Popular electrochemical nanosensors for nitrate detection include metal/metal oxide nanoparticles (NPs), graphene, carbon nanotubes (CNTs), electropolymerized films, and combinations of these. Most materials in this category act as electrode modifiers, as bulk copper (Cu) and other common conductors are poor nitrate detectors near neutral pH. These contacts also tend to degrade and experience interference with other molecules and ions, especially nitrite (NO₂ ⁻) and chloride (Cl⁻), without proper modifiers (Li et al., 2019). Copper nanoparticles (CuNPs) and nanowires in combination with various substates have nitrate detection ranges in the sub ppb (Essousi et al., 2019; Stortini et al., 2015). CuNPs deposited on graphene catalyzed nitrate reduction to ammonia with analysis by differential pulse voltammetry have been shown to have a detection limit of 0.49 ppm (Wang et al., 2018). More recently, Cu nanowires were grown by galvanic deposition to detect nitrate by linear sweep voltammetry. This cost effective and stable growth technique enabled measurement of nitrate concentrations as low as 0.56 ppm and performed well in natural water samples (Patella et al., 2021). An undesirable characteristic of these sensors is that they require a pH between two and three for proper electrocatalytic reduction of nitrate. Silver (Ag) nanostructures have low limits of detection (as low as 24 ppb) near neutral pH and experience few interferents but are limited to concentrations under 62 ppm (Chen Legrand et al., 2017; Hu et al., 2013). Potentiometric sensors utilize ISEs including multiwalled carbon nanotubes (MWCNTs), graphene, and polypyrrole (Cuartero et al., 2018; Schwarz et al., 2018; Pięk et al., 2016; Gil et al., 2024). Work with organic nanotubes has also shown low detection limits of 0.02 ppm (Kundu, 2023). Most ISEs are based on three nitrate ionophores: quaternary ammonium, nitrate ionophore V and VI, and tridodecylmethylammonium nitrate (TDMAN). Very low nitrate concentrations between 10⁻⁷–10⁻² ppm have been detected by measuring the potential difference between the ISE and a reference electrode (Singh et al., 2022; Liu et al., 2020). Electrochemical sensors can be complicated in their construction, but recent work has highlighted alternative constructions with linear ranges between 10 and 100 ppm with a detection limit of approximately 2 ppm (Concepcion et al., 2024) while other work has highlighted possibilities for environmental sustainability with linear ranges of 1–100 ppm (Sarwar Inam et al., 2023).

Spectroscopic nitrate detection is the most popular method in the lab due to the highly precise nature of spectrometers and other optical instruments. By performing a nitrate to nitrite reduction via a Griess assay, nitrate concentrations as low as 10⁻⁵ ppm can be measured with high resolution fluorescence spectroscopy (Yang et al., 2015). Vanadium (III) chloride (VCl₃) in hydrochloric acid (HCl) has also been used with limits of detection as low as 0.006 ppm, though sensing times range from 3–60 min (Garcia-Robledo et al., 2014; Wang et al., 2016). CNTs and CuNPs and been utilized in tandem with optical fibers for both surface plasmon resonance and ultraviolet-visible (UV-Vis) absorption detection techniques (Zhang et al., 2019; Parveen et al., 2017; Moo et al., 2016). Surface-enhanced Raman spectroscopy also offers large linear detection ranges spanning multiple orders of magnitude and low limits of detection of ppm or sub ppm (Gajaraj et al., 2013; Li et al., 2024). While these measurements take only milliseconds, have very low detection limits, and are highly reproducible, they require expensive optical systems and calibration training.

Biosensors based on nitrate reductase (NR) enzymes can detect nitrate with high specificity and sensitivity at neutral pH through adsorption onto electrodes. The primary drawbacks of NR sensors are their high cost and the low temperature required for storage (Singh et al., 2022). Some methods for improving the sensitivity and stability of NR sensors through biological enhancements include incorporating NR from plants and fungi (Kalimuthu et al., 2015; Kalimuthu et al., 2021), combining biosensing elements with nanomaterials such as carbon nanotubes and zinc oxide (ZnO) nanostructures (Can et al., 2012; Ahmad et al., 2017), and utilizing whole-cell organisms (Machado et al., 2022). While nanomaterials can improve the capabilities of biological nanosensors, enzymatic sensors are still susceptible to degradation under environmental conditions (Singh et al., 2022). To overcome the limitations of enzymatic sensors, there is a need for more research focused on developing novel designs and synthesis methods that can minimize degradation under environmental conditions.

