OVERVIEW

Small, low-cost, robust, reliable, and sensitive sensors are needed to enable the realization of practical and economical sensor networks. Although there are a large number measurands that are of interest for sensor-network applications (e.g., seismic, temperature, light, sound, magnetic, chemical, etc.), appropriate commercial sensors exist for many of these measurands. However, one prominent exception is the fact that appropriate chemical sensors are not available. It is for this reason that the sensor technology effort within CENS is researching the design, fabrication, and implementation of chemical sensors that have the specifications needed for sensor networks.

In order to have a targeted effort to develop chemical sensors for sensor networks, we have focused on one sensor-network application, namely monitoring soil contamination / habitat monitoring. To model the flow of contaminants in soil, miniaturized lab-based soil systems (i.e., designed on the scale of meters) are first used to simulate real-world macroscopic field tests. These lab-based test systems need an array of miniaturized chemical sensors to accurately monitor the flow of contaminants in the model. Microsensors developed for this application will first be used in the lab-based system that has a controlled environment. Once the sensors have the robustness needed they will be used for field tests. To further focus and simplify the initial sensor technology development, the specific chemical contaminant compound that will be monitored is nitrate.

APPROACH

Our nitrate sensing system is based on an electroanalytical technique (i.e. amperometry). It incorporates micro-fabricated silver working electrode, silver-oxide reference electrode, and platinum counter electrode. Also, microfluidic channels were fabricated as flow paths to the microelectrochemical cell for the eluent (0.01 M NaOH) and ground-water sample. Linear calibration curves have been obtained (1 µM to 1000 µM ) and the detection limit is ~1 µM. An anion-permeable membrane is used to filter ground water sample for sensor selectivity.

PROBLEMS ENCOUNTERED

• The iterative electrochemical-activation process gradually erodes the silver electrodes. The narrow silver trace leading to electrode increases the current density and expedites the etching process. Eventually, these lead wires are cut and the working electrode is disconnected. Another problem that was observed is that etched silver particles on the electrode surface sometimes bridge the working electrode to the reference electrode. Consequentially, short circuiting and erratic responses were observed.
• The small electrode area of the counter electrode (0.8×10^-3 cm²) increases the current density and generates hydrogen bubbles that hinder electrochemical measurements.

Despite the simple fabrication of silver-oxide reference electrodes, they have limited stability. For example, the reference potential drifts about 50 mV over 2 hours

SYSTEMS / EXPERIMENTS

Description of Experiments and Systems for the Development of Micromachined Amperometric Nitrate Sensors

A newly designed nitrate sensor unit was tested for calibration curves (i.e., sensitivity and linearity). The electrochemical measurements were conducted using a CH Instrument 660B potentiostat and a data-acquisition computer. Figure 3.C.5-2 depicts the bench-top experimental setup. A sensor unit is electrically connected to the computer-controlled potentiostat. Simulated ground-water samples and eluents are injected into the sensor unit by 20-mL syringes. Electrolyte (0.01-M NaOH) was freshly prepared with reagent-grade chemicals and triply-distilled DI water. A commercial-off-the-shelf (COTS) 0.1-M nitrate standard (Thermo Orion Ionplus) was used for nitrate measurements. The CH Instrument 660B potentiostat was also used to fabricate and test the reliability of a polymer-coated Ag/AgCl reference electrode. To form the reference electrode, pure silver wire was chloridized galvanostatically in 0.1-M HCl solution.

ACCOMPLISHMENTS

Two main accomplishments made during this research period are:

• Newly designed nitrate microsensors were tested and demonstrated much improved reliability.

• A feasibility study of polyurethane-coated Ag/AgCl reference electrode as been performed and these electrodes show much improved stability.

Second-Generation Nitrate Microsensor

The research we performed on micromachined amperometric nitrate sensors revealed that there were serious robustness issues that needed to be addressed. Specifically, we had observed hydrogen bubble formation, silver lead-wire degradation, and short-circuiting problems. One of the first improvements we made was to enlarge the counter electrode from 0.8×10^-3 cm² to 7.7×10^-3 cm². In addition, a polyimide layer was deposited and patterned in order to protect the edges of the microelectrodes. Figure 3.C.5-6 shows a new layout for a nitrate-sensing chip and an SEM image (Figure 3.C.5-7) depicts the micromachined electrodes, polyimide insulation, and microchannels on the newly fabricated microsensor chip.

