS system.
Aerosol System. The aerosol system was composed of a concentric nebulizer (Glass expansion, 1161870 AR30-1-FM02E), cyclonic spray chamber (Meinhard, ML148030), quartz tube, gas, and syringe pump (New Era Pump Systems, NE-1000 One Channel Programmable Syringe Pump). The primary aerosols were produced by the concentric nebulizer with the help of argon gas (0.7 slm for gas rate) and a syringe pump (500 ml/min for liquid rate). The cyclonic spray chamber was connected with a concentric nebulizer to filter out large diameter aerosols. The fine aerosols were uniform and sized at the outlet. To ensure the stability of the aerosol, the filtered aerosols are diverted into a quartz tube with a diameter inside of 7 mm.
LIBS System. The LIBS equipment mainly includes a Q-Switched Nd: YAG laser (Beamtech Optronics, Dawa-300, 532 nm, 10 Hz, 7 ns), and a spectrometer (Princeton Instruments, Spectra Pro2750, 1200i@300 nm) equipped with an ICCD (Princeton Instruments, PI-MAX 1024). The plasma was produced by focusing the laser vertically on primary aerosols (Method 1), collinearly on the filtered aerosols at the 3 mm from the spray chamber outlet (Method 2), and collinearly on the filtered aerosols in quartz tube (Method 3). The focal length of the focusing lens is 25 cm. For Method 3, the distance between the laser focus and the entrance of the quartz tube is 19 mm. Plasma emission spectrum was fully imaged into the entrance slit of the spectrometer by a pair of plano-convex lenses (f = 75 mm and 100 mm). To ensure the intensity and stability of the signal, each spectrum was accumulated for 300 shots. The plasma image was recorded by an ICCD camera (Andor Technology, iStar DH334T-18H-13) attached with a 1:1 microfocus lens (Tamron, SP AF90 mm F/2.8).
Sample Preparation. The stock solution of P was prepared by dissolved the chemical reagent KHPO4 (Aladdin, China) in distilled water. The concentration of P in the stock solution is 1000 ppm. Standard solutions were obtained by diluting the stock solution with distilled water. The concentration range of P in the standard solutions was 0-1000 ppm.
3. Results and discussion
3.1 Qualitative analysis
The aqueous solution with 100 ppm P was introduced to the LIBS system by Method 1, Method 2, and Method 3. LIBS spectra and plasma images are shown in Fig. 2. As shown, the length of the plasma for Method 3 was 3 times and 2 times that of Method 1 and Method 2. And the length of the plasma is 5.27, 8.73, and 17.58 mm for Method 1, Method 2, and Method 3, respectively. This shows that Method 3 can enlarge the size of the plasma and plasma amplification is achieved. Moreover, both Method 2 and Method 3 had stronger intensities than Method 1. The main reason is that the laser ablation efficiency of fine aerosols is higher by filtering out the large diameter aerosols with the spray chamber. Compared to Method 2, Method 3 with extra quartz tube had a lower background by reducing air interference. Due to the strongest intensity, P I 213.618 nm was selected as the analytical line.
3.2 Optimization of the parameters
For optimization of the parameters, the concentrations of P for Method 1, Method 2, and Method 3 were 1000, 100, and 100 ppm, respectively. The effects of laser energy and delay on the signal-to-background ratio (SBR) and relative standard deviation (RSD) of SBR of P I 213.618 nm for three sampling methods were investigated, as shown in Fig. 3 and Fig. 4.
As shown in Fig. 3, the SBR values increased with the laser energy as the result of the increased laser ablation mass. However, there had saturation energy for Method 1 and Method 2. The results are attributed to plasma shielding or aerosol scattering of laser energy, which prevent or reduce the laser energy further interaction with the plasma. Moreover, the fluctuation of SBR for Method 3 is minimal. This result means that the spectral intensity stability of Method 3 is better. The results showed that the extra quartz tube of Method 3 not only reduced the plasma threshold but also increased the stability of aerosol transport. Taking into account SBR and RSD of SBR, optimal laser energies were chosen to be 216 mJ, 230 mJ, and 216 mJ for Method 1, Method 2, and Method 3, respectively.
To reduce the interference of the spectral background with the analytical line, the delay time was optimized [28]. As shown in Fig. 4, the trend of SBR increased first and then decreased with delay time. This is mainly because the faster decay rate of background than spectral intensity. And the lifetime of Method 3 is shortest. This was ascribed to the complete ablation of the plasma by the laser, and the higher plasma–ambience interaction due to the larger plasma volume [29]. As the delay time increased, the plasma expanded and cooled, the plasma was unstable. As a results, the RSD of SBR increased with the delay time. Give consideration to SBR and RSD, optimal acquisition delay times were 4.0 μs, 10 µs, and 2.6 µs for Method 1, Method 2, and Method 3, respectively.
