Enabling Smart Agriculture through Sensor-Integrated Microfluidic Chip to Monitor Nutrient Uptake in Plants

The soil microenvironment greatly influences a plant’s ability to absorb nutrients and germinate. Sensing these changes in soil medium is critical to understand plant nutrient requirements. Soil being dynamic represents changes in nutrient content, element mobility, texture, water-holding capacity, and microbiota which affects the nutrient levels. These minor changes affect the plant in early growth and development and studying these changes has always been challenging. Microfluidics provides a platform to study nutrient availability and exchange in small volumes of liquid or media resembling plant microenvironments. Here, we have developed a novel microfluidic chip-embedded molecular imprinted sensor for sensing nitrate and phosphate in the media. For data acquisition and recording we have implemented a potentiostat controlled via a microcontroller allowing data storage and transfer via a long-range radio module (LoRA). The microfluidic device’s functionality was validated by germination of the legume crimson red and recoding the nitrate and phosphate levels in media for 7 d. The MIP-based sensor measures nitrate and phosphate, in the range from 1 to 1000 mM. The accuracy of detection for nitrate and phosphate showed 99% and 95% respectively. The chip coupled with MIP based sensor for nutrient analysis serves as a platform technology for studying nitrate and phosphate nutrient exchange and interaction. This chip in the future can be implemented to study plant deficiencies, drought resistance, and plant immunity.

The complex and heterogeneous nature of the soil microbiota provides a dynamic biochemical habitat that influences several plant functions, such as root growth, respiration, and nutrient exchange.Among the many nutrients present, nitrogen and phosphorus play a critical role in plant growth and development, including protein synthesis, photosynthesis, ATP(Adenosine Tri-phosphate) production, signaling metabolic changes, and converting biochemical reactions.Although these nutrients exist in the natural ecosystem, they are primarily exploited as fertilizers in agriculture to improve soil fertility and plant growth.Nevertheless, most fertilizers applied to the soil are lost to leaching or runoff rather than absorbed by plants. 1 Hence, real-time monitoring of these nutrients in fabricated microenvironments can allow a better understanding of root uptake efficiency, thereby allowing better regulation of fertilizer use and reducing the environmental impact.Traditionally, soil nutrient analysis is carried out in laboratories that involve complex, timeconsuming procedures. 2 Despite their accuracy, laboratory techniques do not provide real-time monitoring of dynamic biochemical variations.This poses a challenge to plant phenotype researchers, as plant biochemical concentrations related to nutrient uptake fluctuate at a given time, influencing its mechanisms and metabolism.Plant phenotyping evaluates plants' physical characteristics, like growth, development, adaptation, yield, quality, tolerance, resistance, and measurement of like parameters. 3,46][7][8][9] Hence researchers need a low-cost real-time platform to understand various phenomena associated with plant germination, growth, metabolism, nutrient uptake, and survival.7][18][19][20] The high throughput of microfluidic platforms enables them to deliver microliters of fluid to specific outputs, making in situ monitoring of minute specimens possible. 21leklett et al. discussed the various applications of microfluidic devices, including simulating soil heterogeneity and studying nutrient interactions. 22Similarly, microfluidics and imaging technologies have demonstrated outstanding potential for advanced plant phenomics.For example, attempts to visualize root growth and cellular behaviors such as host-pathogen interactions were made by Busch et al. 23 High throughput imaging has also been implemented with microfluidics to understand plant response to drought and observe plant root and shoot development. 24,25Phenotype and genotype interactions of Arabidopsis thaliana roots were demonstrated using the RootArray device. 26Current microfluidic devices are fabricated using advanced photolithography techniques, which are tedious, time-consuming, and employ expensive machines. 27hese methods utilize expensive chemicals and reagents and in addition, require a specialized cleanroom facility that is not typically available.Hence, we present a low-cost, facile approach to fabricating sensor-integrated microfluidic devices to study plant-nutrient interaction and dynamic nutrient uptake pathways using MIP (Molecular Imprinted Polymer) based electrode printing and polymer sheet layering.MIP provide high selectivity when integrated with electrochemical sensors, The high selectivity of MIP based sensor is attributed to the selective binding potential of the template and target molecule.Some of the conventional sensors for assessing nitrate and phosphate are listed in Table I.These techniques have shown low accuracy and range of detection and therefore MIP based technique offers much more sensitive and accurate output.The chip with MIP based sensor for nutrient analysis serves as a platform technology for studying nitrate and phosphate.
For data processing and transfer the developed chip is connected to a potentiated and long-range radio module (LoRa) controlled by a microcontroller for real-time data acquisition using low power.Furthermore, we carried out biochemical tests to evaluate the performance of the developed microfluidic chip as compared to the conventional method or analysis, and the device was assessed by monitoring legume nutrient uptake and plant phenotypic traits.Incorporating the sensors in a microfluidic chip allows for the study z E-mail: vkamat@fiu.edu= These authors contributed equally to this work.
ECS Sensors Plus, 2023 2 043201 of nutrient and plant microenvironments which serves as a platform technology for a wide range of agriculture and soil-based studies.

