© 2021 MJH Life Science and Spectroscopy Online. all rights reserved.

© 2021 MJH Life Sciences™ and Spectrum Online. all rights reserved.

In our annual report on X-ray analysis trends, experts in this field provided the latest information on the latest developments in X-ray analysis, focusing on the technology and related applications.

Topics discussed include the comparison of total reflection X-ray fluorescence (TXRF) and graphite furnace atomic absorption spectroscopy (GFAAS) for the analysis of trace diets in biological materials, the use of grazing incidence X-ray diffraction to study nanostructured layers, and multi-element analysis by TXRF Apatite crystals, and track the absorption and transportation of nutrients sprayed on the foliage of plants.

Martina Schmeling and Michelle Gende, Loyola University Chicago, Chicago, Illinois

A sample preparation procedure was developed for the analysis of trace metals in biological materials using E. coli as a model organism. A specific trace metal concentration is added to the sample, and the prepared sample is analyzed by total reflection X-ray fluorescence (TXRF) and graphite furnace atomic absorption spectroscopy (GFAAS). Both measurement techniques are equally effective and produce similar results, but due to the naturally occurring concentration of the element, there is some concentration difference in one of the added target elements. More experiments with different standard elements are planned to verify these results.

Trace metals are used as cofactors and stable proteins for enzymes and participate in many biological functions. Excess and deficiency of trace metals can be harmful to organisms and cause malfunctions and disorders. Trace metals may also be toxic and disrupt basic enzymatic processes. Just like lead, lead can interfere with calcium-dependent enzymes and cause oxidative stress and other effects (1). Therefore, it is very important to carefully monitor the levels of trace metals in biological systems. However, this is not an easy task, because the high organic matrix content, the presence of high concentrations of metabolic elements, and the natural variability of biological materials must be considered. The development of robust and reliable procedures can be used as the basis for future research. Such a procedure should be based on a model system that is easily available in a laboratory environment and has enough natural variability to be representative. Escherichia coli was discovered by T. Escherich in 1885 and is the most studied microorganism. Its genome has been completely sequenced, and many breakthrough discoveries in genetics and microbiology have used E. coli as a model system (2). It is the basis of biotechnological processes and is often used in treatments. In addition, the advantage of Escherichia coli over more complex organisms is that it is easy to grow in the laboratory; therefore, it is suitable for research purposes.

We present here the preliminary results obtained from the development of a sample preparation method for the determination of trace metals in E. coli by total reflection X-ray fluorescence (TXRF) and graphite furnace atomic absorption spectrometry (GFAAS). A predetermined concentration of chromium and copper are added to the sample. We chose chromium and copper because these two elements are essential for many biological systems, and both methods can easily detect these levels. For TXRF quantification, we use gallium as an internal standard and add it to the sample along with chromium and copper, while for GFAAS, we use external calibration for quantification.

All samples were prepared from the same E. coli BL21 cell line and cultured in Luria-Bertani broth (Fisher Bioreagents) at 37°C for 12-18 hours. (3). Harvest the cells after centrifugation at 5000 rpm for 5 minutes, and wash them twice with 5 mL of 50 mM Hepes ((4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; C8H18N2O4S; pH = 7.5; VWR) For the second wash, resuspend the cells in 500 μL of Balanced Salt Solution (BSS), and then disperse 50 μL aliquots into different 1.5 mL centrifuge tubes. Take these 50 μL aliquots of cells Dry overnight at 95 °C, then store the dried cell pellets at room temperature until they are ready for final sample preparation.

For analysis by TXRF and GFAAS, 500 μL 1:4 H2O2 (30%):HNO3 (70%) (all from Sigma Aldrich) solution was added to the cell pellet along with standards containing chromium, copper, and gallium (EMD) In different concentrations, as shown in Table I. Gallium is used as an internal standard. Each group of samples were prepared in triplicate, and a reference substance containing 100 ng/mL gallium was prepared as an internal standard. The resuspended sample was digested in a microwave at 1200 W (MW535OW, Samsung) for 1 minute, then dried again at 100°C for 48 hours, and then 500 μL 1:4 H2O2:HNO3 solution was added. The final sample is then divided into equal parts for TXRF and GFAAS analysis.

For TXRF analysis, pipette 5 μL of each sample onto a clean, pre-checked quartz carrier and dry it in a desiccator overnight. S2 PicoFox (Bruker) was used for analysis with molybdenum excitation at 50 kV and 600 μA. The counting time is 2000 s. Use factory-installed software to evaluate spectra with gallium as an internal standard.

For GFAAS analysis, transfer 20 μL of each sample to the graphite furnace platform and use the protocol shown in Table II (AA-7000 Shimadzu) for measurement. External calibration in the form of a calibration curve is used for quantification.

In Figure 1, the TXRF spectrum of one of the samples containing 100 ng/mL chromium, copper, and gallium is shown, the latter being used as an internal standard.