3.3 Analytical performance of the quantitative analysis
Under the optimal parameters, the quantitative analysis of P in standard solutions for three sampling methods were performed. Figure 5 shows the calibration curves established by the absolute intensity of P I 213.618 nm and the RSD values of absolute intensities with different concentrations. As shown, the determination coefficient R2 of Method 2 was improved from 0.9063 to 0.9923 by Method 3. Moreover, the slope of Method 1 was improved from 17.05 to 60.79 Counts/ppm by Method 2 with the aid of cyclonic spray chamber to filter out the large diameter aerosol. With the aid of tube, the slope of Method 3 was similar to that of Method 2. However, the average RSD (RSDAV) of intensity with different concentration was reduced from 15.22% to 8.32% by Method 3. And the RSD of Method 3 was lowest than Method 2 and Method 1, when the concentration was 100 ppm. Furthermore, the LoD of Method 1 was improved from 17.75 to 6.13 ppm by Method 2, which was further improved to 3.60 ppm by Method 3. The above results shown the precision and sensitivity was improved by Method 3.
To study the error between the predicted concentration and the prepared concentration, the predicted concentration of P was obtained by Leave-one-out cross-validation. On the basis of the predicted concentration, the relative error (RE) between predicted and prepared concentration, and the root mean square error of cross-validation (RMSECV) of the predicted concentration could be calculated. Figure 6(a) shows the correlation curves of the prepared concentration versus the predicted concentration of P. And the RE values of different concentrations are shown in Fig. 6(b). As shown in Fig. 6(a), the R2 value for Method 3 was 0.9971, which is the best. Note that the R2 value of Method 2 was poorer than that of Method 1. The main reason is that the filtered aerosols from the spray chamber outlet are rarefaction and susceptible to ambient turbulence. As a results, the coupling efficiency of laser and aerosol fluctuates. As shown in Fig. 6(b), RE values of Method 3 with different concentrations were lower than those of the other two sampling methods. And the average RE (REAV) values of Method 1, Method 2, and Method 3 were 10.23%, 23.84%, and 6.17%, respectively. The 1-R2, REAV, and RMSECV values are listed in Table 1. As shown, the RMSECV value of Method 2 was improved from 16.68 to 3.24 ppm. And the RMSECV value of Method 1 was 64.29 ppm, which was improved by about 94.96% by Method 3. Therefore, Method 3 has the best analytical accuracy than Method 1 and Method 2.
Fig. 6. Correlation curves of the prepared concentration versus the predicted concentration (a) and RE values of different concentration (b) for three sampling methods.
Table 1. The 1-R2, REAV, and RMSECV values for three sampling methods.
4. Conclusion
For analyzing phosphorus element in an aqueous solution, the traditional liquid aerosol system was improved by using a spray chamber and tube. The cyclonic spray chamber was used to filter out the large diameter aerosols produced by the nebulizer. A quartz tube was connected to the entrance of the cyclonic spray chamber to guide the filtered aerosols to interact with the collinear laser. By this method, the length of the plasma was amplified about 3 times. For determining the P element, the analytical performance of LIBS was improved by plasma amplification LIBS. For example, the analytical sensitivity was improved more than 66.72% and the LoD was reduced by an order of magnitude. Moreover, precision for aerosols produced by chamber was improved about 45.34%. Furthermore, accuracy was improved more than 39.00%. Thus, plasma amplification LIBS without sample preparation is suitable for the sensitivity, precision, and accuracy detection of trace phosphorus elements in water.
Funding
National Natural Science Foundation of China (12074004, 61805002); Foundation of the State Key Laboratory of Pulsed Power Laser Technology (SKL2022ZR10); Foundation of Anhui advanced laser technology laboratory (AHL2021ZR04); National Defense Basic Scientific Research Program of China (JCKY2023230C010); Science and Technology Project of Henan Science and Technology Department (232102220014); Open Reasearch Fund of Advanced Laser Technology Laboratory of Anhui Province (AHL2020KF03); University Synergy Innovation Program of Anhui Province (GXXT-2021-029); Anhui Provincial Key Reasearch and Development Program (1804a0802193).
