Source: UNIV OF WISCONSIN submitted to NRP
REAL-TIME MONITORING OF PESTICIDES IN AGRICULTURAL RUNOFF USING HOT SPOT NORMALIZED SURFACE-ENHANCED RAMAN SPECTROSCOPY
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1029922
Grant No.
2023-67021-39748
Cumulative Award Amt.
$299,934.00
Proposal No.
2022-08555
Multistate No.
(N/A)
Project Start Date
Jul 1, 2023
Project End Date
Jun 30, 2026
Grant Year
2023
Program Code
[A1511]- Agriculture Systems and Technology: Nanotechnology for Agricultural and Food Systems
Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218
Performing Department
(N/A)
Non Technical Summary
Pesticides, including herbicides, insecticides, and fungicides, are toxic chemicals that are used to control pests during crop production. According to the United States Geological Survey, a half million tons of pesticides have been applied to crops annually in the United States. Rather than staying in the fields, the applied pesticides can be washed intosurrounding water systems via agricultural runoff formed by rain, irrigation, and melting snow. Particularly, the pesticide loads in agricultural runoff can increase substantially during extreme rainfall events that are occurring more and more frequently because of the changing climate. Such agricultural runoff from the over 300-million-acre agricultural land in the United States is, arguably, the biggest non-point pollution source for lakes, rivers, and groundwater, which threatens the safety of aquatic ecosystems and human health, and the sustainability of agricultural production. Unfortunately, we have little knowledge about this non-point pollution because the current methods for quantifying pesticides in agricultural runoff are very expensive and time-consuming.In this proposed research project, we aim to develop an inexpensive sensing platform for real-time analysis of pesticides in agricultural runoff. As contaminated water flows through the sensing chamber, pesticides in water will be captured by the gold nanoparticles restrained in a bacterial cellulose scaffold, which generates fingerprinting Raman spectra for pesticide identification and quantification. To further improve the quantitative performance of the sensor, we will use light as an internal standard to reduce the variation of Raman intensities across the sensor surface as a result of the uneven distribution of gold nanoparticles. The performance of this proposed real-time sensing platform will be tested and optimized using artificial agricultural runoff spiked with the five most used pesticides and their major metabolites, and finally validated using real agricultural runoff collected from the Central Sands region in Wisconsin. Analysis of pesticides by theproposed sensing platform is much cheaper and faster than the current methods based on solid-phase extraction and liquid chromatography-tandem mass spectrometry. The proposed technology will allow people to capture the temporal and spatial variation of pesticide loads in agricultural runoff, which enables the stakeholders to make data-driven action plans for minimizing the adverse impacts of pesticides on agroecosystems.
Animal Health Component
30%
Research Effort Categories
Basic
30%
Applied
30%
Developmental
40%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
13352202000100%
Knowledge Area
133 - Pollution Prevention and Mitigation;

Subject Of Investigation
5220 - Pesticides;