Table III lists the results obtained for each concentration and its standard deviation. Figure 2 shows the percentage deviation from the increased target value. For chromium, the measured concentration is consistent with the target concentration, only the lowest value is slightly overestimated. In addition, considering the differences in quantification methods and analysis types between the two technologies, the consistency between TXRF and GFAAS data is excellent. For copper, the situation is slightly different: TXRF overestimates the minimum value, and GFAAS underestimates it. For intermediate values, both TXRF and GFAAS return much higher results than expected. For the above data point, TXRF is accurate, and GFAAS again underestimates it.

These differences from the copper target value may have multiple reasons: First, copper is naturally present in the sample at a considerable concentration, even if the cells are from the same strain and grown under the same conditions, natural variation may also play a role. TXRF measured The average copper concentration of the unspiked samples was 75.3 ± 88 ng/mL and the average copper concentration measured by GFAAS was 102 ± 80 ng/mL. On the other hand, for both methods, the chromium concentration of the unspiked sample was always below the detection limit. Secondly, the standard concentration added to the sample does not fully meet the target, which may be the case with the intermediate value of copper, because both methods seriously overestimate it. Finally, the distribution of copper on the TXRF sample holder of the homogenizer and GFAAS furnace is insufficient, and the procedure needs to be better optimized. For the above reasons, the combination of all these factors is likely to actually affect the copper data. Therefore, additional experiments with one or more different standard elements are required to confirm the results of chromium.

The initial data provided here shows that TXRF and GFAAS have reached agreement on the sample preparation method developed for chromium, but not for copper. The reason for the latter may be due to the larger intrinsic and variable copper concentration. Therefore, it is necessary to conduct more experiments on the standard elements whose intrinsic concentrations are lower than the detection limit of the two technologies to consolidate the chromium data.

(1) G. Flora, D. Gupta, and A. Tiwari, Interdisciplinary Toxicology 5(2), 47–58 (2012).

(2) ZD Blount, elife 4, e05826; DOI:10.7554/elife.05826 (2015).

(3) AR Tuttle, ND Trahan and MS Son, current agreement 1, e20; doi: 10.10021cpz1.20 (2021).

Martina Schmeling and Michelle Gende work at Loyola University Chicago in Chicago, Illinois. Contact directly: Mschmel@luc.edu

Cosmin Romanitan, National Institute of Microtechnology Research and Development Voluntari, Romania

Grazing incidence X-ray diffraction technology allows scanning along the entire length of the nanowire array. The rocking curves, ω-scan and φ-scan recorded at different angles of incidence allow us to obtain bending and torsion profiles along the z-axis. These profiles help us prove that the strain relaxation process does not exist or exists. In addition, the density of screw and edge thread dislocations can be quantified within the framework of bending and torsion energy distribution. The results show that there is a close relationship between the nanowire morphology and the strain relaxation process, and the strain relaxation process can lead to the occurrence of thread dislocations or other structural defects.

Due to its unique and customizable optical and physical properties, the application of low-dimensional materials in various fields such as electronics, photonics and biological devices has undergone a revolution. The key role at the boundary between microstructure and physical properties is represented by strain and latent processes. Strain has always been a double-edged sword for heterogeneous structural design. On the one hand, the constraints of certain materials can be combined to form a coherent structure. On the other hand, strain engineering provides a way to modify material properties (1). In this regard, X-ray-based methods provide valuable information for a wide range of materials and promote the development of material science knowledge. So far, the X-ray analysis of low-dimensional layers has been limited, mainly due to the difficulty of data interpretation, which is attributed to the strong three-dimensional local inhomogeneity of the studied materials (2-4).

This article describes our findings related to one-dimensional silicon nanowires (SiNW) obtained by metal-assisted chemical etching. Specifically, we are interested in extended structural defects, such as edges and thread dislocations in high-density nanowire arrays. Since thread dislocation density is a key aspect of physical properties, special attention should be paid to their formation.

Nanostructured silicon needs new formalism

Transmission electron microscopy has always been one of the most commonly used techniques for studying strain and related defects in nanowire systems (5). Its disadvantage is that the strain state can be changed by accelerating the flux of the electron beam. On the other hand, spectroscopic techniques using single nanowires (1, 4, 9) and nanowire arrays, such as Raman spectroscopy (6, 7), photoluminescence (8) and X-ray diffraction (XRD) (10-12) ). One of the main problems in characterizing the strain of non-planar structures is to solve the interaction between structural defects, bending and torsion effects in Raman mode, photoluminescence bands and the different contributions given by XRD peaks in the standard framework. In addition, considering the strong anisotropy of strain along the z-direction, it is necessary to develop a suitable spectroscopic method to distinguish the strain along the z-direction.

