Available via license: CC BY 4.0
Content may be subject to copyright.
sensors
Article
Dual-Mode Gas Sensor Composed of a Silicon
Nanoribbon Field Effect Transistor and a Bulk
Acoustic Wave Resonator: A Case Study in Freons
Ye Chang 1, Zhipeng Hui 2, Xiayu Wang 1, Hemi Qu 1,*, Wei Pang 1and Xuexin Duan 1,*ID
1
State Key Laboratory of Precision Measuring Technology & Instruments, Tianjin University, Tianjin 300072,
China; cy0803@tju.edu.cn (Y.C.); xy_wang@tju.edu.cn (X.W.); weipang@tju.edu.cn (W.P.)
2China Marine Development and Research Center (CMDRC), Beijing 100049, China; hzpeng117@163.com
*
Correspondence: hemi.qu@tju.edu.cn (H.Q.); xduan@tju.edu.cn (X.D.); Tel.: +86-22-2740-1002 (H.Q. & X.D.)
Received: 27 December 2017; Accepted: 14 January 2018; Published: 25 January 2018
Abstract:
In this paper, we develop a novel dual-mode gas sensor system which comprises
a silicon nanoribbon field effect transistor (Si-NR FET) and a film bulk acoustic resonator
(FBAR). We investigate their sensing characteristics using polar and nonpolar organic compounds,
and demonstrate that polarity has a significant effect on the response of the Si-NR FET sensor,
and only a minor effect on the FBAR sensor. In this dual-mode system, qualitative discrimination can
be achieved by analyzing polarity with the Si-NR FET and quantitative concentration information can
be obtained using a polymer-coated FBAR with a detection limit at the ppm level. The complementary
performance of the sensing elements provides higher analytical efficiency. Additionally, a dual mixture
of two types of freons (CFC-113 and HCFC-141b) is further analyzed with the dual-mode gas sensor.
Owing to the small size and complementary metal-oxide semiconductor (CMOS)-compatibility of the
system, the dual-mode gas sensor shows potential as a portable integrated sensing system for the
analysis of gas mixtures in the future.
Keywords:
freons detection; bulk acoustic wave resonator; field effect transistor; dual-mode sensing;
gas sensor
1. Introduction
Micro sensors are small-sized devices with flexibility, high sensitivity, moderate cost, and reduced
power consumption. During the past decades, a variety of micro sensors have been employed
in gas-sensing applications [
1
,
2
], including gravimetric [
3
], electrochemical [
4
,
5
], and optical [
6
]
sensors. Recent studies in systems based on micro gas sensors have demonstrated their capabilities in
analyzing breath samples for disease diagnosis [
7
,
8
], assessing the authenticity of premium products [
9
],
evaluating the freshness and maturity of food [
10
,
11
], assuring quality in product manufacturing [
12
],
etc. However, in most applications, the sensing system is composed of one type of sensor or
utilizes one type of transduction mode, which only discloses partial information about the analytes.
The combination of different types of sensors or transduction modes could provide complementary
information to improve the sensor’s selectivity and facilitate the discrimination of gas mixtures [
13
–
15
].
A field effect transistor (FET) is a type of electrochemical sensor which offers charge-based
information, such as the dipole moment inherent in analytes. To date, different types of nanostructure-
based FETs, such as nanowires [
16
,
17
], nanorods [
18
,
19
], nanotubes [
20
,
21
], and nanoribbons [
22
]
have been applied for gas sensing. In particular, silicon nanoribbon field effect transistors (Si-NR
FETs) fabricated using complementary metal-oxide semiconductor (CMOS)-compatible processes have
demonstrated high effectiveness and a superb signal-to-noise ratio, with the limit of detection (LOD) as
low as sub-ppm [
23
]. Thus, the device is reliable in polarity sensing even when the analytes are at the
Sensors 2018,18, 343; doi:10.3390/s18020343 www.mdpi.com/journal/sensors
Sensors 2018,18, 343 2 of 12
trace level. A bulk acoustic wave resonator (BAW) is a type of gravimetric sensor. The sensitivity and
selectivity of a BAW can be tuned by chemical surface coating. The gravimetric-based measurement
leads to a sensing signal, which carries information about the adsorption of analytes [
24
]. In recent
years, rapid progress has been made in the field of gas sensing using thin film bulk acoustic resonators
(FBARs) [
25
–
29
]. Due to their GHz-level resonant frequency and high quality-factor, FBARs show
higher sensitivity than conventional acoustic wave devices [30,31].
