The rapid detection of bioavailable micronutrients Cu/Fe/Zn/Mn in soil using laser-induced breakdown spectroscopy combined with solid–liquid–solid transformation

Abstract

Bioavailable fractions of micronutrients—including copper (Cu), iron (Fe), zinc (Zn), and manganese (Mn)—which comprise only 1–20% of the total soil micronutrient content—are the primary forms accessible for plant absorption and physiological processes. Accurate and timely quantification of these fractions is vital for site-specific nutrient management in precision agriculture. This study presents a novel analytical method that combines laser-induced breakdown spectroscopy (LIBS) with a solid–liquid–solid transformation (SLST) protocol to address conventional detection technique limitations. This integrated method facilitates rapid, in situ, and highly sensitive detection of bioavailable micronutrients (Cu, Fe, Zn, Mn) in complex soil matrices. The results demonstrated that bioavailable micronutrients could be accurately and efficiently detected using the proposed method. The limits of detection (LoDs) for Cu, Fe, Zn and Mn were determined to be 0.06, 0.20, 0.98, and 0.71 mg kg⁻¹, respectively, meeting the classification standards of I–III categories in the classification of bioavailable elements in Chinese soil. Furthermore, the total detection time was reduced to less than 20 minutes, highlighting the method's efficiency for rapid soil assessment. This analytical approach offers a practical and innovative solution for real-time monitoring of soil nutrients, facilitating data-driven fertilization strategies for precision agriculture.

---

1 Introduction

Copper (Cu), iron (Fe), zinc (Zn), and manganese (Mn) are vital micronutrients required for optimal plant growth and development. These micronutrients participate in key physiological processes such as photosynthesis, respiration, and enzyme activation, directly influencing plant morphology, yield, and nutritional quality.¹ Deficiencies in these micronutrients can lead to physiological disorders, reduced productivity, and inferior crop quality. In addition, the concentration and spatial variability of these elements play crucial roles in determining soil fertility and maintaining ecosystem balance.² Consequently, accurate assessment of soil micronutrient availability is fundamental for developing site-specific fertilization strategies, optimizing cropping systems, and improving agricultural product quality, all of which support sustainable agricultural practices.³

Various analytical methods for detecting soil elements are currently available. Among them, inductively coupled plasma mass spectroscopy/optical emission spectrometry (ICP-MS/OES), atomic absorption spectroscopy (AAS), and ion chromatography (IC) are widely employed to obtain high sensitivity and accuracy.^4–6 Despite their analytical precision, these methods are limited to offline laboratory environments and are unsuitable for rapid in-field detection. This limitation highlights the need to develop a fast, portable, and efficient method capable of real-time soil element analysis under field conditions.

