the backwashing and cleaning regimes. In wastewater treatment, TSS or the closely
related mixed liquor suspended solids (MLSS) are used to monitor sludge concentrations in aeration tanks and to control sludge recirculation and removal.
5.1.5 Other Direct Parameters
Various other substances have a strong and/or characteristic absorption signal that
allows direct measurement. Those relevant for water and wastewater treatment
include the treatment chemicals ozone (O 3 ) and permanganate (MnO 4
À ) and pollutants such as BTEX (benzene, toluene, xylene), phenol, iron and chromium, which
are indicators for contamination of water with hydrocarbons or industrial waste. In
sewer systems, hydrogen sulphide (H 2 S) can be measured. H 2 S is formed under
anoxic conditions and is both dangerous (it is highly toxic) and corrosive, leading to
biogenic sulphuric acid corrosion of pipe materials.
5.1.6 Indirect Parameters
Not all parameters can be measured directly using UV/Vis spectroscopy. In many
cases either the concentrations of the target analytes are too low to detect, or they do
not absorb enough light at the wavelengths monitored. In some cases, however, the
covariance of the invisible analytes with other, detectable components in the
medium allows the building of a calibration model that exploits this relationship.
Such calibration models for the indirect measurement of parameters have been
reported for a wide range of parameters, including ammonium, total nitrogen and
orthophosphate in wastewater, assimilable organic carbon (AOC) in drinking water
and bacteria (E. coli) in surface waters and drinking water. As these models rely on a
consistent relationship between the visible components and the invisible target
analyte, they are often specific for a particular monitoring location, and their validity
needs to be checked regularly.
Another type of indirect parameter is the process parameter. In this case, a
calibration model is built between (variations) in the spectrum and the (desired)
states in a treatment process. These parameters are used as real-time control inputs
and facilitate optimisation of water treatment through reduction of chemical and
energy consumption while safeguarding or improving treatment effectiveness.
Examples of such process parameters include prediction of coagulation dose
[15], prediction of chlorine demand and prediction of disinfection by-product
formation [19].
One method which has been applied especially for the building of models for
process parameters is differential spectroscopy, i.e. the subtraction of spectra measured before and after a process. The resulting “delta-spectrum” reflects the change
in the composition of the water as a result of the process [9].
Spectroscopic Methods for Online Water Quality Monitoring
297
related mixed liquor suspended solids (MLSS) are used to monitor sludge concentrations in aeration tanks and to control sludge recirculation and removal.
5.1.5 Other Direct Parameters
Various other substances have a strong and/or characteristic absorption signal that
allows direct measurement. Those relevant for water and wastewater treatment
include the treatment chemicals ozone (O 3 ) and permanganate (MnO 4
À ) and pollutants such as BTEX (benzene, toluene, xylene), phenol, iron and chromium, which
are indicators for contamination of water with hydrocarbons or industrial waste. In
sewer systems, hydrogen sulphide (H 2 S) can be measured. H 2 S is formed under
anoxic conditions and is both dangerous (it is highly toxic) and corrosive, leading to
biogenic sulphuric acid corrosion of pipe materials.
5.1.6 Indirect Parameters
Not all parameters can be measured directly using UV/Vis spectroscopy. In many
cases either the concentrations of the target analytes are too low to detect, or they do
not absorb enough light at the wavelengths monitored. In some cases, however, the
covariance of the invisible analytes with other, detectable components in the
medium allows the building of a calibration model that exploits this relationship.
Such calibration models for the indirect measurement of parameters have been
reported for a wide range of parameters, including ammonium, total nitrogen and
orthophosphate in wastewater, assimilable organic carbon (AOC) in drinking water
and bacteria (E. coli) in surface waters and drinking water. As these models rely on a
consistent relationship between the visible components and the invisible target
analyte, they are often specific for a particular monitoring location, and their validity
needs to be checked regularly.
Another type of indirect parameter is the process parameter. In this case, a
calibration model is built between (variations) in the spectrum and the (desired)
states in a treatment process. These parameters are used as real-time control inputs
and facilitate optimisation of water treatment through reduction of chemical and
energy consumption while safeguarding or improving treatment effectiveness.
Examples of such process parameters include prediction of coagulation dose
[15], prediction of chlorine demand and prediction of disinfection by-product
formation [19].
One method which has been applied especially for the building of models for
process parameters is differential spectroscopy, i.e. the subtraction of spectra measured before and after a process. The resulting “delta-spectrum” reflects the change
in the composition of the water as a result of the process [9].
Spectroscopic Methods for Online Water Quality Monitoring
297