Freons are halocarbon-based chemicals extensively used as refrigerants, foaming agents,
and cleaning solvents for domestic and industrial applications. The excessive emission of freon gases,
including partially halogenated hydrochlorofluorocarbons (HCFCs) and chlorofluorocarbons (CFCs),
is considered destructive to the ozone layer and contributes to the greenhouse effect [
32
]. Table 1
lists the physical properties of two typical freons (CFC-113 and HCFC-141b) and their effects on the
atmospheric environment. Since those two freons possess similar physical properties, they are difficult to
distinguish [
33
]. Conventional methods of analyzing freons are gas chromatography-mass spectrometry
(GC-MS) [
34
,
35
] and Fourier transform infrared (FTIR) spectroscopy [
36
,
37
]. These methods can provide
qualitative and quantitative analysis of freons; however, they are limited by their bulky size and high
capital cost for on-line or portable gas analysis [38,39].
Table 1.
Atmospheric lifetimes, ozone depletion potentials (ODPs), global warming potentials (GWPs)
[40,41], and other typical physical properties of CFC-113 and HCFC-141b.
Freon Atmospheric
Lifetime (Years) ODP GWP Volume 1(Å3)Boiling Point (K) Critical
Pressure (MPa)
CFC-113 93 0.81 6080 96.142 320.754 3.820
HCFC-141b
9.4 0.11 717 74.304 300.129 4.386
1Volume was calculated as the Connolly solvent excluded volume.
In this work, we fabricated p-type Si-NR FETs as well as 2.44 GHz FBAR sensors, and incorporated
these two types of sensors into a dual-mode gas sensing system for the analysis of CFC-113 and
HCFC-141b. Experimental results show that qualitative discrimination can be realized by the Si-NR
FET due to its polarity-sensitivity, while the quantitative concentration detection of freons can be
realized by a polymer-coated FBAR through gravimetric measurement. Moreover, freon mixtures are
analyzed based on the complementary characteristics of the two types of sensors. Our results suggest
that the proposed dual-mode gas sensor has potential for the quick and accurate detection of CFC-113
and HCFC-141b, as well as other dual-component gas mixtures, in the future.
2. Materials and Methods
2.1. Chemicals
Freons (CFC-113 and HCFC-141b), ethanol, hexane, and polyisobutene (PIB, Mw = 600,000)
utilized in this work were purchased from J&K Chemical (Beijing, China) in analytical purity (99%)
and used without further purification.
2.2. Experimental Setup
Figure 1shows the experimental setup used in this work. The setup consists of two parts,
a dual-line vapor delivery system and a testing system. In the vapor delivery system, vapors were
bubbled out of liquid with carrier nitrogen gas (99.999%). Freon vapor in varied concentrations was
prepared by adjusting the flow rate of nitrogen from the dilution line via a mass flow controller (MFC,
5850e, Brooks, Pacific Grove, CA, USA). A flowmeter was used to monitor the real-time flowrate at
the end of the system. Experiments were performed with exposure to freon vapors at concentrations
in terms of P/P
0
from 0.1 to 0.6, where P and P
0
represent the partial pressure and the saturated
vapor pressure of the vapor of interest, respectively. According to [
42
], concentrations of vapors at
Sensors 2018,18, 343 3 of 12
P/P
0
from 0.1 to 0.6 are calculated to be at a several thousand ppm level. Freon vapors at lower
concentrations ranging from 1 to 20 ppm were produced by a commercial vapor generator system
(MF-3D, NIM, Beijing, China). In the testing system, a Si-NR FET was wirebonded onto a 28-pin
dual in-line package and source-drain as well as gate voltages (V
g
) were applied using a source
meter (2636, Keithley, Taunton, MA, USA). The source-drain current (I
ds
) was measured with the
source meter. Meanwhile, a FBAR sensor was wirebonded onto an evaluation board and connected
to a vector network analyzer (VNA, E5071C, Agilent, Palo Alto, CA, USA) for frequency domain
measurements. The two types of sensors were fabricated independently and packed together in a
stainless-steel chamber after fabrication. Real-time sensing data of both sensors were recorded by a
personal computer (PC). Ultraviolet light illumination was performed after each experiment to induce
freon molecules desorption and keep the sensor surface fresh.