Laser-induced breakdown spectroscopy (LIBS) has emerged as a promising analytical technique that has recently attracted significant attention.^7–11 This technology acts on the surface of the sample with high-energy pulsed laser, causing it to vaporize instantly and form plasma. Subsequently, the plasma emits spectra of characteristic wavelengths. By analyzing the characteristic spectra, the composition and content of elements in the sample can be reflected.^12,13 Notable advantages of LIBS include rapid real-time analysis, minimal sample preparation, and simultaneous multi-element detection. Furthermore, in recent years, LIBS technology has been combined with machine learning algorithms and advanced mathematical models, enabling a leap from “seeing” spectra to “understanding” information. These characteristics have enabled the application of LIBS technology to expand in various substrates such as metals, liquids, plants, foods, and plastics.^14–20 LIBS technology has also achieved rapid development in the field of soil analysis, capable of effectively detecting a wide range of elements in various soil types (such as calcium, magnesium), trace elements (such as iron, manganese), and heavy metals like lead and chromium.^21–23 Particularly, laser-induced breakdown spectroscopy (LIBS) shows great potential in the analysis of trace elements in soil. Recent studies have adopted the pellet sample preparation method to improve sample uniformity, thereby using single-pulse LIBS, dual-pulse LIBS, and microwave-assisted LIBS (MA-LIBS) technologies to quantitatively analyze trace nutrients such as iron, manganese, copper, and zinc. Experimental results show that these improved LIBS configurations can achieve a detection limit of one part per million (ppm). For example, He et al.²⁴ developed single-pulse and dual-pulse LIBS systems and combined partial least squares regression to improve quantitative performance. The detection limit of single-pulse LIBS for iron and manganese was 112 milligrams per kilogram and 70 milligrams per kilogram, respectively. In contrast, dual-pulse LIBS significantly increased sensitivity, reducing the detection limits of iron and manganese to 41 milligrams per kilogram and 37 milligrams per kilogram. In another study, Liu Y. et al.²⁵ introduced MA-LIBS technology, further optimizing the detection limits of copper and silver to 30 milligrams per kilogram and 23.3 milligrams per kilogram. It is worth noting that the detection sensitivity of copper in MA-LIBS is approximately 23 times higher than that of traditional LIBS systems. In the field of soil element detection, machine learning methods have also been extensively studied. For instance, Han B. et al.²⁶ utilized LIBS elemental imaging and RGB visualization, combined with K-means clustering and principal component analysis, to achieve the visualization of soil heavy metal spatial distribution and the classification of pollution levels. Further research indicates that machine learning not only systematically alleviates the matrix effect and quantitative problems of LIBS, but also can utilize algorithms such as deep belief networks to achieve high-precision identification of heavy metal pollution, fully demonstrating its potential in enhancing the analytical capabilities of LIBS.^27,28

Previous studies have primarily focused on analyzing total soil micronutrient concentrations. However, the actual absorption and use of plant-derived bioavailable fractions offer more significant agricultural insights. Bioavailable micronutrients, which account for 1–20% of the total soil content, represent the forms that plants can absorb. These fractions directly influence the efficiency of crop nutrient use. Therefore, detecting bioavailability fractions can accurately evaluate the soil's fertilization potential and provide a scientific basis for precision fertilization.^29–31

The main objective of this study was to propose a combination of laser-induced breakdown spectroscopy (LIBS) technology and solid–liquid–solid transformation (SLST) to detect micronutrients that are bioavailable in soil. During the experiment, the basic parameters were continuously optimized, and a calibration curve was constructed and verified using real soil samples. This novel method provides a reliable and efficient means for rapidly and accurately detecting bioavailable micronutrients in soil.

2 Materials and methods

2.1 Experimental instruments

The LIBS experimental setup used in this study is illustrated in Fig. 1. A Q-switched pulsed Nd:YAG laser (Quantel Ultra Model 100, the laser spot size is 150 micrometers and the pulse width is 8 ns) with a maximum energy of 80 mJ per pulse, a wavelength of 1064 nm, and a frequency of 20 Hz was used to generate plasma on the surface of the target sample. The light emitted by the plasma was coupled into an optical fiber through a collector and then directed to an Andor spectrometer (Andor Technology Shamrock 500i, 2400 line grating per millimeter) equipped with an intensified charge-coupled device (ICCD, iStar DH334T, Andor Technology). A digital delay generator (SRS, DG645) was used to trigger both the Nd:YAG laser and the ICCD camera. A high-precision three-dimensional platform was used to carry and adjust the sample's position using a stepper motor. In this experiment, the laser pulse energy was set to 80 mJ. To minimize the influence of continuous background noise, the delay time for spectral collection was optimized to 1 μs, with a gate width set to 5 μs. The time-integrated spectra of the laser-induced plasma were obtained in the spectral ranges of 205–214 nm (Zn II 206.20 nm), 255–264 nm (Fe II 259.94 nm), 257–266 nm (Mn II 257.61 nm), and 321–330 nm (Cu I 324.75 nm). Each spectrum was obtained by accumulating over 5 pulses. A total of 30 spectra were collected from each sample, and their average values were calculated.

Fig. 1 LIBS experimental setup.

