JSSSJournal of Sensors and Sensor SystemsJSSSJ. Sens. Sens. Syst.2194-878XCopernicus PublicationsGöttingen, Germany10.5194/jsss-6-327-2017In situ measurements of O2 and COeq. in cement kilnsDriesnerOlgaolga.driesner@enotec.deGumprechtFredGuthUlrichENOTEC GmbH, Höher Birken 6, 51709 Marienheide, GermanyDepartment of Chemistry and Food Chemistry, Technische Universität Dresden, 01062 Dresden, GermanyOlga Driesner (olga.driesner@enotec.de)30August20176232733029May201718July201722July2017This work is licensed under the Creative Commons Attribution 3.0 Unported License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/3.0/This article is available from https://jsss.copernicus.org/articles/6/327/2017/jsss-6-327-2017.htmlThe full text article is available as a PDF file from https://jsss.copernicus.org/articles/6/327/2017/jsss-6-327-2017.pdf
The simultaneous in situ measurement of O2 and COeq. in
cement kilns is a great challenge due to the high process temperatures and
high dust load. The standard method for measurement for flue gas in cement
kilns is extractive. Extractive measurements have a higher response time due
to the flue gas conditioning including the length of heated extraction lines
for electrochemical or optical analysis. This delayed response is not optimal
for fast process control.
A probe was developed for this purpose in which the in situ solid electrolyte
oxygen sensor and an in situ COeq. mixed potential sensor are
implemented. Due to the high temperatures, the probe is cooled by a
water–coolant mixture. In order to prevent deposits of raw material forming
and sintering on the probe, it rotates 90∘ in programmable intervals.
In addition, an automated probe plunger pneumatically removes plugging at the
probe flue gas entrance, also in programmable intervals. These self-cleaning
functions allow the probe to continually stay in the process for combustion
optimisation (low excess O2 and CO) and enable the plant operator to
measure additional process-related gas components (NO, SO2, HCl etc.)
and optimise the SNCR (selective non-catalytic reduction) for NOx
reduction. Combustion air supply can be adapted very quickly due to the in
situ sensors, which has been demonstrated by a
CEMTEC® probe over years (Märker Cement
Harburg, 2017).
Overview of the CEMTEC gas analysing system with air/coolant re-cooler.
Top view of the mixed potential sensor; electrodes are printed on
YSZ electrolyte, contacts made of gold.
Introduction
Combustion control in cement kilns is an enormous challenge as
real-time data of the O2 and COeq. gas concentrations are
necessary. In a typical cement kiln, various fuels with different calorific
values are used, resulting in flue gas with varying compositions. The optimal
fuel / air ratio varies depending on the fuel in use. Coal, gas or oil are
the most common fuels used at the kiln burner. Besides these fuels, domestic
and industrial waste consisting of a variety of materials is burnt. For
optimal combustion, excess O2 and COeq. should be kept at a
minimum. For this purpose only solid electrolyte sensors are suited as they
are stable enough to withstand the harsh process conditions and exhibit a
fast response in milliseconds. In order to measure under these circumstances,
a water-cooled probe was developed which can be installed as near as possible
to the kiln burner. Usually oxygen sensors based on the Nernst principle are
utilised. While the application of oxygen sensors for this process is common,
COeq. sensors are not usually implemented for this purpose.
Set-up of the gas probe
The CEMTEC system consists mainly of the probe itself (1), a compressed air
tank (3), a re-cooler for the cooling water (4), a coolant control
cabinet (5), a local control box (2) and a programmable logic controller (PLC) cabinet (6) – see the
CEMTEC® system overview in Fig. 1.
Inserted in the process, the probe oscillates 90∘ to prevent deposits
of raw material forming and sintering on the probe surface, to prevent the
probe from sticking and to ensure homogeneous heat distribution. The
plunger/filter tube removes plugging at the probe's flue gas entrance and
pulverises any internal deposits which are then removed during purging. The
O2 and COeq. sensors are situated at the back of the probe
tube and measure in situ. Additionally, dust-free sample gas is extracted
through the probe's sintered metal filter, through the heated sample lines to
an analyser cabinet for downstream measurement of other gas components
necessary for process control.
Sensor response of the COe MXP sensor.
Oxygen and CO concentrations from the kiln inlet over a
full day (upper curve oxygen, the lower one CO).
The probe is cooled by a coolant (water–glycol mixture) which in turn is
cooled by an air–water re-cooler. A three-way valve provides a constant
coolant return temperature of 85 ∘C. Once a day the probe is
retracted and re-inserted into the process to ensure that the probe can
retract automatically in case emergency retraction is necessary.
All PLC-controlled drives of the CEMTEC®
probe system are completely pneumatically operated, namely, probe rotation,
plunger movement and probe insertion/retraction. A sufficiently large
compressed air buffer tank is used as an “energy reserve” for an emergency
retraction should the main voltage or plant air fail.