Sensors 2018, 18, x FOR PEER REVIEW 3 of 12
of the vapor of interest, respectively. According to [42], concentrations of vapors at P/P
0
from 0.1 to
0.6 are calculated to be at a several thousand ppm level. Freon vapors at lower concentrations ranging
from 1 to 20 ppm were produced by a commercial vapor generator system (MF-3D, NIM, Beijing,
China). In the testing system, a Si-NR FET was wirebonded onto a 28-pin dual in-line package and
source-drain as well as gate voltages (V
g
) were applied using a source meter (2636, Keithley, Taunton,
MA, USA). The source-drain current (I
ds
) was measured with the source meter. Meanwhile, a FBAR
sensor was wirebonded onto an evaluation board and connected to a vector network analyzer (VNA,
E5071C, Agilent, Palo Alto, CA, USA) for frequency domain measurements. The two types of sensors
were fabricated independently and packed together in a stainless-steel chamber after fabrication.
Real-time sensing data of both sensors were recorded by a personal computer (PC). Ultraviolet light
illumination was performed after each experiment to induce freon molecules desorption and keep
the sensor surface fresh.
Figure 1. Illustration of the experimental setup for the detection of freons at P/P
0
from 0.1 to 0.6 with
both silicon nanoribbon field effect transistor (Si-NR FET) and thin film bulk acoustic resonator
(FBAR) sensors. MFC: mass flow controller; VNA: vector network analyzer; PC: personal computer.
2.3. Device Fabrication and Functionalization
As shown in Figure 2a–e, the p-type Si-NR FET in this work was fabricated by top-down
lithographic techniques on a 4 inch silicon-on-insulator wafer as described in [43]. Briefly, the silicon
active layer (boron-doped, carrier concentration = 10
15
cm
−3
) was thinned to 45 nm using thermal
oxidation and a buffered oxide etch. Subsequently, the source and drain regions, as well as the back-
gate, were patterned and further doped by BF
2+
implantation. The nanoribbon mesas (1 × 10 μm
2
sensing area) were then defined by optical lithography and reactive ion etching (RIE) using Cl
2
inductively coupled plasma. An oxidation furnace was used to grow a 20 nm thick layer of silicon
oxide over the wafer. Metallization was then performed by evaporating Al and patterning by lift-off.
After annealing, the metal contacts were measured to ensure ohmic contacts. Finally, a SU-8 layer
was patterned to passivate the devices, with openings to NRs and contact pads.
As shown in Figure 2f–j, the 2.44 GHz FBAR was fabricated with a standard
microelectromechanical system (MEMS) fabrication process as described before [26]. First, an air
cavity was etched on a single-side polished silicon wafer by deep reactive ion etching (DRIE).
Subsequently, the cavity was filled with phosphosilicate glass (PSG) as a sacrificial layer via chemical
vapor deposition (CVD). A sandwiched structure of Mo/AlN/Mo was then deposited as the bottom
electrode, piezoelectric layer, and top electrode, respectively. After the deposition of a passivation
layer (AlN), a Cr/Au composite film was then deposited and patterned to form contact pads. Finally,
PSG in the cavity was removed by immersing the chip in a diluted HF solution. After fabrication, the
FBAR chip was spin-coated with a layer of PIB (1 mg/mL in chloroform; 4000 rpm for 30 s) and
annealed at 65 °C for 1 h. The PIB layer was then characterized by FTIR spectroscopy (Vertex 70v,
Figure 1.
Illustration of the experimental setup for the detection of freons at P/P
0
from 0.1 to 0.6 with
both silicon nanoribbon field effect transistor (Si-NR FET) and thin film bulk acoustic resonator (FBAR)
sensors. MFC: mass flow controller; VNA: vector network analyzer; PC: personal computer.
2.3. Device Fabrication and Functionalization
As shown in Figure 2a–e, the p-type Si-NR FET in this work was fabricated by top-down
lithographic techniques on a 4 inch silicon-on-insulator wafer as described in [
43
]. Briefly, the silicon
active layer (boron-doped, carrier concentration = 10
15
cm
−3
) was thinned to 45 nm using thermal
oxidation and a buffered oxide etch. Subsequently, the source and drain regions, as well as the
back-gate, were patterned and further doped by BF
2+
implantation. The nanoribbon mesas (
1×10 µm2
sensing area) were then defined by optical lithography and reactive ion etching (RIE) using Cl
2
inductively coupled plasma. An oxidation furnace was used to grow a 20 nm thick layer of silicon
oxide over the wafer. Metallization was then performed by evaporating Al and patterning by lift-off.
After annealing, the metal contacts were measured to ensure ohmic contacts. Finally, a SU-8 layer was
patterned to passivate the devices, with openings to NRs and contact pads.