2.2 Sample preparation

Soil samples 1, 2, 3 and 4 were collected from the surrounding areas of Beijing. Soil sample 1 was used to establish the calibration curve model, while soil samples 2, 3 and 4 served as the validation set for the model. The initial concentrations of bioavailable forms of Cu, Fe, Zn and Mn in soil extract samples were quantitatively analyzed using inductively coupled plasma optical emission spectrometry (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS). The samples to be tested were extracted with diethylenetriaminepentaacetic acid–triethanolamine–calcium chloride (DTPA–TEA–CaCl₂, the following text is referred to as DTPA) solution, and the analysis results have been converted to the actual content in the soil and are listed in Table 1. To establish a calibration curve for quantitative analysis, 32 soil samples with the same matrix background were prepared in this study. All samples were processed under consistent conditions, and different concentrations of bioavailable trace elements were introduced by adding Cu(NO₃)₂, Fe(NO₃)₃, Zn(NO₃)₂ and MnCl₂ solutions to the soil extract. Among them, the addition concentrations of Cu, Fe and Mn ranged from 0.5–15 mg kg⁻¹, and the addition concentration of zinc ranged from 10–100 mg kg⁻¹. Furthermore, the materials used in the materials used in the substrate optimization experiment were Al sheets (99.9% purity), Cu sheets (99.95% purity), Zn sheets (99.9% purity), highly oriented pyrolytic graphite (C), single-crystal silicon (Si) and glass slides.

Table 1 The content of bioavailable micronutrients in the soil

Soil sample	Cu (mg kg^−1)	Fe (mg kg^−1)	Zn (mg kg^−1)	Mn (mg kg^−1)
1	0.97	1.37	0.94	4.28
2	0.92	1.80	0.95	3.78
3	1.34	6.29	5.92	12.59
4	1.57	9.30	3.02	7.17

---

To enable the detection of bioavailable micronutrients using LIBS, an SLST pretreatment method was applied, as illustrated in Fig. 2. In this method, the extractant is thoroughly mixed with the ground soil, and the release of micronutrients is facilitated through mechanical oscillation. Subsequently, the mixture is centrifuged, and the supernatant is carefully aspirated and evenly spread on the substrate surface. After the liquid droplets dry, they form thin solid films, which serve as test samples for LIBS. To address the uneven distribution issue caused by the coffee ring effect during droplet drying, this study effectively improved the spectral stability by scanning the entire droplet area and merging all the pulse spectra. Additionally, the addition of triethanolamine to the DTPA extract solution increased the solution viscosity, thereby further suppressing the formation of the coffee ring effect.

Fig. 2 (a–e) SLST sample pretreatment process.

The key parameters for film preparation were optimized as follows: the solution drop volume was 7 μL, the drying temperature was 70–80 °C, and the drying time was 3 minutes. Under these conditions, stable and uniformly distributed film samples could be obtained.

3 Experimental optimization

3.1 The spectral signals of bioavailable micronutrients Cu, Fe, Zn, and Mn in soil

To systematically evaluate the detection effect of bioavailable Cu, Fe, Zn, and Mn using LIBS, we compared the spectral signals of Cu, Fe, Zn, and Mn in soil extract solution and water, as shown in Fig. 3 (blank background refers to the signal from a blank Zn plate or a glass slide). We can see that the spectral signal intensities of Cu, Fe, Zn, and Mn in the soil extract solution are significantly lower than those in pure water. This is mainly attributed to the complexity of the soil extract solution matrix: in addition to the target metals, it also contains high concentrations of DTPA extractant and other dissolved components. These coexisting substances compete with and dissipate part of the laser energy during the laser ablation process, resulting in a significant reduction in the energy actually used to excite the target metals Cu, Fe, Zn, and Mn, thereby causing signal suppression.

Fig. 3 (a–d) Comparison of signal intensities of blank sample, DTPA extract solution, and water solution of target elements.

To obtain the better spectra in soil extract solution, we diluted the soil extract solutions 20 times to reduce the influence of matrix elements on the spectral signals. The comparison results are shown in Fig. S1.