Acknowledgments
This work received financial support from the National Natural Science Foundation of China (No. 61805002 & 12074004); Anhui Provincial Key Reasearch and Development Program (No. 1804a0802193); University Synergy Innovation Program of Anhui Province (No. GXXT-2021-029); Open Reasearch Fund of Advanced Laser Technology Laboratory of Anhui Province (No. AHL2020KF03); Science and Technology Project of Henan Science and Technology Department (No. 232102220014). National Defense Basic Scientific Research program of China (No. JCKY2023230C010); Foundation of Anhui advanced laser technology laboratory (No. AHL2021ZR04); Foundation of the State Key Laboratory of Pulsed Power Laser Technology (No. SKL2022ZR10).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. X. G. Fei, “Solving the coastal eutrophication problem by large scale seaweed cultivation,” Hydrobiologia 512(1-3), 145–151 (2004). [CrossRef]
2. S. R. Carpenter, N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. Smith, “Nonpoint pollution of surface waters with phosphorus and nitrogen,” Ecol. Appl. 8(3), 559–568 (1998). [CrossRef]
3. C. Niu, Z. L. Hu, X. M. Cheng, A. J. Gong, K. Wang, D. Zhang, S. L. Li, and L. B. Guo, “Individual Micron-Sized Aerosol Qualitative Analysis-Combined Raman Spectroscopy and Laser-Induced Breakdown Spectroscopy by Optical Trapping in Air,” Anal. Chem. 95(5), 2874–2883 (2023). [CrossRef]
4. X. Y. Peng, B. H. Xu, Z. Y. Xu, X. T. Yan, N. Zhang, Y. Z. Qin, Q. X. Ma, J. M. Li, N. Zhao, and Q. M. Zhang, “Accuracy improvement in plastics classification by laser-induced breakdown spectroscopy based on a residual network,” Opt. Express 29(21), 33269–33280 (2021). [CrossRef]
5. X. Y. Yang, Z. Q. Hao, M. Shen, R. X. Yi, J. M. Li, H. W. Yu, L. B. Guo, X. Y. Li, X. Y. Zeng, and Y. F. Lu, “Simultaneous determination of La, Ce, Pr, and Nd elements in aqueous solution using surface-enhanced laser-induced breakdown spectroscopy,” Talanta 163, 127–131 (2017). [CrossRef]
6. G. M. Kompitsas, “Phosphate ore beneficiation via determination of phosphorus-to-silica ratios by Laser Induced Breakdown Spectroscopy,” Spectrochim. Acta, Part B 61(12), 1253–1259 (2006). [CrossRef]
7. X. B. Xu, F. Ma, J. M. Zhou, and C. W. Du, “Applying convolutional neural networks (CNN) for end-to-end soil analysis based on laser-induced breakdown spectroscopy (LIBS) with less spectral preprocessing,” Comput. Electron. Agr. 199, 107171 (2022). [CrossRef]
8. F. Y. Ma and D. M. Dong, “A Measurement Method on Pesticide Residues of Apple Surface Based on Laser-Induced Breakdown Spectroscopy,” Food Anal. Methods 7(9), 1858–1865 (2014). [CrossRef]
9. C. Pereira de Morais, G. Nicolodelli, M. Corso Mitsuyuki, K. S. Gaioso da Silva, F. F. Mauad, S. Mounier, and D. Marcondes Bastos Pereira Milori, “Total phosphorus determination in eutrophic tropical river sediments by laser-induced breakdown spectroscopy techniques,” Anal. Methods 13(1), 77–83 (2021). [CrossRef]
10. S. Sanchez-Esteva, M. Knadel, R. Labouriau, G. H. Rubaek, and G. Heckrath, “Total Phosphorus Determination in Soils Using Laser-Induced Breakdown Spectroscopy: Evaluating Different Sources of Matrix Effects,” Appl. Spectrosc. 75(1), 22–33 (2021). [CrossRef]
11. Y. Tian, Q. Chen, Y. Q. Lin, Y. Lu, Y. Li, and H. Lin, “Quantitative determination of phosphorus in seafood using laser-induced breakdown spectroscopy combined with machine learning,” Spectrochim. Acta, Part B 175, 106027 (2021). [CrossRef]
12. W. Sha, P. A. Nin, C. H. Zhen, C. P. Lu, and Y. C. Jiang, “Analysis of Phosphorus in Fertilizer Using Laser-Induced Breakdown Spectroscopy,” J. Appl. Spectrosc. 85(4), 653–658 (2018). [CrossRef]
13. C. M. Li, Z. M. Zou, X. Y. Yang, Z. Q. Hao, L. B. Guo, X. Y. Li, Y. F. Lu, and X. Y. Zeng, “Quantitative analysis of phosphorus in steel using Laser-induced breakdown spectroscopy in air atmosphere,” J. Anal. At. Spectrom. 29(8), 1432–1437 (2014). [CrossRef]
