A submersible UV/VIS spectrometer for in-situ real-time measurements is presented. It utilises the UV/VIS range (200-750 nm) for simultaneous measurement of COD, filtered COD, TSS and nitrate with just a single instrument. A global calibration is provided that is valid for typical municipal wastewater compositions. Usually high correlation coefficients can be achieved using this standard setting. By running a local calibration improvements concerning trueness, precision and long term stability of the results can be achieved. The calibration model is built by means of PLS, various validation procedures and outlier tests to reach both high correlation quality and robustness. This paper describes the UV/VIS spectrometer and the calibration procedure.

An in-situ UV spectrometer was applied to the effluent of a WWTP in Switzerland and calibrated using a multivariate calibration algorithm based on PLS regression. Except for nitrite, the calibration was based on comparative measurements of the effluent in the plant laboratory. Samples made of stock solution added to three different matrices prepared in the EAWAG laboratory were used for the nitrite calibration because the effluent concentrations were always in the range of 0.06-0.26 mg/l. The results show very good precision for nitrite and nitrate. The measuring range for COD and DOC was not completely covered by the measurements, so the meaningfulness of the results is limited. Nevertheless the precision obtained for soluble COD is high enough for most applications at WWTPs. The accuracy of the TSS measurement is unsatisfactory as regards effluent limits since the spectrometer used does not cover the wavelength region up to 700 nm, which gives better signals for TSS calibration due to its strong correlation with turbidity.

A submersible UV/VIS spectrometer was used to monitor a paper mill wastewater treatment plant. It utilises the UV/VIS range (200-750 nm) for simultaneous measurement of COD, filtered COD, TSS and nitrate with just a single instrument. The instrument measures in-situ, directly in the process. Paper mill wastewater shows typical and reproducible spectra at various process measuring points. There is a relative maximum at 280 mm due to the absorbance by dissolved organic substances, mainly ligninic acids. Comparison of absorbance spectra distinctly shows the decrease of this peak, indicating biological degradation throughout the treatment process. Summarising, one can say that paper mill wastewater cannot be monitored by a simple UV probe measuring only the absorbance at a single wavelength. The required information can only be gained from the whole spectra. Regarding plant control it is suggested that only the overall spectral information is used. Calibrations to conventional parameters are now merely carried out for purposes of reference-checking.

A prototype mid-infrared sensor system for the determination of volatile organic pollutants in groundwater was developed and tested under real-world conditions. The sensor comprises a portable Fourier transform infrared spectrometer, coupled to the sensor head via mid-infrared transparent silver halide fiber-optic cables. A 10 cm unclad middle section of the 6-m-long fiber is coated with ethylene propylene copolymer in order to enrich the analytes within the penetration depth of the evanescent field protruding from the fiber sensor head. A mixture of tetrachloroethylene, dichlorobenzene, diethyl phthalate, and xylene isomers at concentrations in the low ppm region was investigated qualitatively and quantitatively in an artificial aquifer system filled with Munich gravel. This simulated real-world site at a pilot scale enables in situ studies of the sensor response and spreading of the pollutants injected into the system with controlled groundwater flow. The sensor head was immersed into a monitoring well of the aquifer system at a distance of 1 m downstream of the sample inlet and at a depth of 30 cm. Within one hour, the analytes were clearly identified in the fingerprint region of the IR spectrum (1300 to 700 cm(-1)). The results have been validated by head-space gas chromatography, using samples collected during the field measurement. Five out of six analytes could be discriminated simultaneously; for two of the analytes the quantitative results are in agreement with the reference analysis.