step was the characterization of the polymers in respect to the sensor responses
for the analytes. The responses of the 6 sensors coated with the different polymers
were simultaneously measured by the array setup. The sensor responses of the
polymers PUT, PDMS and HBP are shown on the left side of figure
6. These 3 polymers contain polar groups and should therefore show different
response characteristics for analytes with different polarities or polarizabilities
. The sensor responses
of the polymers UE 2010 and M 2400 are shown on the left side of figure
7. Both polymers are amorphous glassy polymers with a microporous structure
whereby the mean size of the pores of M 2400 is 0.1 nm3
and the mean size of the pores of UE 2010 is 0.08 nm3. These polymers can discriminate
analytes due to different sizes of the analytes as only analytes with a smaller
volume of the molecules than the volume of the pores sorb into the pores of
the polymers. Further discussions regarding the pores can be found in chapter
to both figures, all polymers show a fast and reversible swelling when exposed
to R22 whereby the polar polymers PUT and PDMS are reaching an equilibrium state
instantly. These two polymers were measured above their static glass transition
temperature. The glass transition temperature is the temperature, above which
the molecules in the polymer backbone can move relatively to one another resulting
in a quasi-liquid state ,.
Thus, the interactions between vapor and coating can be described as dissolution
of a solute vapor in a solvent coating resulting in a very fast sorption and
desorption of the analytes, which can be modeled by linear solvation energy
relationships (LSERs) -.
Consequently, these two polymers show also an immediate sorption and desorption
of R134a. On the other hand, exposed to R134a the microporous polymers UE 2010
and M 2400 do not reach an equilibrium state within 30 minutes and the signals
need 2 hours to return to the baseline. Due to the bigger volume of the molecules
of R134a, the sorption process is kinetically inhibited and the molecules are
less (and more slowly) sorbed into the polymers.
figure 6: Sensor responses,
calibration curves and standard deviations of 3 measurements of the polar polymers
recorded with the array setup.
figure 7: Sensor responses,
calibration curves and standard deviations of 3 measurements of the microporous
polymers recorded with the array setup.
sides of figure 6 and figure 7
show the signals of the 6 sensors versus the concentrations of the 2 analytes.
This type of plots is often referred to as calibration curve. The correlation
of the polymer swellings (the sensor signals) with the analyte concentrations
is often described as a Henry sorption , a Langmuir sorption
or a combination of both types -.
For all polymers under investigation, the sorption of R134a is best described
as a linear Henry type sorption. This type of sorption process is an indication
of an unspecific sorption process .
The molecules of R134a are too big for the micropores and therefore only an
unspecific sorption process into the polymer matrix of the microporous polymer
can be observed. The sorption causes a swelling of the polymer matrix, which
can be observed as an increase of the thickness of the sensitive layer. The
sorption of R134a and of R22 into the two polar polymers is also of the Henry
type, since both analytes do not have distinctive polar groups and thus do not
specifically interact with the polymers.
other hand, the sorption of R22 into the microporous polymers UE 2010 and
M 2400 shows a calibration curve, which can be best described by the combination
of the Henry and Langmuir sorption. The Langmuir type sorption can be found,
if there is a specific interaction and if the amount of sorption and interaction
sites is limited . The combination of both sorption
types can be best detected when examining the curve for high analyte concentrations,
as the Henry sorption and the Langmuir sorption are identical for small concentrations.
If the sorption is a pure Langmuir sorption, the calibration curve should pass
into saturation for high concentrations whereas the combination of both types
of sorption results in a curve with a positive slope for high concentrations.
In figure 7, both polymers show for R22 this latter
case of a Langmuir sorption with a small portion of a Henry sorption. Furthermore,
the figures demonstrate that the microporous polymers show a much higher slope
for R22 than for R134a. Both findings can be explained by a sorption of the
small molecules of R22 into the micropores. This results in higher signals of
the R22 sorption, as the unspecific Henry sorption into the polymer matrix (which
is also present for R134a) is overlaid by a specific Langmuir sorption into
the pores (which is not present for the bigger R134a molecules). The number
of pores is limited and consequently the sorption of the molecules into the
pores and with it the Langmuir part of the sorption reaches saturation for higher