Field Of Science
2000 - Chemistry;
Goals / Objectives
The overarching goal of the proposed research effort is to develop an inexpensive and field-deployable sensing platform forreal-time quantification of pesticides in agricultural runoff. To achieve this goal, we have developed a research plan that consists of two interconnected research objectives. Objective 1: Quantify typical pesticides and pesticide metabolites at environmentally relevant concentrations using hot spot normalized surface-enhanced Raman spectroscopy (HSN-SERS). In Objective 1, we will leverage the high SERS "hot spot" density of a bacterial cellulose plasmonic passive sampler and the exceptional reproducibility of HSN-SERS to quantify five parent pesticides and their major metabolites down to medium-to-high ng/L level. Objective 2: Develop and validate a flow-through sensing platformfor real-time analysis of pesticides in field samples. In Objective 2, we will integrate the plasmonic passive sampler and HSN-SERS into a flow-through sensing platform that enables real-time analysis of target pesticides in field samples collected from the Central Sands region in Wisconsin. We will also validate the performance of the proposed sensing platform using the standard analytical method based on liquid chromatography-tandem mass spectrometry.
Project Methods
Five pesticides (i.e., imidacloprid, thiamethoxam, clothianidin, atrazine, and metolachlor) and four of their major metabolites (i.e., imidacloprid-urea, de-ethyl atrazine, atrazine TCR, and metolachlor ESA) will be selected as the target analytes because they are the most used and detected pesticides in Wisconsin according to the field-edge groundwater monitoring annual report from WisconsinDepartment of Agriculture, Trade and Consumer Protection. A bacterial cellulose-based plasmonic passive samplerwill serve as the Raman intensity enhancer and hot spot normalized surface-enhanced Raman spectroscopy (HSN-SERS) will be used to improve the reproducibility of conventional SERS. A flow-through chamber for real-time pesticide analysis in agricultural runoff will be built. The SERS substrate will be fixed onto the bottom of a 6-mL acrylic plastic flow cell, which will be placed onto the sample holder of the Raman spectrometer (Horiba LabRAM HR Evolution). To obtain the limit of detection (LOD) for each pesticide, aqueous solutions containing 0, 0.001, 0.01, 0.1, 1, 10, 100, and 1,000 µg/L individual pesticides will continuously flow through the chamber at various flow rates controlled by a programmable peristaltic pump. The laser beam will be focused onto the surface of the SERS substrate and Raman spectra will be collected after 10, 20, 30, 40, 50, and 60 min of exposure to the water.For a specific experimental setup, the ratio between the intensities of Raman scatterings from the pesticides and the elastic scatterings from the laser amplified spontaneous emission(IRaman/IElastic) is proportional to the number of target analytes on gold nanoparticle (AuNP)surfaces. We hypothesize that the relationship between pesticide concentrations in the solution at equilibrium (Ceq) and IRaman/IElastic can be described by the adsorption isotherm models. To test this hypothesis, the IRaman/IElastic acquired from Raman measurement and Ceq acquired from liquid chromatography-tandem mass spectrometry (LC-MS/MS) measurement will be fitted using the Langmuir, Freundlich, and Polanyi theory-based equations. Association constants that describe the affinity between pesticides and AuNPs will thus be obtained. We hypothesize that the ratio between the association constants acquired above can be used to predict the selectivity of the sensing platform for one pesticide over the others. To test this hypothesis, a water sample spiked with all five pesticides and their metabolites (1 µg/L for each compound) will flow through the chamber. At each time point, the SERS spectrum of the mixture will be deconvoluted using the multivariate curve resolution-alternating least squares model that is commonly used to differentiate individual components in a complex Raman spectrum. The performance of the sensing platform for pesticide analysis will be validated using the water samples collected from the Central Sands and Fox River drainage basins in Wisconsin. Groundwater-fed streams in this region provide a broad spatial sampling scale (headwaters to confluence). They will be sampled repeatedly during growing seasons and drawn from groundwater that underlies different compositions of surrounding crop and non-crop agroecosystems. Water will be sampled from streams by pumping approximately 2 L of the stream (surface) water or water from piezometers (groundwater) into brown glass bottles for later analysis. The collected field samples will be analyzed using the standard method based on solid-phase extractionand LC-MS/MS (Agilent 1260 HPLC coupled to a 6460 Triple Quad MS). Outcomes of LC-MS/MS measurement will be compared directly to those obtained with the proposed HSN-SERS sensing platform.

Progress 07/01/23 to 06/30/24

Outputs
Target Audience: The PD presented research findings from this project to Wisconsin farmers at the Wisconsin Potato and Vegetable Growers Association Grower Education Conference in February 2024. The PD connected with and discussed the project with Carla Romano, a staff member from the Wisconsin Department of Agriculture, Trade and Consumer Protection (WI DATCP), at the Water@UW Think Tank Workshop in June 2024. The PD presented project-supported research findings to staff and scientists at the US EPA during the Innovations in Environmental Science Seminar Series in May 2024. The PD presented research findings from this project to representatives from state agencies, industry, academia, and the general public at the Water@UW Think Tank Workshop in June 2024. A Ph.D. student supported by this project presented findings to peers in academia and industry at the ACS Spring Conference in March 2024. Changes/Problems:The project's spending in the first year was delayed due to challenges in recruiting a Ph.D. student in Fall 2023. However, this issue was resolved when a new Ph.D. student joined the group in Spring 2024. Despite the delay, the project's progress was not significantly impacted, as the PD dedicated substantial time to conducting experiments, analyzing data, and processing results. What opportunities for training and professional development has the project provided?This project supported one Ph.D. student, who was directly mentored by the PD. The student met with the PD weekly and attended bi-weekly group meetings. Throughout the first-year of this project, the student received training in experimental design, implementation, data analysis, scientific writing, and presentation skills. The student gave an oral presentationtitled "Quantitative Insights into the Relationship Between the Concentrations and SERS Intensities of Neonicotinoids" at the ACS Spring 2024 National Conference in New Orleans, LA. Additionally, the student published a paper titled "Emerging Investigator Series: Quantitative Insights into the Relationship Between the Concentrations and SERS Intensities of Neonicotinoids in Water" in Environmental Science: Nano. This project also provided opportunities for the Ph.D. student to mentor two undergraduate students, who were trained to conduct rigorous scientific experiments and collect data under the Ph.D. student's supervision. How have the results been disseminated to communities of interest?The PD engaged with Wisconsin farmers, primarily potato growers who heavily rely on neonicotinoids as pesticides, by delivering a presentation at an outreach event organized by the UW-Madison Division of Extension. This event, the Wisconsin Potato and Vegetable Growers Association Grower Education Conference, was held in Stevens Point, Wisconsin, in February 2024. Additionally, the PD presented the project's findings to staff and scientists at the US EPA during the Innovations in Environmental Science Seminar Series in May 2024. Further outreach included a presentation at the Water@UW Think Tank Workshop, where the PD shared research findings with a diverse audience, including representatives from state agencies, industry, academia, and the general public. The PD's group also participated in the Engineering Expo, organized by the UW-Madison College of Engineering, where they demonstrated a handheld Raman spectrometer for pollutant analysis. This expo primarily engaged K-12 students from Dane County, Wisconsin. What do you plan to do during the next reporting period to accomplish the goals? We will continue to lower the limit of detection (LOD) for pesticides using AuNP/BC plasmonic passive samplers. Preliminary data indicates that the LOD can be further enhanced by reducing solution pH to facilitate citrate layer desorption or by adding halides that act as a bridge between pesticides and gold nanoparticles. We will conduct a systematic study to explore how these factors influence SERS sensor performance, not only for the three neonicotinoids we've focused on but also for the other pesticides and metabolites listed in our proposal. Additionally, we will investigate the underlying mechanisms of these improvements using advanced techniques, such as TEM EELS, to better understand gold nanoparticle surface chemistry. We aim to develop a flow-through device for real-time monitoring of pesticides using AuNP/BC plasmonic passive samplers. Preliminary findings show that increasing the sample volume improves the adsorption of neonicotinoids onto AuNP surfaces, ensuring more molecules interact with the concentrated AuNPs in the samplers. This suggests that a flow-through device could enhance SERS sensor performance by continuously supplying large volumes of water, simulating real-world conditions where water volume is not limited. We will continue refining and optimizing data processing algorithms to extract quantitative information from SERS spectra. In the first year, we demonstrated the potential to retrieve quantitative data from noisy or background-distorted spectra of typical neonicotinoids. Moving forward, we will further customize pre-processing procedures for different pesticides and apply a range of multivariate statistical tools, such as partial least squares regression and principal component analysis. Additionally, as our dataset grows, we will explore the application of machine learning models for more robust analysis. We will conduct several field campaigns in Wisconsin's Central Sands region to collect agricultural runoff samples potentially contaminated with the pesticides listed in our proposal. These field sampling efforts, supported by co-PD Groves, will allow us to further refine the sensors and data processing algorithms, focusing on minimizing interferences from real-world field samples.