While potentiometric and optical sensors have very low detection limits, they require sample preparation including pH balance and control calibration. New chemiresistors, capacitors, and FETs avoid these problems, showing great aptitude as nitrate detectors largely due to their ‘lab-on-a-chip’ design. In one recent study, graphene nanowire was created by melting high density polyethylene (HDPE) between two bulk Cu contacts. By measuring current-voltage response, nitrate concentrations between 50–5,000 ppm were successfully determined (Ahmadi et al., 2021). FETs in particular can provide extremely low detection limits (45 ppb) with no interfering species (Minami et al., 2016). By replacing the gate metal on a standard FET with a nitrate sensing material, researchers created chemically sensitive FETs (CHEMFETs) in the early 1970s (Janata, 2022). The amount of nitrate present modulates the electric field in the gate, changing current flow across the device. As in all transistors, various forms of CHEMFETs allow minute field changes in the gate to produce large currents through the device. This sensitivity allowed for ppb detection in some ion selective FETs (ISFETs) and organic FETs (OFETs) over the past decade (Minami et al., 2016; Kim et al., 2020a). Other ISFETs have used nitrate selective membranes on chemical vapor deposited graphene and nitrone coated polyvinyl chloride for detection with impressive results (Kim et al., 2020b; Chaisriratanakul et al., 2020). These electrical sensors can easily be integrated into ‘internet of things’ devices and are likely to see largescale field deployment with wireless data communication (Agir et al., 2021; Alahi et al., 2018).

Different user groups will have different requirements relating to form factor (i.e., size, shape, and other physical considerations). Limnologists or oceanographers will likely find a device that falls into the probe or sonde category, as something 2-liter bottle sized, amenable to their needs, while a community member or farmer interested in well water testing will likely be interested in something less bulky. The nanomaterial-based sensors described previously in this paper all contain small components. However, the sensors integrated system can exist across a wide range of device sizes. The addition of extra components and detectors can hamper portability, and the devices best suited for small form factors would be those that have integrated readout electronics (and display) on the sensor, and do not rely on bulky external components to read results. The intended audience and uses of device, along with the necessary system components, will play a large part in shaping the appropriate form factor of the device.

Device robustness is important to consider during the development as this will affect the appropriate uses and audience. The robustness of a sensor is important for technical applications like wastewater testing and extended field sampling events by scientists and water managers. Factors such as the length of time until failure, number of measurements until required calibration or maintenance, structural suitability for long term deployment, and accuracy, precision, and sensitivity with time all compose the idea of robustness. These factors will need to be investigated in different capacities depending on the intended application. Sensors based on nanomaterials, such as nanotubes and nanoparticles deposited on electrodes, show promise with respect to resistance to structural damage and chemical degradation, as such devices could still function if a certain portion of the nanomaterial is damaged or degraded. Other users of nitrate sensing devices may not require the same level of robustness, and sensors that excel in alternative categories may be more appropriate. Electrical FET sensors based on a single sheet of nanomaterials are among the least robust with respect to structural damage and chemical degradation, as damage of the nanomaterial can easily render the device unusable. In general, the anticipated device users need to be consulted regarding electrical FET design and packaging, as these devices are among the most sensitive.

Cost is an important consideration in the development of any device, although there is a market for nitrate sensing devices that span a broad price spectrum. Scientists and wastewater managers may need relatively expensive sensors, driven by requirements for low detection limits and high precision. In contrast, community-based science projects or water monitoring will likely require more affordable sensor options with detection limits and precision being less of an integral issue. Spatiotemporal coverage issues will also play a role in dictating the appropriate price point. If a scientific or regulatory question required occasional sampling or sampling in only one place, then devices that are expensive but reliable and rugged would be appropriate. Other applications may be willing to sacrifice accuracy and precision for smallness and inexpensiveness because of the need to deploy many sensors at one time. While it can be difficult to discern the exact cost of a device still in a proof-of-concept stage, the overall cost can be estimated with the given components. Sensors containing integrated electronic readout circuitry, which can be mass-produced, would end up being more cost effective than those requiring additional equipment for every single measurement. Additionally, sensors requiring materials that are easy to procure in scalable quantities or are commercially available would lend themselves to a lower final cost. A variety of devices across a price continuum are needed to meet the unique needs of users related to nitrate sensing.

Much of the literature surrounding the development of nitrate sensors neglects the user interface, but the usability and interfacing of a device is important for its transition to the field. Although some work does consider the user interface (Agir et al., 2021; Alahi et al., 2018), the majority of current devices fail to mention the user experience or the collection of data by a new user in the field. The difficulty and complexity of interface should reflect the intended audience a. The interface for scientists could include more data options and even statistics, while an interface for the general public should be simple, possibly even including easy to interpret colors and icons to be usable across literacy and languages. A major expense associated with sensors installed in the environment is associated with operation and maintenance (e.g., checking, resetting, and recalibrating) in the field because of the travel time and human hours expended. Thus, technicians, interns, and students often fulfill these duties, and the user interface needs to consider the appropriate complexity for people at these career and educational stages. In order to complete a transition from a laboratory tested device to a field utilized product, data and results must be available to the users of the sensors and therefore user interface must be considered.

The necessary sensitivity of a device is directly related to the purpose of the sensing effort. Falling within the category of sensitivity are also important considerations of accuracy, precision, and reproducibility. Sensitivity is important for ecological applications in less impacted areas, especially in cases like alpine lake ecology monitoring, or in the low nitrate concentrations of the open ocean. Accuracy and reproducibility are critical in regulatory situations, which are typically associated with human-impacted areas and therefore less likely to need low detection limits. Instead, they need reliable measurements which can meet regulatory standards and legally binding agreements. Less precise sensors can also still be relevant to community science projects or personal home monitoring, such as a sensor that simply indicates the presence or absence of nitrate to the user. Electrical devices that are 2-dimensional material-based FET sensors may be among the sensors that can provide the best sensitivity in the low ppb range, however the consumer and intended use are important to consider when developing highly sensitive devices. Future work will need to consider the tradeoffs between sensitivity, accuracy, and robustness and would benefit from a comprehensive evaluation of the device and user.