The performance of the microelectrode is now much improved as shown in the calibration curves that have been obtained with the new second-generation chips. The nitrate standards were measured in 0.01-M-NaOH electrolyte using the standard addition method. The measurements were performed with the working electrode biased to -0.9 V (vs. Ag/AgCl) and the chronoamperometric response was integrated for 500 msec. Figure 3.C.5-8 shows the calibration curves of the highest sensitivity and lowest sensitivity sensors. The detection limit was 1 µM and it should be noted that the linearity of the sensor has also been greatly improved, as compared to that of our first-generation nitrate microsensors (e.g., from r²=0.970 to 0.993).

Polyurethane-Coated Ag/AgCl Reference Microelectrodes

To simplify fabrication, the present chip design uses a silver-oxide reference electrode (i.e., thin-film silver anodized at 0.4 V for 12 minutes). However, a silver-oxide reference electrode has limited stability. The reference potential drifts about 50 mV over 2 hours as shown in Figure 3.C.5-9.

Although a Ag/AgCl reference electrodes can be used, it is well known that its potential drifts severely as the AgCl dissolves away in the electrolyte. Bindra et al. (1991) suggested that a Ag/AgCl reference electrode with a polyurethane (PU) protection layer may have much better stability. Harrison (1994) and Kounaves (1997) demonstrated the advanced stability of PU-coated Ag/AgCl electrodes, as compared to uncoated electrodes. Before developing a microfabrication process to integrated the PU-coated reference electrode, we prepared Ag/AgCl wires with following process and tested them: (a) chloridize silver wire (99.99%, Alfa Aesar) galvanostatically (0.249 mA for 30 minutes); (b) dissolve 5% PU (Thermedics, SG-85A) in 98% tetrahydrofuran (Aldrich) and 2% dimethlyformide (Johnson & Matthey); (c) dip-coat chloridized silver wire three times; (d) dry PU in desiccators overnight; and (e) store PU-coated wire in saturated KCl solution. The change in potential was recorded in a 0.1-M-KCl solution with respect to a commercial Ag/AgCl reference electrode (Bioanalytical Systems). As illustrated in Figure 3.C.5-10, there is an initial potential increase (i.e., from 0.0792 V to 0.0886 V over 30 minutes) and then the potential decreases very slowly. The longest test performed was 27 hours and the potential drift was only ~3 mV. The results from this feasibility experiment demonstrate the enhanced stability of PU-coated Ag/AgCl reference electrodes. Stability tests in 0.01-M-NaOH electrolyte are also underway. If such positive results are obtained again, we will then work to integrate this form of electrode into our second-generation microsensor. The result will be a miniature nitrate sensor with a long-term stability sufficient for field tests.

FUTURE DIRECTIONS and Objectives

Integration of Ag/AgCl reference Electrode

If PU-coated Ag/AgCl reference electrodes show enhanced reliability in a NaOH electrolyte, we will then work to integrate this form of electrode into our microsensor. 

Long-Term Qualification Test

Currently, only short-term tests have been performed. With computer-controlled peristaltic pumps, integrated fluidic system, and a data logger, automated long-term qualification tests will be performed. 

Connecting to a Wireless Network

The CENS-made interface board will be used to interface the present sensor setup to Crossbow Motes (www.xbow.com). Data transmission and sensor control will be tested wirelessly in a sensor network.

Quantitative Analysis of Sensor Selectivity

Sensor selectivity will be quantified with standard selectivity test methods (i.e., find concentration of each interfering chemical species that gives a 100% increase in sensor signal obtained for nitrate alone). The ion chromatography system at UC Merced will be used to quantify such chemical species. Real ground-water samples will be tested and compared with results from ion chromatography. 

Nitrate-Sensor Field Test

After bench-top performance tests are completed, nitrate microsensor units will be deployed in the field to measure concentrations of nitrate in real ground water.

Explore Chip-Scale and Board-Scale Potentiostats and Integrate them as Appropriate with the Sensor Unit

For a nitrate sensor to be portable, low-power, high-performance, and wireless-communication capable, a miniature and inexpensive potentiostat is needed – instead of a large and expensive laboratory potentiostat. We will explore the current technology of chip-scale and board-scale potentiostats. If needed, we will establish collaborations to facilitate the integration of the best potentiostat with our sensor.

PEOPLE

Faculty:

Prof. Jack W. Judy

Staff:

Dr. Ira Goldberg

Graduate Students:

Dohyun Kim