Skip Header NavigationIntranet 
CENTER FOR EMBEDDED NETWORKED SENSINGContactDirectionsEmploymentEventsNews
HomeAbout UsResearchEducationResourcesPeople

Research Project

Potential Electrochemical Detection

Technology > Sensors > Creating Scaleable and Deployable Nitrogen-Cycle Sensors for Environmental Systems > Potential Electrochemical Detection

On this page: Overview | Approach | Systems/Experiments | Accomplishments | Future Directions | People

Overview

This aspect of the CENS Sensor Group effort builds on an ion selective electrode (ISE) microsensor for nitrate that was reported on in previous years. This ISE is created using a conducting polymer (polypyrrole, Ppy) doped with nitrate ions on a carbon electrode. In the past year work has emphasized (1) testing potentiometric sensors in situ, in the context of CENS contaminant observation deployments, and (2) increasing the stability (longevity) of the potentiometric sensors in the environment.  The potentiometric nitrate sensor is fabricated via electropolymerization of pyrrole in the presence of nitrate as the dopant ion.  The result, a nitrate-doped polypyrrole (PPy) membrane is highly selective for nitrate and amenable to fabrication in attractive form factors, such as that of wispy fibers mimicking root structures.  The sensor works well for weeks to months under controlled laboratory conditions.  Unfortunately, this type of sensor has consistently failed to maintain its sensitivity for more than a few hours in flow experiments.  In many instances the sensor can be reconditioned to restore sensitivity.  However, this level of longevity remains a serious limitation when it comes to real-world deployments.

Approach

We have previously tested the potentiometric nitrate sensors in laboratory test beds, including simple water flow-through systems and flow-through soil boxes.  Over the past year we endeavored to incorporate the CENS nitrate sensors into contaminant observation deployments in comparison with commercial nitrate sensors. Given the previous life cycle indications for these sensors, we did not anticipate long-lasting responses in situ.  However, we did expect the exercise to yield valuable information.  To address the longevity issue with the nitrate ISEs, we attempted to encase the dopant ions in the PPy membrane by depositing a layer of insulating material (described below) atop the PPy membrane.  This fabrication technique is known as the bi-layer approach.

Systems/Experiments

Field deployments for the PPy-based nitrate microsensors were undertaken at three CENS test beds:  Palmdale, Merced, and the James Reserve.  In each case, the sensors were buried in the soil adjacent to commercial nitrate sensors (DirectIon 9 mm diameter combination electrodes, Sentek Ltd, UK).  For reasons of brevity, this section focuses on the Palmdale results. 

Figure 1

Palmdale Field Tests.  A photo from the May 2006 Palmdale deployment is shown in Figure 1. Typical results for the Palmdale test are plotted in Figure 2. Two nests of 4 nitrate sensors were installed adjacent to moisture sensor nests (see Contaminant Observation section of report for moisture data). 

Figure 1 - Microclimate station adjacent to dual nitrate senor access tubes (CENS and commercial nitrate sensors deployed side-by-side).

Each nitrate sensor nest was comprised of two commercial nitrate sensors and two CENS-fabricated sensors.  The Sentek sensors are roughly 10 cm long and 9 mm in diameter, while the UCM electrodes were prototypes that consisted of the working electrode and a commercial reference electrode (ESA no leak model miniature reference electrode) housed in a protective clear plastic casing.  The overall size of the CENS-fabricated sensors was approximately 4-5 mm in diameter (including protective epoxy coating) and 2-3 cm in length.  All of the nitrate sensors were deployed through a 1.5 inch PVC pipe that was placed at a 45 degree angle into the ground.  The pipe was open-ended allowing the sensors to protrude into the soil at the desired depth.   The above-ground end of the PVC pipe was sealed to prevent irrigation water from entering the tube.  This deployment strategy is being tested as a convenient means of inserting chemical sensors and periodically removing them for maintenance. 

Figure 2

Figure 2.  Comparison of commercial and fabricated nitrate sensor results at the Palmdale reclaimed water irrigation test bed.  Voltage drops denote irrigation-driven arrival of nitrate which was applied in a controlled release to perturb the system (tinted windows and broken lines represent extended and single-pass irrigation events, respectively).

Both the fabricated and Sentek sensors behaved erratically during the high intensity irrigation events.  This is the first time we have tested them under such conditions and the reason for this behavior is unclear.  To speculate, the highly variably readings may real, the result of contact with alternating high and low concentration water.  Alternatively, the variation may be an artifact of the sensors’ equilibrating under the highly dynamic flow and transport conditions.

