Due to the preparation process, water
for pharmaceutical applications does not
contain any salt, which could raise the conductivity of the water. The conductivity of
this water is mainly influenced by dissolved
carbon dioxide (TIC) from the atmosphere
and elements of organic carbon compounds
An example with rounded off figures:
• Conductivity at the first Sensor (COND 1)
0.6µS/cm at 42°C
• Conductivity at the second Sensor (COND
2) 0.8µS/cm at 42°C
The absolute conductivity of water at 42
°C is approximately 0.12µS/cm. The difference of 0.48µS/cm to the measured conductivity at the first sensor arises from soluble
carbon dioxide. TIC and TOC can be calculated from the measured conductivities of
the sample because of the direct relation
between the conductivity and the total carbon dioxide content. The corresponding values are tabulated for different temperatures
and stored in the instrument.
The organic content in the sample is
oxidized by the UV lamp and converted to
CO2. The additional carbon dioxide raises
the conductivity of the sample and there-
fore also the measurement value at the
second sensor. Subtracting the first conduc-
tivity measurement from the second mea-
surement gives the conductivity that is cre-
ated from the oxidized carbon compounds:
[COND 2] - [COND 1] = 0.2µS/cm.
The TOC-content can be calculated
with the same tabulated values as the TIC
(COND1). This determination of TIC and
TOC is an absolute method under the conditions presented above because a comparison of the actual values with similar conditions takes place.
The extent of the potential deviation is
recorded in the menu of the measurement
converter. There is no adjustment.
The calibration is carried out with a specific sample/solution of 1ppm. If the reading
has a greater deviation from the set target
value, there are a number of explanations: the
necessary conditions are not fulfilled, the operating parameter is not correctly adjusted, or
the instrument has a technical fault.
This method is meaningful when there is
INCREASE OF RADIATION
The currently used UV
lamps (Hg-Low pressure
quartz lamps), perform with
full power only in a narrow
temperature range, between 40
and 50°C. A deviation of 10°C
in the operating temperature
can result in a loss of power of
up to 20%.
From the graph in Figure 5,
the importance of keeping the
temperature of the lamp within
the optimal range is clear.
Outside influences (
environmental temperatures, location) and
the sample water itself (
changing sample temperatures) can
lead to lamp temperatures outside the optimal
range. The heat exchanger solution minimizes
the difference between the two conductivity
sensors to less than 0.2°C. However, in order
to maintain the temperature in the optimal
range between 40 and 50°C, further measures
are necessary. Additional heat-exchangers or
sample coolers (C) enable a target temperature of 42°C to be precisely sustained. Thus,
the maximum radiation efficiency is achieved,
resulting in optimal and consistent oxidation.
(See Figure 6)
OPTIMIZING THE SAMPLE FLOW
The aforementioned measures already
show a substantial improvement to the system. The full potential is only realized in
combination with an optimized sample flow.
The classic design of a UV lamp and the
sample is flowed around the light source.
Divergence loss and reflection
are practically unavoidable.
The formation of deposits on
the directly radiated surface,
which decreases radiation den-
sity, cannot be fully eliminated
in long running operations.
This side effect can only
be avoided by establishing a
direct contact between the UV
lamp and the sample.
Figure 7 demonstrates the
sample flow in the newly engi-
neered UV-Reactor. The sample
flows directly along the lamp.
The furthest distance from the
middle of the lamp is 8mm and
the sample layer is only 0.5mm
thick. Divergence loss and
ozone production is prevented
because of the enclosed construction. This
optimization leads to a distinct increase of
radiation density, resulting in complete oxida-
tion of the carbon compounds present in the
sample. The UV lamp forms a single unit with
the reactor case. The entire reactor can be
replaced in the case of a failure and is recy-
clable. Maintenance is easier and faster.
Stable temperatures and a consistent, in-
tense radiation are the foundations for exact
and reliable measurement values. Equally
important, however, is the compensation
method, i.e., how the measured values are
converted to the standard temperature.
The conversion to standard temperature
( 25°C) can be calculated by various algorithms,
although these formulas are not absolute, only
approximations. For this reason, SWAN applies
two different compensation methods, which
are selected according to the field of operation.
Figure 6: Reactor with a heat
exchanger and a heating cartridge
Figure 7: Cross section of a UV reactor.