Materials
All the reagents, including electrode preparation and testing solutions, were prepared in Millipore distilled water (17.8 mΩ).Analytical grade acetone, isopropyl alcohol (IPA), hydrochloric acid (HCl), sulfuric acid, Anthrone reagent, sodium nitrate, di-hydrogen phosphate, poly-pyrrole, agar, and Murashuge Skoog media, were purchased from Sigma Aldrich.Microfluidic chips were prepared using 25 × 75 × 1 mm glass slides from Thermo Fisher Scientific and were cleaned with 70% IPA, treated with DI water, and dried under nitrogen gas before use.Sheets of Polydimethylsiloxane (PDMS) sheets of approximately 1 mm thickness were fabricated cut and layered to fit the device.PDMS sheets offer advantages due to their less wastage and cost-effectiveness when fabricating small and large devices.

Methods
Fabrication of microchip.-Fabrication of the microchip was done using a prior protocol described by Kamat et al. 34,35 In brief, 10:1 (Base: curing agent) PDMS was mixed, degassed for 30 min under a vacuum, and poured into a mold.The base serves as a precursor silicone and the curing agent works as a crosslinker used for fabricated PDMS chips.The mixture was allowed to cure at 70 °C for 3 h before being cut into strips.Glass slides were cleaned using 70% IPA and then deionized water remove additional impurities, such as grease and smaller foreign particles.A three-electrode configuration was then patterned onto a glass slide.PDMS sheets 2.5 mm thick were sandwiched between the glass slides, whose outer edges were treated with a Corona plasma torch.The thickness can be tailored according to the seed type, allowing the sowing of most seeds for plant phenotype studies.The PDMS sides of the device were then punched with holes (0.5 microns) for introducing the media and collecting spent media.Once the glass slides were bound, the cavity was filled with 1% agarose and 1/4 strength MS media (Murashuge Skoog media, Sigma Aldrich) to support plant growth.The media was then passed into the chip using a suction syringe.
Fabrication of sensor.-Electrodeswere fabricated and applied to glass slides using the screen-printing method.The stencil template for the three-electrode design was created using SolidWorks, then printed onto transferable vinyl adhesive using Silhouette Cameo and Studio software version 4.3.263.The adhesive was attached to a glass slide and smoothed to remove air bubbles.The printed electrodes consisted of silver/silver chloride (Ag/AgCl) (60:40) and carbon graphene (Gr), with the first layer being Ag/AgCl and the second layer being Gr.After applying the first layer of Ag/AgCl, the glass was dried at 70 °C for 2 h and then allowed to cool dry overnight.The second layer of carbon graphene was applied similarly using the screen-printing method and then dried at 65 °C for 1 h.After drying, the validity of the fabricated device was tested using cyclic voltammetry of 5 mM ferri-ferrous solution (−1 V to 1 V).Nitrate and phosphate-doped polymer sensors were prepared using electrochemical methods.A 1:1 ratio of 1 M of poly-pyrrole (Ppy) and 0.5 M of nitrate was mixed in a 5 ml vial, where the solution was purged with nitrogen for 15 min to displace oxygen.Pyrrole is photosensitive and susceptible to contamination with oxygen; hence stored in lightsensitive vials and refrigerated.The doped polymer membranes were prepared using electrochemical methods.The molecularly imprinted polymer (MIP) was electro-synthesized using a PGSTAT in ten cycles at a scan rate of 50 mV s −1 while applying a constant sweep voltage from 0-0.9 V.During polymerization, as the polymer binds, a conducting polymer matrix is obtained with nitrate molecules imprinted within it on the surface of the working electrode.Then elusion is carried out using 70% ethanol to extricate the nitrate molecule from its entrapment, thereby creating a surface complementary to its shape and function (Fig. 1).
