Project Title: In-situ monitoring of a reactive trace metal in riverine and estuarine mixing zones
Principal Investigator(s): Dr. Robert H. Byrne, Dr. Michael R. Callahan, Dr. Eric T. Steimle

Accomplishments
Scheduled Tasks:
Objectives for the current reporting period were to include moving from in-lab testing of SEAS-Cu to (i) deployment and in-field testing and (ii) optimization of the SEAS-Cu sensor for near-shore copper analysis.

Progress on Tasks
Beginning in March 2002, deployment our SEAS-Cu sensor commenced in Tampa Bay, FL. To date a total of 20 field deployments of the SEAS-Cu sensor have been completed. SEAS-Cu deployments have been used to determine copper in Tampa Bay, and optimize SEAS-Cu colorimetric chemistry for estuarine analysis. The modified chemistry has been validated, and a new series of SEAS-Cu field deployments are currently underway.

Difficulties Encountered
During the current reporting period two main obstacles were encountered. Both obstacles were associated with the pump configuration of the current generation SEAS(I) sensor. The first problem was associated with pre-mixing of the various reagents used in the Cu-BDS (copper-bathocuproine disulfonic acid disodium salt) colorimetric chemistry. During reagent stability studies, it was discerned that two Cu-BDS reagents (0.0208 M sodium dodecyl sulfate (SDS) and 1.02 M sodium citrate) formed a visible precipitate within several hours of mixing at the stock concentrations used in the cited Cu-BDS method1. The SDS-sodium citrate precipitates formed were large enough to clog the in-line flow-check value used in SEAS dye pumping system (see Figure 1). Obstruction of the flow-check valve by SDS-sodium citrate precipitates was overcome by reducing the concentration of the sodium citrate-HCl buffer.
The second problem encountered involved measurement protocol issues. SEAS(I) sensor colorimetric absorbance measurements are obtained as ratios of the transmitted light intensity (A = log I_o/I) using environmental samples mixed with (I) and without (I_o) the Cu-BDS reagents. Accurate measurements require that the inherent optical properties of the environmental sample remain consistent between I and I_o. Field deployment data obtained using the SEAS-Cu sensor in Tampa Bay, demonstrated that the transmittance of bay water samples (which are rich in colored dissolved organic matter (CDOM)) were effectively altered by changes in pH when Cu-BDS reagents were mixed with the sample. The problem of baseline alteration is being addressed by modifying the Cu-BDS reagents in order to minimize their effect on sample pH.

Anticipated Success in Meeting Project Objectives in Scheduled Project Period
We anticipate that all project objectives will be meet within the allotted project timeline.

Preliminary Results
The absorbance (A484) versus copper concentration response of the Cu-BDS complex using 50% less buffer was evaluated using standard regression analysis. The standard regression curve is shown in Figure 2. The absorbance response was linear (R² = 0.9980) for added copper concentrations between 0 and 200 nM. The molar absorptivity (13,795 L mol^-1 cm^-1) of the Cu-BDS complex determined from the slope of the regression (_ = 98 cm) was in excellent agreement with the value (13,863 + 45 L mol^-1 cm^-1) observed for the standard LPAS-Cu method.

Preliminary deployments of the SEAS-Cu sensor in Tampa Bay during the months of April and May yielded promising results. Figure 3 shows the results of SEAS-Cu field deployments in Tampa Bay: (a) Bayboro Harbor, a deep water harborage and (b) Sunshine Skyway Bridge, an open bay location (see Figure 3). SEAS-Cu data showed lower Cu concentrations for the open bay site relative to the more anthropogenically-impacted harbor.

During subsequent field deployments of the SEAS sensor in June and July 2002 it was determined that baseline alterations can interfere with Cu-BDS measurements. Figure 4 shows a baseline offset that occurs when non-colorimetric reagents are added to a bay water sample (see Figure 4). The offset shown in Figure 4 results in a 60% reduction in the calculated copper concentration. By referencing the Cu-BDS spectrum to bay water plus the non-colorimetric reagents, baseline alterations can be eliminated. A second generation SEAS(II) sensor (1st prototype due autumn 2002) will have three reagent pumps. In this case, potential baseline offsets will be eliminated by introducing non-colorimetric reagents to the reference measurement.

Prior to the acquisition of SEAS(II), we are exploring the use of modified Cu-BDS chemistries which will mitigate baseline alterations. Figure 5 shows the absorbance versus pH response for two series of bay water samples (see Figure 5). The data shows that by matching the pH of bay water samples with that of pure bay water baseline offsets can be eliminated. This is achieved by adjusting the reagent pH in order to minimize pH changes as the sample and reagents are mixed.

Based on these experiments, a modified Cu-BDS reagent mix (1.7E-5 M BDS, 7.32E-3 M NaOH, 7.3E-3 M hydroxylamine hydrochloride, 0.014 mM Tween 20) optimized for near-shore seawater samples is being evaluated. The absorbance (A₄₈₄) versus concentration response (Figure 6) for the modified Cu-BDS reagent mixture tested in the SEAS-Cu sensor was linear (R² = 0.9998) for added copper concentrations between 25 and 150 nM (see Figure 6). The molar absorptivity (14,167 L mol^-1 cm^-1) determined from the slope of the regression (_ = 98 cm) was in reasonable agreement with the value (13,863 + 45 L mol^-1 cm^-1) observed for the standard LPAS-Cu method.

On September, 4th 2002, the modified Cu-BDS chemistry was tested using four discreet grab samples from Bayboro Harbor in Tampa Bay, FL (Figure 7). Samples were analyzed using the SEAS-Cu measurement protocol. Copper concentrations ranged between ~ 20 and 30 nM with slightly higher concentrations occurring at the west-end of the harbor. On the following day, real-time copper measurements were analyzed in-situ at site #2 using the SEAS-Cu instrument. A grab sample from site #2 was also taken and measured in the lab using a desk-top LPAS system. Data from the SEAS-Cu deployment and grab sample are given in Table 1. The data shows good agreement between the real-time in-situ measurements ([Cu] = 17 + 2 nM) and laboratory measurements ([Cu] = 19 + 1 nM).

Tasks and activities for next reporting period

Tasks for the next reporting period
Deployment of our SEAS-Cu sensor along Tampa Bay transects is the principal task for the next reporting period. Data collected during SEAS-Cu deployments in Tampa Bay and Rookery Bay, FL will eventually be disseminated through publication in a peer-reviewed journal.

Work plan to accomplish tasks
A series of in-situ SEAS-Cu deployments will be conducted at and between 5 locations in Tampa Bay and Rookery Bay, FL. Selected locations, including transects between the open bay and anthropongentically impacted harbors, will provide a range of copper concentrations. From at least two of the selected sites in-situ deployments will include time series data.

Concerns or difficulties
Approximately four weeks were required to resolve issues associated baseline variations in high CDOM waters. This was not anticipated in our scheduling. However, we now have useful measurement protocols, and will soon have a modified SEAS sensor (SEAS II), that should make our measurements quite robust.

Expenditures
Expenditures were in the range anticipated for the work completed to date.