For example, in the case of nanowires, the defect area cannot be treated by the free defect area alone, because the protocol usually gives misleading results, whether using Raman spectroscopy, photoluminescence, or X-ray rocking curves. The above aspects prompted us to find suitable spectroscopic tools to characterize one-dimensional systems that have the inherent anisotropy of strain-related phenomena along the z-direction.

Overcome the problems associated with nanowires

In order to obtain information at different points along the z-axis, we explored the ability of X-rays to encode information at different penetration depths through grazing incidence X-ray diffraction (GI-XRD) technology. This technique allows the information depth to be adjusted in the range from tens of microns for symmetric oblique diffraction to tens of nanometers for strongly asymmetric oblique geometry, where the angle of incidence is lower than the critical angle of total external reflection (13). In this study, GI-XRD (111) reflection was used, and the incident angle (αi) of the source was appropriately adjusted to obtain the X-ray penetration depth equivalent to the length of the array, while the detector was kept at 2θB. The schematic diagram of grazing incidence X-ray diffraction is shown in Figure 1.

We prefer (111) reflection for these types of studies because of the high structure factor and high inclination angle, χ = 54.7°, which corresponds to the angle between the (111) and (001) directions. It is worth noting that in addition to the (224) or (115) reflections of χ = 35.26 and 15.79°, respectively, the high χ angle allows us to obtain a smaller penetration depth and evaluate the area close to the surface. The X-ray penetration depth is calculated according to the Beer-Lambert formula, where the path length is assumed to be the sum of the incident path and the scattering path. More details of X-ray penetration depth calculation can be found in our previous work (14-16), where grazing incidence X-ray diffraction is applied to silicon nanowires, and GaN is applied to heteroepitaxial layers and porous silicon, respectively.

In addition, by recording multiple rocking curves at different incident angles, the length of the nanowire array should be adjusted appropriately, and we can fully describe the strain-related phenomena. For example, the ω scan measures the tilt (φtilt) of the nanowire array, which can be regarded as a curved profile. At the same time, φ-scan encodes the twist (φtwist) to further express the twist profile.

Bending and torsion curves in nanowire arrays

Figures 2a-c show the bending and torsion curves of two nanowire arrays with lengths of 1.5 μm and 9.5 μm, further denoted as short nanowires and long nanowires.

It can be observed that for short nanowires, both φtilt and φtwist monotonously increase as the penetration depth decreases. In contrast, for long nanowires, both φtilt and φtwist show a decline. In our extensive work (14), we showed that this decline is related to the coalescence area on the upper part of the nanowire array, and that there are bending and torsional relaxation mechanisms in the long nanowires. At the same time, for short nanowire arrays, there is no relaxation process.

Quantification of thread dislocations in nanowire arrays

In order to obtain a quantitative description of the strain relaxation processes in the long nanowires, they are clearly shown, and the energy lost through these processes is evaluated. The energy loss is evaluated from the bending and torsion energy distribution, as shown in Figure 2b. Although there is no strain relaxation process for short nanowire arrays, in the case of long nanowires, the entire structure observed in bending and torsion profiles is now reflected in the energy distribution along the length of the nanowire. We evaluated the energy loss from the start of relaxation to the point of relaxation, as shown in Figure 2d. The energy accumulated in nanowires is usually released by the generation of dislocation networks (17). In order to calculate the dislocation density in the nanowire, we believe that the bending (ΔEbending) and twisting (ΔEtorsion) energy in the nanowire is actually consumed by the formation of spiral and edge thread dislocations. The complete derivation of the thread and edge thread dislocation density is given in (14), which is not given here. Finally, it is found that the thread dislocation density (3.64 × 107 cm-2) is higher than the edge dislocation density (8.48 × 106 cm-2). These results show that the number of dislocations passing through the nanowire caused by the tilt relaxation is higher than the number along the nanowire, which is based on theoretical research carried out within the framework of the elastic theory of cylindrical thin rods [17].

We demonstrated the ability of grazing incidence X-ray diffraction at different angles of incidence, providing us with reliable information about bending and torsion curves, and helping to improve our understanding of the strain relaxation process in high-density nanowire arrays. Grazing incidence X-ray diffraction studies are realized on asymmetric (111) reflections at different incident angles, allowing us to penetrate the X-ray depth from tens of nanometers to 10 microns, corresponding to the length of the array.

(1) RB Lewis, P. Corfdir, H. Kupers, T. Flissikowski, O. Brandt and L. Geelhaar, NanoLett. 18, 2343–2350 (2018).

(2) A. Davtyan, D. Kriegner, V. Holy, A. AlHassan, R. Lewis, S. McDermott, etc., J. Appl. Crystallography 53, 1310–1320 (2020).

(3) AA Hassan, WA Salehi, RB Lewis, T. Anjum, C. Sternemann, L. Geelhaar and U. Pietsch, nanotechnology. 32, 205705 (2021).