On the carbon working electrode, during electrodeposition, nitrate molecules migrate towards polypyrrole and get entrapped on the polymer matrix.Then the molecules are extracted from the conducting polymer to produce a surface that is complimentary in shape and functionality to nitrate by elution.Then the electrochemical detection  ECS Sensors Plus, 2023 2 043201 of nitrate by the MIP sensor was performed by placing 5 μl of nitrate solution at the working electrode for rebinding.Similarly, the phosphate sensors were prepared under the same conditions, utilizing a 1:1 ratio of polypyrrole and potassium dihydrogen phosphate.A wireless monitoring system was used to collect data in realtime using the fabricated sensor device (Fig. 2).As a part of the development of the electronic module, the microcontroller was interfaced with the LoRa radio module and dipole antenna to facilitate communication with the cloud platform. 36The chip was then connected to a potentiostat, controlled by a microcontroller (Arduino chip) for data acquisition and processing.The data was then transmitted to a PC module (Fig. 2).
Assessing plant growth parameters.-Toanalyze the plant growth parameters in a microfluidic chip, we chose Crimson Red Clover due to its relatively small seeds, considerable growth within a short time frame, and agricultural importance.Legume crops are popular among intercropping operations and crop rotations because they provided nitrogen fixation improving the overall health of the soil and reducing expenses on inputs for increased growth and yield.The experimental design consisted of two treatment groups, each containing ten seeds: a treatment group containing ten conventionally grown seeds and a treatment with ten seeds grown in the microfluidic chip.Conventionally, seeds are moistened on a biostrate fabric (cotton gauze) and observed after 24 h for root/shoot development.Plant roots and shoots are generally examined and analyzed by removing them from their natural state, which damages them.By observing the microfluidic chip under a stereomicroscope without disturbing the microenvironment, root shoots can be analyzed without being removed or damaged.
Assessing total chlorophyll, total nitrogen, and total soluble sugar.-Chlorophylland nitrogen are two of the most vital components of the plant.During photosynthesis, chlorophyll absorbs energy from the Sun and transforms it into chemical energy for the plant.The chemical structure of each chlorophyll molecule contains a porphyrin (tetrapyrrole) nucleus with a chelated magnesium atom at the center and a long-chain hydrocarbon (phytyl) side chain attached through a carboxylic acid group. 31In this method, chlorophyll is extracted in 80% acetone and the absorption peaks at 663 nm and 645 nm are read in a spectrophotometer.The total nitrogen content of the sample was obtained by the wet oxidation digestion method.The sample tissue was weighed and placed in pre-cleaned, autoclaved borosilicate tubes and rinsed with 0.1 M HCl.A 2 ml solution of concentrated sulfuric acid was added, and the tubes were kept at 300 °C for 2 h.Then 5-8 drops of 30% hydrogen peroxide were added which catalyzes to fasten the reaction instead of adding Se, Hg, and Cu as catalyst and incubated again at 300 °C for 10 min.After the extract became colorless, 25 ml of DI water was added for Total N analysis using Vapodest 500 instrument.Determination of total carbohydrates was carried out by the Anthrone method, which assesses sugar in seeds.Carbohydrates are first hydrolyzed into simple sugars using dilute hydrochloric acid.In a hot acidic medium, glucose is dehydrated to hydroxymethylfurfural.This compound forms a red to green-colored product with Anthrone giving an absorption maximum at 630 nm.