3.2 Optimization of experimental parameters

In order to obtain the optimal detection signal of bioavailable micronutrients in the soil, this study systematically investigated the effects of extraction agent types, extraction time, water-to-soil ratio, and substrate on the spectral signal. The extraction of bioavailable components is a key step in soil analysis. Common methods include the use of weak acids (e.g., HCl, HOAc), chelating agents (e.g., DTPA, EDTA), buffer salt solutions (e.g., NaHCO₃), neutral salt solutions (e.g., NH₄OAc), and the Mehlich 3 composite extraction method, etc.^32–35 Through preliminary screening, this study selected 0.05 mol per L ethylenediaminetetraacetic acid (EDTA), 0.05 mol per L DTPA–CaCl₂–TEA solution (DTPA), and 0.1 mol per L hydrochloric acid (HCl) as three extraction agents, and compared their extraction efficiencies. As shown in Fig. S2, the spectral signal intensity of the elements obtained by extracting with the DTPA solution was higher than that obtained by extracting with EDTA and HCl. These results indicate that the DTPA solution is superior to the other two extractants for extracting bioavailable Cu, Fe, Zn, and Mn from the soil. Therefore, the DTPA was selected as the extractant for this study.

Further, we compared the spectral enhanced effects of six substrates including Cu sheet, Al sheet, Zn sheet, Si sheet, C sheet and glass slide. As shown in Fig. 4, the optimal signals of Cu, Fe and Mn could be obtained in Zn sheet, and the optimal signal of Zn was obtained in glass slide (the corresponding spectrogram is shown in Fig. S3). This advantageous effect may be attributed to a triple synergy mechanism: the low melting point property of Zn induces the formation of a surface micro-melting layer, prolonging plasma duration; the low ionization energy facilitates the formation of a high-density “electron pool”, enhancing collision efficiency for elements with high excitation energies; the narrow full width at half-maximum of its spectral lines avoids spectral band interference. Although chemical inertness is considered an advantage of the glass slide, uneven sample crystallization results in inferior signal stability compared to the Zn substrate. High-melting-point metal substrates (e.g., Al, Cu) inhibit plasma excitation due to excessive heat dissipation, while plasma maintenance time is shortened by semiconductor substrates as a result of thermal conductivity imbalance.^36–39 Collectively, these mechanisms establish the Zn sheet/glass slide system as exhibiting optimal detection performance.

Fig. 4 Effect of different substrates on spectral intensity.

Besides, we also investigated the effect of liquid-to-soil ratio (LSR) and extract time. A gradient method was employed in which the soil mass was held constant while the volume of the DTPA solution was increased to examine its influence on the characteristic spectral intensity of the target elements. We can see from Fig. S4, as the ratio of water to soil gradually increased, the spectral signal showed a trend of gradual weakening. This trend indicates that a higher volume of DTPA solution may induce a dilution effect, thereby reducing the concentration of the target element in the extract and weakening the spectral intensity. To balance the extraction efficiency and detection sensitivity, the LSR was optimized to 2 : 1. This ratio ensured sufficient extraction of bioavailable micronutrients while minimizing dilution-induced signal attenuation caused by excess DTPA solution. As shown in Fig. S5, the spectral intensities of Cu, Fe, Zn, and Mn initially increased with oscillation time, peaking at 10 minutes, before either plateauing or slightly declining beyond this point. The initial increase in spectral intensity can be attributed to the improved contact between the DTPA solution and soil particles, which enhanced the extraction of target elements. Therefore, the optimal LSR and extract time were 2 : 1 and 10 min, respectively.

4 The sensitively detection of bioavailable Cu/Fe/Zn/Mn in soil using LIBS-SLST method

To evaluate the quantitative analysis performance of the LIBS-SLST method for determining bioavailable micronutrients in soil, the spectra and the calibration curves of different concentration for Cu, Fe, Zn, and Mn were developed, as shown in Fig. 5 and Table 2. We can see that the determination coefficients (R²) for Cu, Fe, Zn, and Mn were 0.90, 0.97, 0.99, and 0.93, respectively. The LoDs calculated using the 3σ criterion, were 0.06 mg kg⁻¹ for Cu, 0.20 mg kg⁻¹ for Fe, 0.98 mg kg⁻¹ for Zn, and 0.71 mg kg⁻¹ for Mn. The relative standard deviations (RSD) were 7.85%, 9.63%, 4.68%, and 7.94%, respectively, while the root mean square errors (RMSEv) were 1.96 mg kg⁻¹, 3.36 mg kg⁻¹, 2.64 mg kg⁻¹, and 2.60 mg kg⁻¹ for Cu, Fe, Zn, and Mn respectively.