(4) J. Wallentin, D. Jacobsson, M. Osterhoff, MT Borgstrom and T. Salditt, NanoLett. 17, 4143–4150 (2017).

(5) KL Kavanagh, I. Saveliev, M. Blumin, G. Swadener and HE Ruda, J. Appl. Physics 111, 044301 (2012).

(6) N. Begum, M. Piccin, F. Jabeen, G. Bais, S. Rubini, F. Martelli and AS Bhatti, J. Appl. Physics 104, 104311 (2008).

(7) Y. Chen, B. Peng and W. Wang, J. Phys. Chemistry C 111, 5855-5858 (2007).

(8) B. Jenichen, O. Brandt, C. Pfuller, P. Dogan, M. Knelangen and A. Trampert, Nanotechnology 22(29), 295714 (2011).

(9) T. Stankevic, E. Hilner, F. Seiboth, R. Ciechonski, G. Vescovi, O. Kryliouk, U. Johansson, L. Samuelson, G. Wellenreuther, G. Falkenberg, R. Feidenhans'l, R ., and A. Mikkelsen, ACS Nano. 9, 6978–698 (2015).

(10) HV Stanchu, AV Kuchuk, Vice President Kladko, ME Ware, YI Mazur, ZR Zytkiewicz, AE Belyaev and GJ Salamo, Nanoscale Res. Wright. 11, 81 (2016).

(11) D. van Treeck, G. Calabrese, JJW Goertz, VM Kaganer, O. Brandt, S. Fernandez-Garrido, S and L. Geelhaar, Nano Res. 11, 565–576 (2018).

(12) T. Auzelle, X. Biquard, E. Bellet-Amalric, Z. Fang, H. Roussel, A. Cros and B. Daudin, J. Appl. Physics 120, 225701 (2016).

(13) D. Grigoriev, S. Lazarev, P. Schroth, AA Minkevich, M. Kohl, T. Slobodskyy, M. Helfrich, DM Schaadt, T. Aschenbrenner, D. Hommel and T. Baumbach, J. Appl. Crystallography 49, 961–967 (2016).

(14) C. Romanitan, M. Kusko, M. Popescu, P. Varasteanu, A. Radoi and C. Pachiu, J. Appl. Crystallography 52, 1077–1086 (2019).

(15) C. Romanitan, I. Mihalache, O. Tutunaru and C. Pachiu, J. Alloy Compd. 858, 157723 (2021).

(16) C. Romanitan, P. Varasteanu, D. Culita, A. Bujor, O. Tutunaru, J. Appl. Crystallography 54, 847–855 (2021).

(17) VM Kaganer, B. Jenichen and O. Brandt, Physics. Pastor application 6, 064023 (2016).

Cosmin Romanitan works at the National Institute of Microtechnology Research and Development in Voluntari, Romania. Contact directly: cosmin.romanitan@imt.ro

Artem S. Maltsev, Alexei V. Ivanov and Galina V. Pashkova, Irkutsk Institute of Crustal Research, Russian Federation

Elemental analysis of apatite minerals is an important research that can be used in various applications in geology, biology, medicine, agriculture, technology and environmental remediation. Two new methods for rapid and cost-effective multi-element analysis of apatite samples by total reflection X-ray fluorescence (TXRF) spectroscopy have been developed. The first method is used for apatite samples to solve the "source" geological task. The second development was used to analyze apatite crystallites with a size of 50-60 μm, with an emphasis on uranium. Determining the uranium concentration in apatite crystallites is a prerequisite for fission track dating technology. The sample preparation strategy, quantitative analysis procedure, and accuracy evaluation of the two TXRF methods have been considered. The advantages of TXRF spectroscopy compared with other analytical methods are also discussed.

Apatite Ca5(PO4)3(F, Cl, OH) is one of the most common minerals on the earth and exists in many sedimentary rocks, metamorphic rocks and igneous rocks (1). In addition to high calcium and phosphorus content, apatite minerals are also rich in rare earth elements strontium, uranium, and thorium. Their quantitative analysis can be used for provenance analysis (matching sediments with their source igneous and metamorphic rocks) and petrology (deciphering the characteristics of these rocks). Origin) Deciphering the origin in geology (2). Uranium in apatite is important because knowing its concentration is a prerequisite for fission track dating (3). Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is the most commonly used technique for elemental analysis of apatite (4). However, this technology is not cost-effective. Other analytical techniques, such as electron probe X-ray microanalysis (5) and wavelength dispersive X-ray fluorescence spectroscopy (6), are either not sensitive enough or require large samples. Therefore, we use TXRF spectroscopy because it is fast and cost-effective, and can analyze very small amounts of material (tens of micrograms) (7).