14. X. K. Shen, H. Wang, Z. Q. Xie, Y. Gao, H. Ling, and Y. F. Lu, “Detection of trace phosphorus in steel using laser-induced breakdown spectroscopy combined with laser-induced fluorescence,” Appl. Opt. 48(13), 2551–2558 (2009). [CrossRef]
15. H. Kondo, N. Hamada, and K. Wagatsuma, “Determination of phosphorus in steel by the combined technique of laser induced breakdown spectrometry with laser induced fluorescence spectrometry,” Spectrochim. Acta, Part B 64(9), 884–890 (2009). [CrossRef]
16. M. A. Gondal, T. Hussain, Z. H. Yamani, and A. H. Bakry, “Study of hazardous metals in iron slag waste using Laser Induced Breakdown Spectroscopy,” J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng. 42(6), 767–775 (2007). [CrossRef]
17. G. G. Arantes de Carvalho, D. Santos Junior, L. C. Nunes, M. D. Gomes, S. Gomes, F. D. O. Leme, and F. J. Krug, “Effects of laser focusing and fluence on the analysis of pellets of plant materials by laser-induced breakdown spectroscopy,” Spectrochim. Acta, Part B 74-75, 162–168 (2012). [CrossRef]
18. G. Kim, K. Kim, H. Maeng, H. Lee, and K. Park, “Development of Aerosol-LIBS (Laser Induced Breakdown Spectroscopy) for Real-time Monitoring of Process-induced Particles,” Aerosol Air Qual. Res. 19(3), 455–460 (2019). [CrossRef]
19. Y. Shao, S. X. Ma, X. D. Zhao, H. W. Tian, S. L. Chen, D. M. Dong, and J. J. Zhou, “Determination of phosphorus in water using iron hydroxide assisted laser-induced breakdown spectroscopy,” J. Anal. At. Spectrom. 38(3), 638–641 (2023). [CrossRef]
20. A. Kumar, F. Y. Yueh, and J. P. Singh, “Double-pulse laser-induced breakdown spectroscopy with liquid jets of different thicknesses,” Appl. Opt. 42(30), 6047–6051 (2003). [CrossRef]
21. S. Groh, P. K. Diwakar, C. C. Garcia, A. Murtazin, D. W. Hahn, and K. Niemax, “100% Efficient Sub-Nanoliter Sample Introduction in Laser-Induced Breakdown Spectroscopy and Inductively Coupled Plasma Spectrometry: Implications for Ultralow Sample Volumes,” Anal. Chem. 82(6), 2568–2573 (2010). [CrossRef]
22. E. M. Cahoon and J. R. Almirall, “Quantitative Analysis of Liquids from Aerosols and Microdrops Using Laser Induced Breakdown Spectroscopy,” Anal. Chem. 84(5), 2239–2244 (2012). [CrossRef]
23. M. D. Sui, Y. S. Fan, L. L. Jiang, Y. Y. Xue, J. Zhou, and S. L. Zhong, “Online ultrasonic nebulizer assisted laser induced breakdown spectroscopy (OUN-LIBS): An online metal elements sensor for marine water analysis,” Spectrochim. Acta, Part B 180, 106201 (2021). [CrossRef]
24. A. Kumar, F. Y. Yueh, T. Miller, and J. P. Singh, “Detection of trace elements in liquids by laser-induced breakdown spectroscopy with a Meinhard nebulizer,” Appl. Opt. 42(30), 6040–6046 (2003). [CrossRef]
25. X. Y. Yang, L. B. Guo, J. M. Li, R. X. Yi, Z. Q. Hao, M. Shen, R. Zhou, K. H. Li, X. Y. Li, Y. F. Lu, and X. Y. Zeng, “Laser-induced breakdown spectroscopy of liquid solutions: a comparative study on the forms of liquid surface and liquid aerosol,” Appl. Opt. 55(26), 7406–7411 (2016). [CrossRef]
26. R. E. Neuhauser, U. Panne, R. Niessner, G. A. Petrucci, P. Cavalli, and N. Omenetto, “On-line and in-situ detection of lead aerosols by plasma-spectroscopy and laser-excited atomic fluorescence spectroscopy,” Anal. Chim. Acta 346(1), 37–48 (1997). [CrossRef]
27. A. A. Bol’shakov, S. J. Pandey, X. L. Mao, and C. Y. Lou, “Analysis of liquid petroleum using a laser-induced breakdown spectroscopy instrument,” Spectrochim. Acta, Part B 179, 106094 (2021). [CrossRef]
28. A. Emmanuel and H. B. Yao, “Influence of laser energy on the electron temperature of a laser-induced Mg plasma,” Appl. Phys. B 123(1), 22 (2017). [CrossRef]
29. Z. Wang, M. S. Afgan, S. Sheta, Z. Hou, and Z. Li, “Understanding the laser-induced aerosol ablation of sub-micron liquid particles via size-resolved spectral and image analyses,” J. Anal. At. Spectrom. 34(12), 2385–2393 (2019). [CrossRef]