Impacts
What was accomplished under these goals? During the first year of this project, we focused primarily on achieving our firstgoal: quantifying five pesticides and their metabolites down to medium-to-high ng/L concentrations using plasmonic passive samplers. First, we synthesized, optimized, and characterized gold nanoparticle/bacterial cellulose (AuNP/BC) passive samplers designed for surface-enhanced Raman spectroscopic (SERS) applications. These samplers were tested for detecting five pesticides--imidacloprid, clothianidin, thiamethoxam, alachlor, and atrazine--in both deionized water and groundwater-sourced tap water. While SERS is known for its high sensitivity, its quantitative performance was initially limited due to an unclear relationship between pesticide concentrations and corresponding SERS signals. To address this, we conducted fundamental studies to better understand how SERS could be applied for the accurate quantification of neonicotinoid insecticides in water. We selected two widely used neonicotinoids--imidacloprid (IMD) and clothianidin (CLO)--as target analytes and studied their adsorption onto gold nanoparticle surfaces, a critical step for enhancing SERS signals. By investigating the adsorption equilibrium between these pesticides and gold nanoparticles, we modeled their behavior using classic adsorption isotherms, such as the Freundlich and Langmuir models. This allowed us to correlate SERS intensities with pesticide concentrations, enabling more accurate calibration curves. Our findings revealed that the relationship between SERS signals of neonicotinoids and their concentrations in water samples can be perfectly described using a Langmuir adsorption isotherm. Before the SERS "hot spots" were fully saturated, the SERS signals increased linearly with the adsorbed amount of neonicotinoid molecules within SERS "hot spots". This quantitative and mechanistic study was successful because of the use ofhigh-performance liquid chromatograph as a complementary analytical tool and hot spot normalized SERS to improve measurement reproducibility. To improve the detection limit down to high ng/L levels, we developed a custom data processing algorithm that correlates the concentration of neonicotinoids with their SERS spectral patterns. The algorithm assumes that SERS patterns shift as the molecular orientation on AuNP surfaces changes. Using this method, we extracted quantitative data even when background interference distorted the spectra.With citrate-coated gold nanoparticles as substrates, we tested the IMD, CLO, and thiamethoxam (THX)across a range of concentrations in both deionized and tap water, and achieveddetection limits as low as 0.001 μM. We also tackled challenges in SERS analysis with complex water matrices, including tap and lake water, by employing background subtraction techniques to reduce interferences. This process significantly enhanced the signal-to-noise ratio, allowing for clearer identification of the characteristic Raman bands of the three target neonicotinoids, even at environmentally relevantconcentrations. Our method demonstrated highaccuracy in detecting all neonicotinoidsacross multiple samples. In our continued development, we assessed factors like pH and the presence of other pesticides, confirming that the method works reliably in real-world conditions and whenmultiple pesticidescoexist inenvironments.

Publications

  • Type: Journal Articles Status: Published Year Published: 2024 Citation: Liu, J. Lazarcik, and H. Wei,* Emerging investigator series: quantitative insights into the relationship between the concentrations and SERS intensities of neonicotinoids in water, Environmental Science: Nano, 2024, 11, 3294-3300