As shown in Figure 2f–j, the 2.44 GHz FBAR was fabricated with a standard
microelectromechanical system (MEMS) fabrication process as described before [
26
]. First, an air
cavity was etched on a single-side polished silicon wafer by deep reactive ion etching (DRIE).
Subsequently, the cavity was filled with phosphosilicate glass (PSG) as a sacrificial layer via chemical
vapor deposition (CVD). A sandwiched structure of Mo/AlN/Mo was then deposited as the bottom
electrode, piezoelectric layer, and top electrode, respectively. After the deposition of a passivation
layer (AlN), a Cr/Au composite film was then deposited and patterned to form contact pads. Finally,
PSG in the cavity was removed by immersing the chip in a diluted HF solution. After fabrication,
the FBAR chip was spin-coated with a layer of PIB (1 mg/mL in chloroform; 4000 rpm for 30 s) and
annealed at 65
◦
C for 1 h. The PIB layer was then characterized by FTIR spectroscopy (Vertex 70v,
Sensors 2018,18, 343 4 of 12
Bruker Optics, Ettlingen, Germany) and atomic force microscopy (AFM, Dimension Icon, Bruker,
Rheinstetten, Germany) in tapping mode.
Sensors 2018, 18, x FOR PEER REVIEW 4 of 12
Bruker Optics, Ettlingen, Germany) and atomic force microscopy (AFM, Dimension Icon, Bruker,
Rheinstetten, Germany) in tapping mode.
Figure 2. Schematics of the fabrication process of (a–e) Si-NR FET and (f–j) FBAR. (a) Thermal
oxidation of wafer; (b) thinning of silicon active layer; (c) implantation of source, drain, and back-gate
and etching of nanoribbon; (d) deposition of front oxide and contact pads; (e) deposition of
passivation layer; (f) etching of air cavity and deposition of passivation layer; (g) deposition of
sandwiched structure; (h) deposition of passivation layer; (i) deposition of contact pads and release
of phosphosilicate glass (PSG); (j) spin-coating of polyisobutene (PIB).
3. Results
3.1. Sensor Characterizations
As mentioned above, the dual-mode gas sensor used in this work is composed of a Si-NR FET
(Figure 3a) and a FBAR sensor (Figure 3b). Owing to the miniaturization and CMOS-compatible
fabrication, both sensors show potential for large-scale integration into portable dual-mode sensing
systems. Figure 3c presents the Ids-Vg curve of the Si-NR FET sensor. The device shows a typical p-
type behavior, which means that the channel carrier density is tuned with a negative bias voltage
applied to the back-gate. Thus, the adsorption of polar molecules can be directly detected in the
accumulation mode of the device due to the surface potential change. In contrast, the sensing
mechanism of the FBAR sensor is based on the mass loading effect, and the resonant frequency
decreases with the loading of analytes. According to the Sauerbrey equation [44], high operation
frequency gives the FBAR sensor a much higher sensitivity than conventional quartz crystal
microbalance (QCM) [45]. The quality factor of a resonator is a parameter that describes energy losses
over time, and is related to the sensing performance of the device [46]. Figure 3d presents the quality-
factor curve of FBAR. The highest quality-factor is 1158 at the operation frequency of 2.44 GHz,
indicating that the acoustic energy is well-trapped in the piezoelectric layer.
In order to further improve the sensitivity of the bare FBAR sensor [28], a polymer layer of PIB
[47] was spin-coated on top of the device. The surface was characterized with FTIR spectroscopy and
AFM. Figure 4 presents the characterization results of the PIB layer. The spectrum (Figure 4a) shows
reflection peaks at 1366 and 1389 cm−1 due to symmetrical deformation vibrations of C-(CH3)2 as well
as reflection peaks at 2925 and 2952 cm−1 due to the asymmetrical stretching vibrations of -CH2- and
(a)
Silicon PSG Mo
AlN Cr/Au PIB
Implant
Silicon
Al
SiO
2
Active Si SU8
(f)
(b)(g)
(c)(h)
(d)(i)
(e)(j)
Si-NR FET FBAR
Figure 2.
Schematics of the fabrication process of (
a
–
e
) Si-NR FET and (
f
–
j
) FBAR. (
a
) Thermal oxidation
of wafer; (
b
) thinning of silicon active layer; (
c
) implantation of source, drain, and back-gate and etching
of nanoribbon; (
d
) deposition of front oxide and contact pads; (
e
) deposition of passivation layer;
(
f
) etching of air cavity and deposition of passivation layer; (
g
) deposition of sandwiched structure;
(
h
) deposition of passivation layer; (
i
) deposition of contact pads and release of phosphosilicate glass
(PSG); (j) spin-coating of polyisobutene (PIB).