The TXRF analysis of apatite samples was performed using an S2 Picofox spectrometer (Bruker Nano) equipped with a Moonode X-ray tube (working conditions of 50 kV and 0.75 mA). The measurement was taken during 500 seconds. The following reagents and materials were used: Concentrated nitric acid (ultra pure grade, Merck) and ultra pure deionized water (18.2 MΩ, Elga Labwater) were used for sample preparation; the quartz carrier was used as the sample holder; the Durango reference material was used for authenticity evaluation. TXRF was applied to apatite samples from the Slyudyanka metamorphic complex (Siberia), which is known for its mineralogical versatility (8) and the Tanzibei (9) complex (Altai).

Based on previous papers dedicated to TXRF analysis of geological objects, there are two main sample preparation techniques: suspension and acid digestion. Suspension preparation requires careful grinding procedures, and the final particle size is less than <10 μm, which is difficult if only a small amount of sample is available. The acid digestion procedure is more suitable for analyzing apatite crystals. The sample preparation technique for bulk apatite is optimized as follows: 5-10 mg of sample is mixed with 100 μL of HNO3 in a closed PTFE container and heated at 160 °C for 30 minutes; the resulting solution uses 800 μL of ultrapure Dilute with H2O and 100 μL gallium internal standard.

For miniature samples of apatite (crystals with a mass of about tens of micrograms), we prepare the samples directly on the carrier, because the above sample preparation techniques cannot be reproduced with such a small sample. Under the microscope, carefully place the single crystal apatite in the center of the pre-silicided quartz carrier. Then, 500 nL of concentrated nitric acid was deposited on the crystal, and the carrier was heated at a gradually increasing temperature (90 °C) to obtain a uniformly distributed sample. Then add about 1 μL of ultrapure water to dissolve the calcium salt, and dry the sample again. Using Raman spectroscopy allows us to study the formation of minerals and their distribution in crystals. Due to the uniform distribution of phosphorus in the sample, its stoichiometric value is used as an internal standard for quantitative analysis (equal to 170 ± 10 g/kg).

Analyze Durango reference materials to evaluate the authenticity of the two TXRF methods. Since the reference material is not certified, ICP-MS analysis was performed for verification. Table I lists the TXRF and ICP-MS results and reference values ​​of Durango Apatite (10), expressed in terms of average concentration values ​​and related standard deviations. The results of the two TXRF methods for most elements are consistent with the ICP-MS and reference values. The TXRF detection limit for a large number of samples is between 0.1 and 0.7 mg/kg. For microcrystals, the detection limit is between 1.0 and 20 mg/kg. Some deviations in the results may be caused by the inhomogeneity of the apatite sample, especially in the case of crystallite analysis.

The TXRF method of batch samples was applied to apatite minerals from the Slyudyanka complex, and the results were processed by principal component analysis (PCA). Figure 1 shows the three main groups of apatite minerals with different origins. The PCA results are completely consistent with the sample geochemical data.

The TXRF method of apatite crystallites is currently under consideration to determine the fission track dating of uranium. The uranium in 16 apatite crystallites from the Tanzibei complex was analyzed by TXRF and LA-ICP-MS. The uranium concentration value was found to be between 3.0 and 155 mg/kg. The results showed good convergence, the correlation coefficient r2=0.9932, and the average recovery rate was 96±13%. Some results obtained by these methods did not reach agreement, which may be due to the difference between the two methods. LA-ICP-MS is a local method. The measurement takes place in the ablation pit of the crystal about 20-30 μm. Unlike TXRF, TXRF analyzes the entire crystal.

Considering the advantages of TXRF over conventional methods of apatite analysis, including fast, cost-effective, and simple performance, the proposed method may be an excellent tool for applications involving apatite.

The research was carried out within the framework of funding from the Ministry of Science and Higher Education of the Russian Federation (funding number: 075-15-2019-1883). The research was carried out using SB RAS from the Institute of Crustal Research in conjunction with the center's equipment for geodynamics and geochronology. The author thanks S. Panteeva for ICP-MS analysis, A. Marfin for Raman spectroscopy, and L. Reznitskii and Yu. Bishaev provided samples.

(1) Apatite: mineral information, data and location, Mindat.org. http://mindat.org/min-29229.html (accessed June 2, 2021).

(2) A. Zirner, MAW Marks, T. Wenzel, DE Jacob, and G. Markl, Lithos 228-229, 12-22 (2015). https://doi.org/10.1016/j.lithos.2015.04.013.

(3) EA Belousova, WL Griffin, SY O'Reilly and NI Fisher, J. Geochem. explore. 76, 45-69 (2002). https://doi.org/10.1016/S0375-6742(02)00204-2.

(4) DM Chew, MG Babechuk, N. Cogne, C. Mark, GJ O'Sullivan, IA Henrichs, D. Doepke and CA Mckenna, Chemistry. Jell. 435, 35–48 (2016). https://doi.org/10.1016/j.chemgeo.2016.03.028.