Sensor principles
Oxygen is measured by the well-known solid electrolyte cell using platinum
electrodes on yttria stabilised zirconia (YSZ). The concentration can be
calculated according to the Nernst equation:
Ueq=RT4Fln(pO2′′pO2′).
Ueq cell potential
R universal gas constant
T temperature
F Faraday constant
pO2′′ O2 partial pressure
reference air
pO2′ O2 partial pressure
process gas
With air as a reference the volume concentration of oxygen is obtained by
Eq. (2) (Guth, 2012):
p′O2/vol%=20.69⋅exp-46.42(Ueq/mV)/(T/K).
The operating temperature of the solid electrolyte cell was fixed at 800 ∘C.
For the measurements of gas components such as carbon monoxide (CO) or
hydrocarbons (CxHy) in non-equilibrated gas phases, kinetically
determined sensors are used. Depending on the electrode material, the gas
components do not equilibrate on the measuring electrode at temperatures
< 700 ∘C. Thus gas components, which are not
thermodynamically stable, are electrochemically active. In CO and O2
containing gas, at least two electrode reactions can take place:
electrochemical reduction of oxygen and the electrochemical oxidation of
carbon monoxide. The measured open circuit voltage does not obey the Nernst
equation. Therefore such electrode behaviour is often referred to
non-Nernstian electrodes (or mixed potential sensors). The cell voltage
Umix depends logarithmically on the concentrations according to
Eq. (3) (Guth and Zosel, 2004; Shuk et al., 2008):
Umix=U0-A⋅lnφCO.
Umix mixed potential
U0 offset voltage
A material- and temperature-dependent constant
φCO CO concentration
The mixed potential sensor in thick film technology was developed for
measurement of CO in flue gases of combustion chambers. The top view of the
sensor which has the size 9.5 × 3.5 mm is shown in Fig. 2. For
the CO-sensitive electrode, a mixture of different oxides is used. The choice
of electrode materials diminishes the cross-sensitivity to hydrogen and
hydrocarbons. Nevertheless, certain cross-sensitivities cannot be avoided.
Therefore, the concentration of CO is expressed in COeq., which
means CO and equivalent. The specific mixture of oxides allows a wide
measuring range of COeq. The operating temperature of the mixed
potential sensor was adjusted to 700 ∘C. The sensor response is
shown in Fig. 3.
The high sensitivity and fast response are evident whereby a sensor signal of 80 mV per 1000 ppm CO CO is reached. The
calibration of the COeq. sensors is carried out under simulated
process conditions. Offset, span and slope are determined during the
calibration process which enables higher accuracy under process conditions.
Results and discussion
The results of gas analysis, which was carried out in a cement plant, are
shown in Fig. 4. It is clear to see that the sensor response is fast enough
to control the process optimally.
A high oxygen concentration corresponds to a low carbon monoxide
concentration and vice versa. The measured O2 and COeq.
values at the kiln inlet enable the operator to adjust the combustion air
quantity accordingly. If the O2 value is too high, the ID fan must draw
more flue gas through the furnace system than necessary, resulting in
excessive fuel use due to more combustion air being heated than required for
the combustion process. If the O2 value is too low and there is a very
high CO content, the combustion is sub-optimal as the oxidation process is
incomplete. In addition, high CO concentrations will result in destruction of
the refractory lining in the kiln and kiln inlet chamber due to CO corrosion.
Here, the operator can make appropriate changes to the settings, either to
the burner or to the ID fan, to achieve the best possible combustion. After
2 years of operation, it can be concluded that the results of the probe
measurements are very convincing and, when compared to extractive
measurements of the same flue gas, are also plausible. Maintenance consists
of purging the sample line twice a year as well as a general maintenance in
the analyser cabinet with calibration of the analysers for NO and SO2.
The CEMTEC probe has a yearly maintenance interval. No loss in sensor
performance was observed over this 2-year period (Märker Cement Harburg, 2017). This
analysing technique also helps to meet the emission limits of NOx and
NH3, as can be seen in Fig. 4.
The data that support the findings of this study are not publicly available.
The authors declare that they have no conflict of interest.
This article is part of the special issue “Sensor/IRS2 2017”. It is a result of the AMA, Nuremberg,
Germany, 30 May–01 June 2017.
Edited by: Peter A. Lieberzeit
Reviewed by: two anonymous referees
ReferencesGuth, U.:
Gas sensors, 2nd Edn., Springer,
in: Electrochemical Dictionary
edited by: Bard, A. J., Inzelt, G., and Scholz, F., 400–402, 10.1007/978-3-642-29551-5_7, 2012.Guth, U. and Zosel, J.:
Electrochemical solid electrolyte gas sensors – hydrocarbon and NOx
analysis in exhaust gases, Ionics 10, 366–377, 10.1007/BF02377996, 2004.
Märker Cement Harburg: User Report,
in: Global Cement Magazine May, 2017.Shuk, P., Bailey, E., Zosel, J., and Guth, U.:
New advanced in situ carbon monoxide sensor for the process application,
Ionics, 15, 131–138, 10.1007/s11581-008-0274-4, 2008.