Fig. 5 The spectra and calibration curves of Cu (a), Fe (b), Zn (C), and Mn (d).

Table 2 The LoDs, R², RSD and RMSEv of bioavailability of four micronutrients

Detect elements	R ^2	LoD (mg kg^−1)	RSD	RMSEv (mg kg^−1)
Cu	0.90	0.06	7.85%	1.96
Fe	0.97	0.20	9.63%	3.36
Zn	0.99	0.98	4.68%	2.64
Mn	0.93	0.71	7.94%	2.60

---

To further demonstrate the advantages of LIBS-SLST in detecting micronutrients (Cu/Fe/Zn/Mn), we evaluated the detection sensitivity by referring to the classification standard for the content of bioavailable micronutrients in soil, as shown in Fig. 6. As can be seen, the detection limits for Cu, Fe and Mn can reach the first-class standard (Cu: <0.1 mg kg⁻¹, Fe: <2.5 mg kg⁻¹, Mn: <1 mg kg⁻¹), and the detection limit for Zn can reach the third-class standard (Zn: <1 mg kg⁻¹). These results indicate that this method can effectively classify the content levels of bioavailable micronutrients (Cu/Fe/Zn/Mn) in soil.

Fig. 6 Content classification of four elements in soil and the levels of detection limits obtained from the experiment.

Besides, the actual different soil samples 2, 3, 4 were selected to verify the applicability of the method. The Table S1 shows the predicted values and the recovery of Cu/Fe/Zn/Mn in real and artificial samples. We can see that the Cu/Fe/Zn/Mn in real soils could be effectively detected, and the artificial samples could also be accuracy detected. The recovery of Cu/Fe/Zn/Mn in real and artificial samples within the range of 81.6–121.6%. Specifically, the average recovery rates of three soil samples were 109.50%, 98.80 and 92.60, respectively. These recovery rate results confirmed that the established calibration model had wide applicability. The model demonstrated good generalization ability for various soil matrices.

5 Conclusions

This study successfully developed a new LIBS-SLST detection method, effectively overcoming the key limitations commonly found in traditional detection methods for bioavailable micronutrients, such as complex operation, lengthy time consumption, and sample destructiveness. Experimental results demonstrated that this method can efficiently complete the extraction and quantitative analysis of the bioavailable fractions of target elements (Cu, Fe, Zn, Mn), with LoDs as low as 0.06, 0.20, 0.98, and 0.71 mg kg⁻¹. The recovery rates of the three soil validation samples (81.6–121.6%) showed excellent reproducibility and precision of the model. This method has the powerful ability to detect soil bioavailable micronutrients within a wide concentration range of ppb to ppm. Its significantly low detection limit and rapid analysis characteristics provide strong technical support for efficient, non-destructive monitoring of soil micronutrient content and in-depth research on their bioavailability.

Author contributions

Yangrui Li: writing – review & editing, writing-original draft, software, methodology, investigation, formal analysis, data curation. Zhizheng Shi: writing – review & editing, investigation, data curation, formal analysis. Leizi Jiao: funding acquisition, resources, methodology. Ning Liu: writing – review & editing, methodology, funding acquisition, investigation. Zhen Xing: writing – review & editing, investigation. Shixiang Ma: writing – review & editing, investigation, data curation, resources, supervision, methodology, project administration, funding acquisition. Hongwu Tian: writing – review & editing, supervision. Daming Dong: project administration, funding acquisition, supervision, resources.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data supporting this article are available from the corresponding author upon reasonable request.

