-
Procedia Engineering 87 ( 2014 ) 108 – 111
1877-7058 © 2014 The Authors. Published by Elsevier Ltd. This is
an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).Peer-review
under responsibility of the scientific committee of Eurosensors
2014doi: 10.1016/j.proeng.2014.11.394
ScienceDirectAvailable online at www.sciencedirect.com
EUROSENSORS 2014, the XXVIII edition of the conference
series
Investigations on work functions of gasochromic color dyes as
gate materials for FET based gas sensors
Carolin Petera,*, Dominik Zimmermannb, Daniel Knopa, Sven
Rademachera, Ina Schumachera, Ingo Freundb, Jürgen Wöllensteina
aFraunhofer Institute for Physical Measurement Techniques, 79110
Freiburg, Germany bMicronas GmbH, 79110 Freiburg, Germany
Abstract
We present investigations on gasochromic color dyes as gas
sensitive gate materials for field-effect transistor (FET) based
gas sensors. Therefore the work function of two color dyes
N’,N’,N,N-tetramethyl-p-phenylenediamine (TMPD) and bromphenolblue
(BPB) were characterized using a Kelvin probe. TMPD is selective to
NO2 and shows changes in the output of the Kelvin probe of 50 mV
during exposure to 0.5 ppm NO2. BPB is an indicator for ammonia.
Even 3 ppm NH3 cause a change of 10 mV. In addition, the
gasochromic materials are highly selective to only one gas. The
results show, that these dyes are suitable as sensitive gate
materials and offer the possibility to build a low-power FET,
detecting NO2 and NH3 in air at room temperature. © 2014 The
Authors. Published by Elsevier Ltd. Peer-review under
responsibility of the scientific committee of Eurosensors 2014.
Keywords: gas sensor; low-power; gasochromic color dye, gate
material; FET
1. Introduction
Today the use of low-cost sensors with low-power consumption is
becoming increasingly important in many applications. Especially
for battery-operated systems low power gas sensors are required.
Often metal oxide gas sensors (MOX sensor) are in use. They are
cost effective, robust and sensitive, but with the lack of
selectivity and high operation temperatures up to 450 °C.
* Carolin Peter. Tel.: +49-761-8857-731.
E-mail address: [email protected] E-mail address:
[email protected]
© 2014 The Authors. Published by Elsevier Ltd. This is an open
access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).Peer-review
under responsibility of the scientific committee of Eurosensors
2014
http://crossmark.crossref.org/dialog/?doi=10.1016/j.proeng.2014.11.394&domain=pdf
-
109 Carolin Peter et al. / Procedia Engineering 87 ( 2014 ) 108
– 111
GASFETs are a possible alternative to MOX-sensors. GASFETs can
detect changes in conductivity, capacitance
or work function of a gate material due to the gas exposure [1]
at room temperature. Many gate materials can be used as gas
sensitive layers, like metals, oxides, salts, polymers and organic
color dyes.
2. Gasochromic color dye
We investigated changes in the work function of two different
color dyes during exposure to NO2 and NH3, respectively. For NO2
detection we used N’,N’,N,N-tetramethyl- p-phenylenediamine (TMPD).
TMPD is a para-phenylenediamin and belongs to the group of
chinonimin dyes, also known as Kovac’s reagent or Wurster’s blue.
The color change is a double-stage reaction. Due to the oxidation
caused by NO2, the para-phenylenediamin provides an electron and
forms the blue cation. By providing a second electron, the cation
forms colorless chinondiimin. The equilibrium is on the side of the
first, blue cation. Fig. 1 shows the complete double-step reaction
mechanism.
Fig. 1: Two-step oxidation of TMPD due to NO2. The reaction
results in a reversible color change from brown to blue.
For NH3-detection the pH-indicator bromphenol blue (BPB) was
used. Its reaction to gaseous ammonia is shown
in fig. 2. The strong basicity of NH3 leads to a splitting of
the hydroxyl group of acid BPB. This acid-base reaction results in
a visible color change, as the protonated form is yellow and the
deprotonated one blue.
Fig. 2: Acid-base reaction of gaseous ammonia to the
pH-Indicator BPB. The color dye changes its color reversible from
yellow (protonated form) to blue (deprotonated form).
3. Experimental
This section describes the development of the two different
color dye matrices and the sample preparation for Kelvin probe
measurements.
3.1. Sample preparation of kelvin probes
To obtain stable gas sensitive films the color dyes have been
embedded into gas permeable and optically transparent polymer
matrices. TMPD has been embedded into a poly vinyl chloride (PVC)
matrix. Therefore, 10 mg of PVC powder were diluted bit by bit in
100 ml THF under ambient condition. After complete dilution 2 g
TMPD were added. To avoid rough and cracked films, 10 ml
plasticizer (Hexamoll Dinch, provided by BASF) were diluted
too.
