Seawater pollution concentrates at river estuaries, harbor exits, and straits bearing heavy ship traffic. Coastal survey patrols can detect water turbidity or oil contamination during cruising, but they usually cannot analyze molecular pollution in real time unless they use some form of spectroscopy. Fluorescence lidar is a well-known method; however, it is mostly restricted to the detection of a single species at one time—and is costly. Raman lidar has simultaneous multispecies capability, but suffers from the usually low Raman backscatter cross sections.
Surface-enhanced Raman spectroscopy (SERS) is an elegant solution to overcoming the weakness of Raman signals. The substance to be analyzed—the analyte—is adsorbed on a rough surface. Aided by the molecule-surface interaction, Raman signals can be boosted by several orders of magnitude (up to 107). But the technique requires close contact of the sensor with the analyte—for example, the contaminants in seawater.
This situation encouraged a multinational European research team to build an underwater measurement unit equipped with several sensors—one a SERS optode developed by researchers from the Technical University at Berlin (where the project is coordinated)—and to use it for a measurement campaign in the Gulf of Gdansk, which is part of the Baltic Sea.1 That part of the sea is polluted by the largest Polish river, the Vistula, as well as by considerable ship traffic to the Gdansk port, mandating a detailed survey of the water quality. The increased sensitivity of SERS opens the door to the detection of polycyclic aromatic hydrocarbons (PAHs) in situ and to record transient events, such as those caused by leakages in passing ships. Hence, the campaign was organized to perform stationary measurements, experiments in tow, and to record depth profiles.
The underwater measurement unit was connected to the boat by a tether and two cables for electricity and data transmission, and was rated to a depth of greater than 300 m. In addition to the SERS sensor, it housed other optodes for the detection of oxygen based on fluorescence quenching, a fiberoptic salinity sensor based on refractometry, and other commercial instruments for the measurement of conductivity, temperature, and depth. The laser was a 785-nm-emitting external-cavity laser diode installed in the unit's core instrument, which also incorporated an axial "grism" (combined grating-prism spectrometer) covering the visible range from 420 to 680 nm and the near-IR region from 800 to 950 nm for spectral analysis. A thermoelectrically cooled CCD was used as the detector.
For the SERS substrate, the authors chose a configuration of metal colloids suspended in a matrix spin-coated on a glass support or a silicon wafer. They created different SERS substrates by thermally reducing silver nitrate that had been dissolved in the sol-gel precursor mixture.2 The substrate, to be exposed to the seawater, was connected by optical fibers delivering the laser radiation and transmitting the collected Raman scattering to the spectrometer/detector.
Raman spectra obtained from the instrument during a stationary measurement performed over a period of 80 minutes shows the development of strong Raman components (see Fig. 1). A SERS result was obtained with filtered harbor water (see Fig. 2). To assign the peaks of the spectrum, a standard mixture of 16 types of polycyclic aromatic hydrocarbon (PAH) with a concentration of 3 µg/l was added to the sample and the SERS measurement repeated. Containing pyrene, naphthalene, fluorene, chrysene, and other PAHs, that mixture qualitatively matches with the harbor sample; hence, detection of PAHs in polluted water is possible in real time.
REFERENCES
- H. Kronfeldt et al., ISOPE-2004, Toulon, Paper 2004-JFC-01.
- T. Murphy et al., Appl. Phys. B 69, 147 (1999).