(5) H. Teiber, MAW Marks, AA Arzamastsev, T. Wenzel and G. Markl, J. Min. Geochemistry. 192, 151–167 (2015). https://doi.org/10.1127/njma/2015/0277.

(6) L. Muia and R. Van Greeken, anus. Humph. Journal 251, 177–181 (1991). https://doi.org/10.1016/0003-2670(91)87132-Q.

(7) AS Maltsev, AV Ivanov, VM Chubarov, GV Pashkova, SV Panteeva, LZ Reznitskii, Talanta 214, 120870 (2020). https://doi.org/10.1016/j.talanta.2020.120870.

(8) J. De Grave, E. De Pelsmaeker, FI Zhimulev, S. Glorie, MM Buslov and P. Van den haute, Structural Physics 621, 44-59 (2014). https://doi.org/10.1016/j.tecto.2014.01.039.

(9) Slyudyankabaikal Region, Irkutsk Region, Russia, https://www.mindat.org/loc2719.html (accessed June 2, 2021).

(10) KP Jochum, U. Nohl, K. Herwig, E. Lammel, B. Stoll and AW Hofmann, Geostand. Geographical anus. Reservoir 29, 333–338 (2005). https://doi.org/10.1111/j.1751908X.2005.tb00904.x.

Artem S. Maltsev, Alexei V. Ivanov and Galina V. Pashkova work at the Irkutsk Crust Research Institute, Russian Federation. Direct communication to: artemmaltsev1@gmail.com

Thainara Rebelo da Silva, Gabriel Sgarbiero Montanha, Camila Graziele Corrêa, João Paulo Rodrigues Marques and Hudson Wallace Pereira de Carvalho Nuclear Instrument Laboratory, Agricultural Nuclear Energy Center, University of Sao Paulo, Sao Paulo, Brazil

X-ray fluorescence (XRF) spectroscopy can be used to analyze the distribution and fate of fertilizers applied to plant leaves (foliar fertilization). Our team has been developing methods to enable commercial and internally built portable systems to quantify several mineral nutrients in major crops such as soybeans, corn, and coffee through non-destructive and in vivo analysis in the field. In addition, XRF microanalysis has been applied to study the biofortification of grains and the foliar absorption of fertilizers. In this method, XRF technology can describe the anatomical distribution of minerals in the tissue. These studies show that XRF spectroscopy is an important technology for plant nutrition diagnosis and metalomics research. In addition, it is a powerful tool to increase knowledge about how plants absorb, transport, and metabolize foliar fertilizers.

For thousands of years, agriculture and forestry have provided us with food, feed, fiber, energy and materials. Recently, they have also been responsible for providing renewable fuels, such as ethanol, which account for approximately 68% of global biofuel production (1). In addition, concerns related to climate change and possible oil scarcity promote agriculture as a platform for the production of the chemical industry. Coupled with the ever-increasing world population and rising living standards, these factors have led to pressure to increase agricultural production. A more sensible way to respond to these needs is to increase productivity, rather than increase acreage.

Agricultural productivity is a complex function that depends on many factors, one of which is plant mineral nutrition. In addition to carbon, hydrogen, and oxygen, plants need 14 other mineral nutrients (2). Maintaining adequate levels of these nutrients is a key part of the productivity equation. Since the concentration of these nutrients in the soil may be low and will be depleted after each crop season, they must be replenished in the form of fertilizer.

Although soil sowing is the most common method of fertilization, it is not necessarily the most effective way to deliver certain nutrients to plants. Depending on the physical, chemical and biological conditions of the soil, it may be difficult for plants to obtain nutrients (3). Therefore, alternative methods such as foliar fertilization and seed treatment may be more suitable for alleviating plant defects. For example, for plants grown in iron-deficient calcareous soils, iron foliar fertilization is significantly more effective than soil application (4).

Foliar fertilizers can be absorbed by the epidermal cuticle and stomata (pores) of the leaves and reach the phloem tube by diffusion, where they are loaded and transferred first to the petiole and then to other parts of the plant (5,6). This process is shown in Figure 1.

However, leaf tissue is very complex, and this process may vary for different species and environmental conditions. Therefore, tracking the fate of mineral nutrients applied to leaves can be challenging (7).

X-ray fluorescence (XRF) spectroscopy technology can detect most mineral nutrients. However, it is worth noting that the element sensitivity depends on the atomic number; therefore, the detection limit varies for each element. Elements such as phosphorus, sulfur, potassium, and calcium are easily detected because they are present in plant tissues at hundreds to thousands of mg/kg, while micronutrients such as manganese, iron, copper, zinc, and molybdenum, although Their XRF yield is higher, but more challenging because they exist in a lower concentration range, usually from tens to hundreds of mg/kg. On the contrary, due to its low atomic number, it is difficult to detect the relatively high concentration of magnesium in plant tissues. It is good for magnesium detection under vacuum or helium atmosphere. The detection of elements such as nitrogen and boron is not practical.