-
110 Carolin Peter et al. / Procedia Engineering 87 ( 2014 ) 108
– 111
The BPB solution was made under similar conditions. Instead of
PVC we used an ethyl cellulose matrix. Therefore, 125 mg ethyl
cellulose were stirred into 5 ml ethanol under ambient conditions
for 1 hour. After complete dilution we added 425 µl
tributyl-phosphate as plasticizer and 20 mg BPB.
For sample preparation 100 µl of each solution were pipetted
directly on the substrates. After drying for 24 hours to provide
evaporation of solvent residues, the substrates are prepared for
the measurement.
3.2. Screen-printing of NH3 matrix
Investigations have been performed on screen-printing of the
films as an alternative deposition method. Screen-printing offers
the possibility of simultaneous deposition of many samples. For
this technique, the matrix has to by highly viscous. The color dye
was also diluted into a matrix of ethyl cellulose. But, compared to
matrix above, 4 g ethyl cellulose were stirred into 40 ml ethanol.
After complete dilution 1.5 ml plasticizer (tributyl-phosphate) and
20 mg pH indicator were added. For first tests we added the pH
indicator BCG (bromcresol green) instead of BPB. They have the same
working principle, but different points of changes.
The meshes of the used screen have the dimensions of 0.6x1.35
mm2. There are always two meshes directly next to each other with a
space of 0.2 mm. Conventional Si-wafers were used as substrates.
The squeegee was pressed with 2.5 bar over the substrate. Fig. 3
(left) shows a picture of the screen-printed layer directly after
the printing process, fig. 3 (right) the according profilometer
scan in transverse direction. The resulting layers are homogeneous
and have an average thickness of 1.8 µm. The edge regions are, by
the drying of the paste, wavy, with a lateral extent of a few
microns.
Fig. 3: Left: picture of the screen-printed layer directly after
printing. Right: profilometer scan of one rectangle in transverse
direction. The
layers have an average thickness of 1,8 µm.
4. Measurements and results
The following measurements were carried out using a Kelvin probe
in ambient filtered pressurized air with 40% r.H. and ambient room
conditions.
Fig. 4 (left) shows the measured work function changes of TMPD
under NO2 atmosphere, in steps of 0.3 ppm, 0.5 ppm, 1 ppm, 3 ppm
and 10 ppm NO2 each one for 30 minutes. After gas reaction, the
layers were exposed to synthetic air again. The gas exposure leads
to changes of 50 mV for 0.5 ppm and 350 mV for 10 ppm NO2. Lower
gas concentrations than 0.5 ppm generate only a slow and time
shifted output signal.
Fig. 4 (right) shows the according changes in the work function
of BPB exposed to different NH3 concentrations from 3 ppm, 5 ppm,
10 ppm, 30 ppm to 50 ppm, each one also for 30 minutes. The
exposure of 3 ppm leads to changes in the work function of 10 mV,
higher gas concentrations of 50 ppm to higher voltages of 40
mV.
-
111 Carolin Peter et al. / Procedia Engineering 87 ( 2014 ) 108
– 111
Fig. 4: Left: changes in the work function of TMPD according to
the NO2 concentration between 0.3 ppm and 10 ppm. Right:
according
changes in the work function of BPB under NH3 deposition at
different concentrations between 3 ppm and 50 ppm. All measurement
were taken out in filtered pressurized air at 40 % r.H..
5. Conclusions
The results show that the work function change of gasochromic
color dyes are suitable for measuring gas concentrations. In
addition, the gas reaction is highly selective to the target gas. A
changing humidity influences the sensor signal due to the water
adsorption of the polymer matrix. These investigations show the
high capability of gasochromic materials as sensitive layers for
FET based gas sensors.
Acknowledgements
We would like to thank the German BMBF and the Leading Edge
Cluster Microtec Südwest for funding within the Project
Minergy.
References
[1] I. Eisele, T. Doll and M. Burgmair, Low power gas detection
with FET sensors, Sensors and Actuators B 78 (2001) 19-25 [2] C.
Peter, S. Schulz, M. Barth, M. Gempp, S. Rademacher an J.
Wöllenstein, Low-cost roll-to-roll colorimetric gas sensor system
for fire
detection, Soli-State Sensors, Actuators and Microsystems, 2013
Transducers & Eurosensors [3] J. Courbat, D. Briand,
J.Wöllenstein, N.F. de Rooij, Colorimetric gas sensors based on
optical waveguides made on plastic foil, Procedia
Chemistry (2009) 576-579
0,0
0,1
0,2
0,3
0,4
0 1 2 3 4 5 6 7 80306090
Vou
t
TMPD
gas
conc
entra
tion
time / h
humidity / % NO
2 (0,1 ppm)
-0,06
-0,04
-0,02
0,00
0,02
0 1 2 3 4 5 601020304050
Vou
t
BPB
gasf
low
/ pp
m
time / h
humidity / % NH
3