The supplementary information (SI) file contains the key supporting data from this study: specifically, the spectral intensity comparisons of the elements involved in the optimization experiments presented in Fig. S1 to S5, and the relevant data for the model validation set listed in Table S1. See DOI: https://doi.org/10.1039/d5ja00341e.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (32171627), the National Key R&D Program of China (2023YFD1702500), the National Natural Science Foundation of China (32225035), the Innovation Capacity Building Project of Beijing Academy of Agriculture and Forestry Sciences (KJCX20230408) and the China Postdoctoral Science Foundation (No. 2024M760251).

References

1. J. J. Lucena and L. Hernandez-Apaolaza, Iron nutrition in plants: an overview, Plant Soil, 2017, 418, 1–4.
2. S. Alejandro, et al., Manganese in plants: from acquisition to subcellular allocation, Front. Plant Sci., 2020, 11, 300.
3. C. Stanton, et al., Zinc in plants: Integrating homeostasis and biofortification, Mol. Plant, 2022, 15(1), 65–85.
4. A. E. Hartemink, Assessing soil fertility decline in the tropics using soil chemical data, Adv. Agron., 2006, 89, 179–225.
5. R. S. Picoloto, et al., Determination of inorganic pollutants in soil after volatilization using microwave-induced combustion, Spectrochim. Acta, Part B, 2013, 86, 123–130.
6. M. C. Saha, et al., Determination of Ag and Cd in soil and sediment samples by graphite furnace atomic absorption spectrometry (GFAAS), At. Spectrosc., 2015, 36(4), 177.
7. M. A. Aguirre, et al., Hyphenation of single-drop microextraction with laser-induced breakdown spectrometry for trace analysis in liquid samples: a viability study, Anal. Methods, 2015, 7(3), 877–883.
8. M. A. Aguirre, et al., Dispersive liquid–liquid microextraction for metals enrichment: A useful strategy for improving sensitivity of laser-induced breakdown spectroscopy in liquid samples analysis, Talanta, 2015, 131, 348–353.
9. P. Yang, et al., High-sensitivity determination of cadmium and lead in rice using laser-induced breakdown spectroscopy, Food Chem., 2019, 272, 323–328.
10. S. Ma, et al., Determination of trace heavy metal elements in aqueous solution using surface-enhanced laser-induced breakdown spectroscopy, Opt. Express, 2019, 27(10), 15091–15099.
11. J. Bocková, et al., Potential use of surface-assisted LIBS for determination of strontium in wines, Appl. Opt., 2018, 57(28), 8272–8278.
12. K. Yu, et al., Study on Soil Elements Detection with Laser-Induced Breakdown Spectroscopy: A Review, Spectrosc. Spectral Anal., 2016, 36(3), 827–833.
13. Y. Zhang, et al., Application progress of laser-induced breakdown spectroscopy for surface analysis in materials science field, Spectrosc. Spectral Anal., 2012, 32(6), 1441–1446.
14. Z. Wang, et al., Application of laser-induced breakdown spectroscopy to real-time elemental monitoring of iron and steel making processes, ISIJ Int., 2016, 56(5), 723–735.
15. M. Cui, et al., Improved analysis of manganese in steel samples using collinear long–short double pulse laser-induced breakdown spectroscopy (LIBS), Appl. Spectrosc., 2019, 73(2), 152–162.
16. C. Du, et al., Investigation on laser-induced breakdown spectroscopy of MgCL2 solution, Optik, 2019, 187, 98–102.
17. V. K. Singh, et al., Review: Application of LIBS to Elemental Analysis and Mapping of Plant Samples, At. Spectrosc., 2021, 42(1), 99–113.
18. M. G. Nespeca, et al., Detection and quantification of adulterants in honey by LIBS, Food Chem., 2020, 311, 125886.
19. G. S. Senesi, et al., Recent advances and future trends in LIBS applications to agricultural materials and their food derivatives: An overview of developments in the last decade (2010–2019). Part II. Crop plants and their food derivatives, TrAC, Trends Anal. Chem., 2019, 118, 453–469.