A benchtop energy dispersive XRF device using 50 W anode and ~45o geometry competes with atomic absorption spectroscopy (AAS) for traditional laboratory-based foliar nutrition diagnosis. In this case, in the powder tissue of leaf, root, seed or stem material, no sample digestion is required to quantify the nutrient content (8). Therefore, despite the higher detection limits, XRF may be suitable for this application because of its higher speed and lower cost compared to AAS.

In traditional laboratory-based leaf diagnosis, farmers randomly collect a specific number of leaves representing a certain crop plot. The leaves are then sent to the laboratory, where they are washed, dried, ground, digested, and finally analyzed by atomic or molecular spectroscopy. The entire process may take several days to several weeks. However, the time window for detecting and correcting nutritional deficiencies in annual crops (such as soybeans, corn, rice, wheat, and tomatoes) is short. For example, the two weeks between shipping leaf samples and obtaining results are too long.

For this reason, our team has been developing a strategy for foliar diagnosis under field conditions. For this work, we used commercial handheld devices and built-in portable systems. The main feature of the instrument is the short X-ray beam path, or the use of vacuum, for the analysis of macronutrients, and the primary filter for the analysis of micronutrients. Currently, it is possible to quantify phosphorus, sulfur, calcium, potassium, manganese, iron, and zinc in soybean fields within minutes with the same accuracy and precision as traditional laboratories (9 and other unpublished results). The method is currently being tested on coffee trees and corn, and efforts are being made to expand the number of detectable elements.

In addition, XRF spectroscopy has been used to study the absorption and transport of mineral nutrients in plants (10). Figure 2 shows a sample holder for a benchtop device for in vivo XRF determination of foliar fertilizer absorption on soybean petioles. The X-ray beam spot size in the millimeter range provides a pinhole and collimator to accurately illuminate the desired tissue. The method we developed allows to track leaf (11-13) and root absorption (14-16), as well as short- and long-distance transportation of petioles and stems, respectively. This means that the absorption rate of nutrients can be determined without the use of radioisotopes, thereby comparing the performance of commercially available products and supporting the development of more effective nutrient sources. In this case, care must be taken to avoid inducing X-ray beam radiation to damage plant tissues. Irradiation with an unfocused beam provided by an anode operating at 40.5 W for up to 60 minutes did not cause any detectable damage (17).

Finally, XRF spectroscopy can also produce chemical images that reveal the spatial distribution of elements. Capillary optics can focus the X-ray beam downwards, thereby providing high flux density and lateral resolution. This tool supports research that unlocks nutrient pathways in plant tissues and reveals the fate of nutrients. An interesting case study involves the biofortification of food by selenium. This element is not necessary for plants, but it is necessary for animals. Most people live with selenium deficiency. Therefore, together with our collaborators, we have been working hard to increase the selenium content in staple foods such as rice and cowpea. Chemical images of selenium and other elements show that most mineral nutrients are concentrated in the seed coat of cowpea (18) and the aleurone layer (bran and germ) of rice (19, 20). Therefore, biofortification strategies are only effective when the required nutrients are incorporated into the endosperm, which is the most widely consumed part of the grain.

XRF spectroscopy has great application potential in agricultural science. Because it is mostly unknown in this scientific community, we must continue to work intensively to promote it. A key advantage of XRF spectroscopy in such applications is its low invasiveness, which allows it to be used in combination with plant sensors (such as infrared gas analyzers, fluorescence measurement to detect organic molecules) and electrodes for electrophysiological reactions ( 17).

The XRF facility was funded by the São Paulo Research Foundation (FAPESP)’s Multi-User Equipment Program (funded 2015-19121-8) and Financiador de Estudos e Projetos (FINEP), and was funded by the “Core Facility de suportes às pesquisas em Nutrologia e Segurança Alimentar na USP ”Project (authorization number 01.12.0535.0). TRS was founded by Coordenação de Aperfeiçoamento de Pessoal de Pessoal de Nível-Brasil (CAPES)-Finance Code 001 (grant 88887.498165/2020-00). GSM received a doctoral scholarship from the Sao Paulo Research Foundation ( Grant No. 20/07721-9). CGC received a scholarship funded by Ubyfol (fund 10394-0). HWP Carvalho received a Research Productivity Scholarship from the National Council for the Development of Science and Technology (CNPq) of Brazil (fund 306185/2020-2). Thanks to MHF Gomes, TM Soares and BA Machado for providing some pictures of XRF equipment.

(1) IEA-International Energy Agency, Transportation Biofuels-Renewable Energy in 2020-Analysis (IEA, Paris, 2020).

