Significant costs are associated with laboratory analyses of groundwater samples collected at Department of Defense (DoD) sites. Most of these samples are needed to characterize the nature and extent of contamination at a site, evaluate remedial system performance, and track contaminant plume migration via regularly scheduled monitoring events. There is need to replace laboratory analyses with reliable, easy-to-use field methods that produce real-time results. Colorado State University (CSU) has developed fiber-optic biosensors that are ideally suited for field monitoring of groundwater contaminants.

The overall objective of the biosensor demonstration was to provide a basis to justify the use of biosensors to augment or replace conventional analytical methods for measuring selected compounds in groundwater. Specific objectives included:

• Demonstrating the accuracy, reliability, and cost of biosensors
• Demonstrating the effectiveness of on-site field measurements using biosensors
• Determining operational limits associated with using the biosensors
• Transferring the biosensor technology to end users.

Generally, a biosensor is a device that utilizes a biological recognition element (typically enzymes or antibodies) to sense the presence of an analyte and create a response that is converted by a transducer to an electrical or optical signal.

The primary issue regarding the use of biosensors is reliability, i.e., are biosensor results comparable to laboratory analyses? The end user also needs to know whether there are conditions that affect the reliability of biosensor performance. Biosensors also need to be easy to use and calibrate so that reproducible results can be obtained from different users.

Biosensors were used to analyze groundwater sampled from several monitoring wells at Operable Unit 8 (OU8) of the Bangor Naval Submarine Base (SUBASE Bangor) in Kipsap County, Washington, to evaluate biosensor performance under a range of conditions. The target analyte was 1,2-dichloroethane (1,2-DCA). Groundwater samples were collected from monitoring wells spaced throughout the plume to analyze a wide range of 1,2-DCA and co-contaminant concentrations. The samples were analyzed by biosensors and gas chromatography/mass spectroscopy (GC/MS). A flow-through cell also was set up to allow biosensor readings in flowing water similar to the setup typically used to collect pH, conductivity, and turbidity readings prior to monitoring well sampling. Biosensors were lowered into monitoring wells to record down-hole in situ readings.

Performance of the biosensors was evaluated based on the following criteria:

• Accuracy, as demonstrated by a one-to-one correlation between the two analytical techniques (conventional GC/MS and biosensors)
• Range, as demonstrated by a response from less than 5 micrograms per liter (μg/L) to greater than 500 μg/L 1,2-DCA
• Precision, as demonstrated by a low relative percent difference (RPD) between duplicate analyses
• Sample throughput, as demonstrated by short analysis time in the field
• Mechanical reliability, as demonstrated by a low incidence of failure
• Versatility, as demonstrated by acceptable performance under a variety of conditions.

Two performance levels were established with regard to the data that the biosensors might be used to collect: Level 1 was the ability to provide qualitative, screening data with definitive compound identification. Level 2 was the ability to provide definitive compound identification and quantitative concentrations.

The interference of parameters affecting the pH of the groundwater being measured impacted the biosensor’s performance against several performance criteria, including accuracy, precision, sensitivity, and range. The biosensor measures small pH changes produced by the reaction of an enzyme with 1,2-DCA, and techniques are required to distinguish these pH changes from pH changes due to other processes. For vial measurements, this interference can be significantly reduced by proper calibration. However, for flow-through cell and down-hole measurements, calibration procedures have not been developed to reduce the pH interference. Because the biosensor measures small pH changes produced by the reaction of an enzyme with 1,2-DCA, methods are required to distinguish these pH changes from pH changes due to other processes. This can readily be accomplished by adding an optical fiber (bundled with the biosensor) and a second measurement channel to the hardware, thus providing optical pH measurement for correction of the pH changes.

At the present level of development, the biosensors would most appropriately be used to provide semi-quantitative data regarding 1,2-DCA concentrations in groundwater. The biosensors can be used to collect Level 2 quantitative data when used in the vial measurement mode; however, further investigation into development and testing of the biosensors is required for them to be reliable field instruments for all the applications originally intended.