Some examples of methods that have been applied at pilot plants will be briefly reviewed here. Across
most pilot facilities the amine degradation product emitted at highest quantities seems to be ammonia.
The next highest emissions typically are the solvent amines themselves. Both the ammonia and amines
have been successfully measured under certain conditions using traditional Fourier Transform Infrared
(FTIR) spectroscopy methods. However, the FTIR detection limits for these chemicals have been reported
in the range of 1 ppm at pilot facilities, but down to 0.1 to 0.5 ppm under ideal conditions with
substantially less water present, but these values were quite dependent on the facility design. Depending
on the concentration of the solvent amines, they may or may not be detectable at these limits. Improved
detection limits that could resolve these compounds at a wider range of pilot facilities might be possible
with the introduction of longer path lengths and more spectrally-sensitive detectors. These improvements
are theoretically achievable but have not yet been attempted at pilot facilities. Lessons learned from the
application of FTIR suggest strict adherence to the heated sampling line is of particular importance for
amines as even heat gap lengths of less than 1 cm can impact the results. Attempts to remove water from
the flue gas sample are not recommended because most of the chemicals of interest are quite soluble and
would be lost unless this liquid was collected and subjected to offline laboratory analysis. Alternative to
the FTIR technique, the TDLAS technique (Tunable Diode Laser Absorption Spectroscopy) is also suited
for the on-line monitoring at similar (or possibly lower) detection limits. Disadvantage of the latter
technique is the fact that only NH3 (and water depending on the manufacturer) can be measured whereas
with a FTIR analyzer, up to 50 components (inorganic as well as organic) can be measured
simultaneously. Additionally, several manual methods with isokinetic sampling and impingers (often with
acidic solutions to encourage trapping of the chemically basic amines) have been used for these
compounds, most of which bear substantial similarity to EPA Method 5 approaches.
Nitrosamines are a very difficult chemical class to measure at a pilot plant stack because of their expected
low concentrations, difficulty to clearly separate individual compounds in mass spectrometry approaches,
and the frequent requirements in existing methods (usually based on ambient air and occupational health
sampling techniques) for chemical modification of the collected sample in the laboratory before analysis.
Thus far measurement efforts have used manual sampling with cartridges loaded with Thermosorb/N,
extraction, and analysis by gas chromatography, high pressure liquid chromatography, and thermal energy
analysis (TEA). These approaches have followed or slightly modified the OSHA 27 method.
Thermosorb/N has been shown to be very sensitive to water saturated conditions. If water removal
techniques such as condensation are used, the liquid phase must be collected and analyzed in additoin to
the sorbent traps. Although TEA has not been employed for online monitoring, it has been suggested for
use as a screening tool for unknown nitrosamines in flue gas samples from PCCC facilities due to its
selectivity, sensitivity and the fact that it does not require an external calibration standard for each
compound of interest in order to provide an accurate quantitative response. A number of entities are
working on new nitrosamine methods, both to capture numerous individual chemicals at once to approach
a total chemical class approach, or by focusing on individual, or smaller sets of, chemicals. Some of these
are based on other nitrosamine properties, such as their sensitivity to UV exposure.
The working group activities are ongoing; the upcoming report will provide additional detail on the
features and drawbacks of a number of relevant methods for all chemical classes in Table 1beyond what