(2) E. Kirkby, in Marschner's "Mineral Nutrition of Higher Plants" (Academic Press, 3rd Edition, 2012), pp. 3-5.

(3) VD Fageria, J. Plant Nutr. 24(8), 1269-1290 (2001). DOI: 10.1081/PLN100106981.

(4) NK Fageria, MP Barbosa Filho, A. Moreira and CM Guimarães, J. Plant Nutr. 32(6), 1044–1064 (2009). DOI: 10.1080/01904160902872826.

(5) TAB Fenilli, K. Reichardt, OOS Bacchi, PCO Trivelin and D. Dourado-Neto, An. Akkad. bra. Sean. 79(4), 767–776 (2007). DOI: 10.1590/S00013765007000400015.

(6) V. Fernández and PH Brown, front row. Plant science. 4, 289 (2013). DOI: 10.3389/fpls.2013.00289.

(7) V. Fernández, E. Gil-Pelegrín and T. Eichert, Plant J. 105(4), 870-883 (2021). DOI: 10.1111/tpj15090.

(8) E. de Almeida, NM Duran, MHF Gomes, SM Savassa, TNM da Cruz, RA Migliavacca and HWP de Carvalho, X-ray spectroscopy 48(2), 151–161 (2019). DOI: 10.1002/xrs.3001.

(9) GT Costa Junior, LC Nunes, MHF Gomes, E. de Almeida and HWP de Carvalho, X-ray spectroscopy 49(2), 274–283 (2020). DOI: 10.1002/xrs.3111.

(10) ES Rodrigues, MHF Gomes, NM Duran, JG Cassanji, TNM da Cruz, AS Neto, SM Savassa, E. de Almeida, HWP de Carvalho, Front. Plant science. 9, 1588 (2018). DOI: 10.3389/fpls.2018.01588.

(11) MHF Gomes, BA Machado, JPR Marques, R. Otto, T. Eichert and HWP de Carvalho, J. Soil Sci. Plant 20(4), 2721–2739 (2020). DOI: 10.1007/s42729-020-00338-3.

(12) BA Machado, MHF Gomes, JPR Marques, R. Otto and HWP de Carvalho, J. Agric. Food Chemistry 67(47), 13010–13020 (2019). DOI: 10.1021/acs.jafc.9b05630.

(13) MHF Gomes, BA Machado, E. de Almeida, GS Montanha, ML Rossi, R. Otto, FS Linhares and HWP de Carvalho, J. Agric. Food Chemistry 67(44), 12172–12181 (2019). DOI: 10.1021/acs.jafc.9b04977.

(14) TNM da Cruz, SM Savassa, MHF Gomes, E. dos Santos, NM Duran, E. de Almeida, AP Martinelli and HWP de Carvalho, Environ. science. 4(12), 2367–2376 (2017). DOI: 10.1039/C7EN00785J.

(15) TNM da Cruz, SM Savassa, GS Montanha, JK Ishida, E. de Almeida, SM Tsai, J. Lavres Junior and HWP de Carvalho, Sci. Representative 9(1), 10416 (2019). DOI: 10.1038/s41598-019-46796-3.

(16) GS Montanha, ES Rodrigues, SLZ Romeu, E. de Almeida, AR Reis, J. Lavres Junior and HWP de Carvalho, Plant Sci. 292, 110370 (2020). DOI: 10.1016/j.plantsci.2019.110370.

(17) GS Montanha, ES Rodrigues, JPR Marques, E. de Almeida, AR Reis and HWP de Carvalho, Metallomics 12(2), 183–192 (2020). DOI: 10.1039/C9MT00237E.

(18) MGDB Lanza, VM Silva, GS Montanha, J. Lavres, HWP de Carvalho and AR dos Reis, Ecotoxicol. environment. Safety. 207, 111216 (2021). DOI: 10.1016/j.ecoenv.2020.111216.

(19) AR Reis, EHM Boleta, CZ Alvez, MF Cotrim, JZ Barbosa, VM Silva, RL Porto, MGDB Lanza, J. Lavres, MHF Gomes and HWP de Carvalho, Ecotoxicol. environment. Safety. 190, 110147 (2020). DOI: 10.1016/j.ecoenv.2019.110147.

(20) JHL Lessa, JF Raymundo, APB Corguinha, FAD Martins, AM Araujo, FEM Santiago, HWP de Carvalho, LRG Guilherme and G. Lopes, J. Cereal Sci. 96, 103125 (2020). DOI: 10.1016/j.jcs2020.103125.

Thainara Rebelo da Silva, Gabriel Sgarbiero Montanha, Camila Graziele Corrêa, João Paulo Rodrigues Marques, and Hudson Wallace Pereira de Carvalho work in the Nuclear Instrumentation Laboratory of the Agricultural Nuclear Energy Center of the University of São Paulo, Brazil. Direct contact: hudson@cena.usp.br