Results
Characterization of sensor.-Thedevice was prepared in-house with cost-effective methods that can be scaled up for large-scale production.The electrodes were tested in a ferrous solution before electro-polymerization. Figure 3 shows electrodes prepared using screen printing methods.The working and counter electrode consisted of printed Gr, and the supporting reference electrode featured an additional Ag/AgCl layer.After fabrication, electrochemical characterization was performed using ferrous solutions of 1 mM, 2 mM, and 5 mM, as shown in Fig. 4.
Voltametric response of doped PPy sensors.-Figure 5 shows the cyclic voltammograms of doped PPy sensors between voltage ranges of −1 V and 1 V at 1 mM concentration of NaNO 3 − and KH 2 PO 4 at a scan rate of 50 mV s −1 .The graphs demonstrate the sensors to be electroactive to the concentrations with oxidation peaks at approximately 0.6 and 0.7 V respectively.Subsequently, the voltage at the oxidation peak was chosen to deliberate the calibration test with nitrate solutions from 1 micromole to 1 millimolar.
Calibration plots of doped PPy sensors.-Theresponse of the doped sensors was tested for nitrate and phosphate ions in respective calibration solutions.Figure 6 shows the calibration plot for nitrate   ECS Sensors Plus, 2023 2 043201 and phosphate, created by extrapolating the peak currents for each concentration.The linear fit to the nitrate (Eq. 1) and phosphate (Eq.2) forms are as follows: Nutrient sensing and real-time monitoring.-Thedoped PPy sensors were employed to measure phosphate and nitrate within the growth medium for seven days.Nutrients were replenished every five days in the microfluidic root chamber after nutrient uptake over seven days.The plant uptake of nitrate was 700 μM in 3.5 d compared to phosphate, which was 30 μM (Fig. 7b).This could be because the seedling requires more nitrate during the growth phase, which leads to increased nitrate uptake. 37,38In the post-5-day period, we see a similar trend with higher nitrate consumption.Nitrate and phosphate sensors were integrated into microfluidic root chambers and immersed in agar nutrient media.On day seven, we observed mature shoot and root systems (Fig. 7a).Accordingly, the chip could detect nitrates and phosphates for up to seven days, providing realtime nutrient uptake data.A humidifying environment with 85% relative humidity was maintained to prevent evaporation loss.
Assessing plant growth parameters.-Toanalyze the plant growth parameters in a chip, we chose Crimson Red Clover due to its relatively small seeds, considerable growth within a short time frame, and agricultural importance as a legume and ground cover.The experimental design consisted of two treatment groups, each containing ten seeds: a treatment group containing ten microfluidic chip-grown seeds (set 1) and control seeds (set 2).The conventional growth process involves moistening seeds on biostrate fabric for germination and evaluating for root/shoot development, total chlorophyll, total sugar, SPAD count, and total nitrogen post-7 d (Fig. 8).The sensors demonstrated the ability to measure in real-time nutrient uptake of legume plants in the early stages of their growth.The graph below shows marginal change as compared to conventionally  grown plants, which shows that the microfluidic chip is successful in monitoring N and K nutrients safely and plants are healthy and show no side effects.The chip is successful in maintaining comparable microenvironment and liquid and gas exchanges mimicking real-world scenarios.Although a 10% increase in germination for the microfluidic plants was observed as compared to the control.Moreover, total nitrogen, chlorophyll, and soluble sugar were similar in both groups but the SPAD avg was higher in microfluidic-grown plants.