To link prospective nitrate sensors to users, we identified four broad user groups: wastewater managers, farmers, scientists, and community members. For each user, we assigned numerical scores to each FOCUS parameter on the scale of 1–5, where one is least important to the user and five the most, and plotted these in Figure 2A. Wastewater managers prioritize operational robustness (O = 5) the most because monitoring systems must function reliably in harsh environments with varying conditions. Sensitivity (S = 4) follows next as accurate measurements are needed to ensure regulatory compliance and early detection of relevant issues. Cost (C = 3) takes middle priority as while budgets matter, reliable equipment justifies higher expenses. User interface (U = 2) is less critical since staff are trained in complex systems, and form factor (F = 1) ranks lowest as treatment plants have adequate installation space. Scientists place S highest because research demands precise, reliable data for experimental validity. U is also crucial for detailed control over measurements and comprehensive data access. O ranks third as scientists work in a mix of controlled and uncontrolled environments, while F is less important as lab setups are adaptable to research needs and end goals. C ranks lowest because most scientists value data quality data over quantity, justifying higher equipment expenses. This contrasts with farmers, whose priorities center on practical and economic factors, with C ranking highest. O follows next as their equipment must withstand outdoor conditions and physical impacts. F and U share lower priority. While portability and ease of use matter, they are secondary to cost and robustness. S ranks lowest since basic accuracy typically meets agricultural monitoring needs. Community members prioritize C as the main barrier to adoption in voluntary monitoring situations. U ranks second as systems must be accessible without technical training, while F takes middle priority to as portability and power needs are often important for community engaged sampling. O ranks lower as short-term durability often suffices for citizen groups, and S is least important since community monitoring typically emphasizes large numbers of less precise data.

We applied the FOCUS analysis to all relevant sensors cited in this perspective (Supplementary Figure S-2). The analysis of the entire dataset of 37 entries indicates that U and S show a clear and consistent negative correlation (Supplementary Figure S-3). This reflects a fundamental design challenge: Highly sensitive sensors require more complex controls, calibration options, and detailed readouts, which lead to less user-friendly interfaces. The relationship demonstrates the inherent challenge of balancing sophisticated measurement capabilities with user friendliness. We performed a cluster analysis on the FOCUS ratings and, as shown in Figure 2B, it reveals three clusters. We plot the results as functions of two principal components (PC1 and PC2), which is a dimensionality reduction technique to reduce the number of variables while preserving as much variance as possible in the data. PC1 is measuring a trade-off between sensitivity/operational robustness versus user interface/form factor, while PC2 compares cost/form factor versus operational robustness. Cluster 1, with positive PC1 and PC2 values, is a collection of sensors with high sensitivity, good operational robustness, strong form factors, but more complicated user interface and higher cost. Cluster two includes sensors that have moderate scores across the FOCUS board, while cluster three comprises sensors with strong operational robustness at lower cost.

Next, we performed a similarity analysis to find which of the 37 entries align best with preferences of each of the end users, based on the profiles for them in Figure 2A. Our results implementing a recommendation system based on cosine similarity reveals the best matches for each user type and are indicated in Figure 2B relative to the three main clusters. Gajaraj et al. (2013) describes a SERS-based nitrate detection system using commercially available gold nano substrates. This scores highest for both wastewater management and scientific research, offering precision equivalent to ion chromatography, suitable detection range (1–100 mg/L), and reliability when faced with interfering compounds. Its quick analysis time, minimal preparation needs, and non-destructive approach support high-throughput monitoring and research requirements. Despite initial equipment costs, the lower per-test expenses and reduced preprocessing make it cost-effective for both routine wastewater testing and scientific studies.

However, while technically sophisticated, the SERS-based approach is less suitable for farmers and community members primarily due to its complexity and operational requirements. It demands complex sample preparation and advanced Raman spectroscopy instrumentation. While highly sensitive, its technical requirements added to high equipment costs make it impractical for non-technical users. Instead, we find the sensor described in Ahmad, et al. (2017) to be ideal for farmers, featuring durable zinc oxide nanorods for field conditions, rapid response time, and minimal sample preparation. The reported detection range suits agricultural needs, covering trace to excess nitrate levels. With proven reliability in real samples, interference resistance, and month-long stability, it enables quick on-site testing for timely fertilizer and irrigation decisions. And finally, Machado et al. (2022) describes the best sensor for community users, with affordable materials and simple construction, while also maintaining good sensitivity. Its cartridge-based design, minimal sample preparation, and Arduino-based system make it user-friendly for citizen scientists. With reliable reproducibility and real-time measurements, it enables effective community water monitoring without requiring technical expertise.