20. H. Meng, W. Gao and Y. Ye, et al., Multimodal LIBS-FLIPA fusion with frame segmentation for robust plastic classification via advanced LIPA processing, Opt. Lett., 2025, 50(9), 3038–3041.
21. B. H. Zhang, et al., Quantitative Analysis of Mn in Soil Samples Using LIBS, Spectrosc. Spectral Anal., 2015, 35(6), 1715–1718.
22. L. Xiaomei, et al., Quantitative analysis of Cr in soil by laser-induced breakdown spectroscopy, Spectrosc. Spectral Anal., 2021, 41(3), 875–879.
23. Z. K. Zheng, et al., Research on laser induced breakdown spectroscopy for detection of trace Cu in polluted soil, Spectrosc. Spectral Anal., 2009, 29(12), 3383–3387.
24. Y. He, et al., Quantitative analysis of nutrient elements in soil using single and double-pulse laser-induced breakdown spectroscopy, Sensors, 2018, 18(5), 1526.
25. Y. Liu, et al., Improvement of the sensitivity for the measurement of copper concentrations in soil by microwave-assisted laser-induced breakdown spectroscopy, Spectrochim. Acta, Part B, 2012, 73, 89–92.
26. B. Han, W. Gao and J. Feng, et al., Laser-induced breakdown spectroscopy for imaging and distribution analysis of heavy metal elements in soil, J. Hazard. Mater., 2025, 139284.
27. Y. Huang, S. S. Harilal and A. Bais, et al., Progress toward machine learning methodologies for laser-induced breakdown spectroscopy with an emphasis on soil analysis, IEEE Trans. Plasma Sci., 2023, 51(7), 1729–1749.
28. Y. Zhao, M. Lamine Guindo and X. Xu, et al., Deep learning associated with laser-induced breakdown spectroscopy (LIBS) for the prediction of lead in soil, Appl. Spectrosc., 2019, 73(5), 565–573.
29. K. Van Sundert, et al., Towards comparable assessment of the soil nutrient status across scales—Review and development of nutrient metrics, Global Change Biol., 2020, 26(2), 392–409.
30. C. Dimkpa, et al., Methods for rapid testing of plant and soil nutrients, Sustainable Agriculture Reviews, 2017, pp. 1–43.
31. K. Zhong, et al., A novel near infrared spectroscopy analytical strategy for soil nutrients detection based on the DBO-SVR method, Spectrochim. Acta, Part A, 2024, 315, 124259.
32. M. Jalali, et al., Risk assessment of available and total heavy metals contents in various land use in calcareous soils, Environ. Earth Sci., 2023, 82(12), 298.
33. R. Zhang, et al., Extraction methods optimization of available heavy metals and the health risk assessment of the suburb soil in China, Environ. Monit. Assess., 2023, 195(10), 1221.
34. M. Iatrou, et al., Determination of soil available phosphorus using the Olsen and Mehlich 3 methods for Greek soils having variable amounts of calcium carbonate, Commun. Soil Sci. Plant Anal., 2014, 45(16), 2207–2214.
35. R. Yi, et al., Determination of trace available heavy metals in soil using laser-induced breakdown spectroscopy assisted with phase transformation method, Anal. Chem., 2018, 90(11), 7080–7085.
36. T. Ge, et al., Effect of metal substrate temperature on the determination of trace metal elements in water by combined LIBS with electro-deposition, J. Anal. At. Spectrom., 2022, 37(12), 2510–2516.
37. S. Ma, et al., Determination of trace heavy metal elements in aqueous solution using surface-enhanced laser-induced breakdown spectroscopy, Opt. Express, 2019, 27(10), 15091–15099.
38. M. A. Aguirre, et al., Elemental analysis by surface-enhanced Laser-Induced Breakdown Spectroscopy combined with liquid–liquid microextraction, Spectrochim. Acta, Part B, 2013, 79, 88–93.
39. D. Bae, et al., Spreading a water droplet on the laser-patterned silicon wafer substrate for surface-enhanced laser-induced breakdown spectroscopy, Spectrochim. Acta, Part B, 2015, 113, 70–78.

---