After the high intensity episodes, the Sentek sensors appeared to capture the dynamics of the nitrate transport well relative to the fabricated sensors.  Looking first at the shallow (7 inch) Sentek, this sensor observed roughly 100 ppm after the initial high intensity event, while the deeper sensor lags a few hours behind in response to the flooding.  After this early period, the nitrate appears to be redistributed into both the 7 and 14 inch zone upon arrival of the pivot at midnight May 11.  A still larger peak is brought about on the next revolution of the pivot.  However, both zones are relatively depleted of nitrate after the 06:30 May 11 pivot event.  Following the second high intensity event, both sensors read relatively low levels until the next pivot revolution, where the shallow sensor remains relatively constant, apparently measuring the residual nitrate concentration.  In contrast, the 14 inch sensor returns and maintains elevated nitrate concentrations, suggesting that the nitrate pulse has migrated and remained at the lower depths. 

The fabricated sensors did not agree well with the commercial sensors.  The clearly responded to irrigation events of both the high intensity and regular type.  Both of the fabricated sensors seemed to respond to the first high intensity event with significant increases in nitrate concentration readings.   However, neither exhibits the prominent increases shown by the Sentek sensors.  Furthermore, the two fabricated sensors exhibit increases rather than decreases following the second high intensity event.  Both then show a steady decline in nitrate concentrations until the end of the test.  These responses are difficult to explain and merit further investigation. 

Sensor Longevity Issue.  To address the longevity problem, CENS investigators Harmon nad Rat’ko visited our collaborators at the Weizmann Institute’s Organic Chemistry Department (under funding from the Bi-National Agricultural Research & Development Fund) to learn to create double-later deposits, where an insulating layer of bis-3.4-ethylenedi-oxyhtiophene (bis-EDOT) is deposited atop the PPy surface in an effort to retain the dopant nitrate anions which may be lost in flowing water.  All measurements were carried out using direct potentiometry technique. An Ag/AgCl saturated no-leak electrodes model EE-0009 from Cypress Systems (a division of ESA Inc,USA) was used as a reference electrodes in potentiometric cells. Potentiometric measurements were conducted using Fluke 111 True RMS multimeter and HOBO U12 4 channel Data logger. The plots in Figure 1 show the linear response (mV) to logarithmic increases in nitrate concentration.  The plots demonstrate performance in terms of calibration with the double-layer added was found to be as good as or better than the microsensors fabricated with a single layer of PPy (shown previously).

Figure 3

Figure 3. Potentiometric response of bi-layer (PPy(NO)3+Bis-EDOT) electrodes to NO3- - ion (22 °C). Each curve title indicates the date electrode was made (MM/DD/YY format) and the number of the electrode.

 In a series of comparative flow-through experiments in water and soil-box systems, the bi-layer microsensors (results in Figure 4) consistently outperformed the single-layer PPy microsensors.  However, while their longevity was significantly greater, the bi-layer sensors also lose their sensitivity in a relatively short period of time when placed in a flowing stream.   

Alternative Sensor Fabrication Strategies.  After the bi-layer sensor experiments, several polyvinyl chloride (PVC) nitrate-selective membranes were investigated in the literature.  The fabrication technique is different from the electro-polymerization technique we employ for the PPy-version of the microsensors.  For the PVC sensors, a mixture of PVC, plasticizer, and quaternary ammonium ions is dissolved in tetrahydrofuran (THF).  The solvent is then evaporated off the mixture, leaving a membrane behind that is selective for nitrate.  We have successfully replicated a recently published fabrication technique, and preliminary results from this sensor are promising.  However, we have not yet investigated the useful lifespan of this new type of potentiometric sensor under environmental conditions.

Figure 4

Figure 4. Results of flow-through tests comparing bi-layer CENS potentiometric microsensor with commercial sensor (1.5 l/min flow rate; perturbations are from spikes of 20 ml of 5 M NaNO3 3 times during the experiment).

Accomplishments

The main accomplishments in the potentiometric nitrate microsensor development project include:

Future Directions

People

Faculty:

Postdocs:

Graduate Students:

Undergraduate:

External Research Partnerships

Department of Organic Chemistry, Weizmann Institute, Rehovot, Israel (current)