Conclusions
In summary, we presented and demonstrated a low-cost, simple method for fabricating multi-ion selective sensors integrated into the microfluidic chip.The device fabrication was achieved by layering PDMS sheets between glass slides to obtain the desired microfluidic root chamber.The printed sensors measured in situ nitrate and phosphate concentrations inside the growth medium as legume plant roots grew within the device.The selective sensors demonstrated high sensitivity and can continuously monitor nutrients for approximately 7 d.The high selectivity of MIP based sensor is attributed to the selective binding potential of the template and target molecule.The preparation of MIPs involve polymeric structures formed by the target molecule (template), the monomer selected accordingly, the cross-linker, and the initiator, works to create highly specific cavities that can recognize the template with high selectivity.
Overall, we demonstrated that legume plants could be grown in uniquely fabricated microfluidic chips without hindering typical performance and growth.This is possible as the developed microfluidic chip enables precise control of fluid flow within plant systems, allowing accurate measurements of vital parameters such as nutrient availability, water stress, and disease presence.The developed chip demonstrates the efficacy of plant sensors that are non-invasive and allow for continuous monitoring of plant health and growth.The electrochemical nitrate and phosphate sensors, which encompass electrical and molecular-based systems, offer rapid detection of nutrient changes.The functionality of the developed microfluidic chip was successfully validated through the germination of the legume crimson red and the observation of nutrient uptake.The fabricated sensor demonstrated good performance in measuring nitrate and phosphate concentrations within a wide range of 1 to 1000 mM.For nitrate measurement, the sensor exhibited a sensitivity of 0.001 M and an R 2 value of 0.99.Compared with the standard solution this indicates a strong correlation between the sensor readings and the actual nitrate concentrations in the growth medium.Similarly, for phosphate measurement, the sensor displayed a sensitivity of 0.001 M and an R 2 value of 0.95, further highlighting its reliability.With the successful validation of our microfluidic chip, we have opened avenues for investigating dynamic plant-nutrient interactions in real time.The device enables the study of nutrient content specifically nitrate and phosphate in soil thereby providing valuable insights into the phenotypic and genotypic variations associated with nutrient deficiency.With the ability to integrate nutrient sensors, optimized fertilization can be established, thereby maintaining a sustainable agricultural environment.The microfluidic approach can help in studying nutrient microenvironment and associated deficiencies for overall better productivity and yield of crops as well as reducing the cost of large-scale ex situ research.The microfluidic approach provides a bridge for testing various plant-related physiological and biochemical interactions before conducting field trials.Further, with the simplicity of this device, it can be used by plant scientists for a wide range of real-time applications in the future, including rootpathogen interaction, drought-resistant plant selection screening, nutrient uptake efficiency, and monitoring the soil micro-environment.

Figure 1 .
Figure 1.(a) Schematic diagram of nitrate-specific MIP films (b) Chemical equation: Binding of nitrate molecule with the polypyrrole.

Figure 2 .
Figure 2. (a) Schematic Illustration of real-time circuit setup which features a microcontroller interfaced with a mux for switching between plant sensor devices and an oxidation/reduction potentiometer (ORP) that performs the measurements.All electrical components are connected using a breadboard.Data can be viewed on the serial monitor or sent to the cloud using LoRa.(b) Flow diagram illustrating the multiplexed plant sensing devices communication to cloud, smartphone, and personal computer.

Figure 3 .
Figure 3. (a) Schematic illustration of the fabrication process; (b) Illustration of germination wells with integrated sensors and PDMS sheets; (c) View of the sensor-integrated microfluidic chip facilitating the growth of legume plant; (d) An enlarged view of legume roots.

Figure 5 .
Figure 5. Cyclic voltammogram of the doped PPy sensors in different concentrations of (a) NaNO 3 − and (b) KH 2 PO 4 − at a scan rate of 50 mV s −1 .

Figure 6 .
Figure 6.(a) Calibration plots for nitrate and (b) phosphate doped PPy sensors with peak currents vs concentration.

Figure 7 .
Figure 7. (a) Legume plant shoot and root growth in the device for 7 d.(b) nutrient uptake trend by legume plant for approximately 7 d with fresh nutrients recharged after day five (1 ml nutrient MS media was added).

Table I .
Recent development in soil-based (Nitrate and Phosphate) nutrient sensors.