3. Results
3.1. Sensor Characterizations
As mentioned above, the dual-mode gas sensor used in this work is composed of a Si-NR FET
(Figure 3a) and a FBAR sensor (Figure 3b). Owing to the miniaturization and CMOS-compatible
fabrication, both sensors show potential for large-scale integration into portable dual-mode sensing
systems. Figure 3c presents the I
ds
-V
g
curve of the Si-NR FET sensor. The device shows a typical p-type
behavior, which means that the channel carrier density is tuned with a negative bias voltage applied to
the back-gate. Thus, the adsorption of polar molecules can be directly detected in the accumulation
mode of the device due to the surface potential change. In contrast, the sensing mechanism of the
FBAR sensor is based on the mass loading effect, and the resonant frequency decreases with the
loading of analytes. According to the Sauerbrey equation [
44
], high operation frequency gives the
FBAR sensor a much higher sensitivity than conventional quartz crystal microbalance (QCM) [
45
].
The quality factor of a resonator is a parameter that describes energy losses over time, and is related
to the sensing performance of the device [
46
]. Figure 3d presents the quality-factor curve of FBAR.
The highest quality-factor is 1158 at the operation frequency of 2.44 GHz, indicating that the acoustic
energy is well-trapped in the piezoelectric layer.
In order to further improve the sensitivity of the bare FBAR sensor [
28
], a polymer layer of
PIB [47] was spin-coated on top of the device. The surface was characterized with FTIR spectroscopy
and AFM. Figure 4presents the characterization results of the PIB layer. The spectrum (Figure 4a)
shows reflection peaks at 1366 and 1389 cm
−1
due to symmetrical deformation vibrations of C-(CH
3
)
2
as well as reflection peaks at 2925 and 2952 cm
−1
due to the asymmetrical stretching vibrations of
Sensors 2018,18, 343 5 of 12
-CH
2
- and -CH
3
. In the AFM image (Figure 4b), a thickness of 58 nm and separate domains with
6~8
µ
m gaps can be found. These characterization results confirm a uniform coating of PIB on the
FBAR sensor.
Sensors 2018, 18, x FOR PEER REVIEW 5 of 12
-CH
3
. In the AFM image (Figure 4b), a thickness of 58 nm and separate domains with 6~8 μm gaps
can be found. These characterization results confirm a uniform coating of PIB on the FBAR sensor.
Figure 3. Illustration of the dual-mode gas sensor. (a,b) present the optical microscope images of the
Si-NR FET array and the scanning electron microscopy (SEM) image of an FBAR configuration; (c)
and (d) describe the I
ds
-V
g
characteristic of the Si-NR FET sensor and the quality-factor curve of the
FBAR sensor.
Figure 4. (a) FTIR spectroscopy result and (b) AFM image of the PIB layer coated on the FBAR sensor.
3.2. Si-NR FET Results
In the dual-mode gas sensor, the Si-NR FET sensor was incorporated to provide polarity
information for the analytes. We first examined the influence of polarity on the FET response using
ethanol and hexane as test samples. Figure 5 shows the real-time detection results of polar ethanol
(dipole moment = 1.85) and nonpolar hexane (dipole moment = 0) at the same concentration with a
Si-NR FET device. With the introduction of ethanol at 1 min, the drain current of the Si-NR FET sensor
significantly increases. In contrast, the drain current does not change even with a continuous flow of
hexane. The results confirm the polarity-sensitivity of the Si-NR FET sensor. For a Si-NR FET device,
the response of the device depends on carrier concentration inside the Si-NR, which can be directly
modulated by electrostatic interaction between a polar vapor molecule and the device. The long
response time of the Si-NR FET sensor to ethanol is attributed to the chemical reaction kinetics
Figure 3.
Illustration of the dual-mode gas sensor. (
a
,
b
) present the optical microscope images of
the Si-NR FET array and the scanning electron microscopy (SEM) image of an FBAR configuration;
(
c
) and (
d
) describe the I
ds
-V
g
characteristic of the Si-NR FET sensor and the quality-factor curve of the
FBAR sensor.
Sensors 2018, 18, x FOR PEER REVIEW 5 of 12
-CH
3
. In the AFM image (Figure 4b), a thickness of 58 nm and separate domains with 6~8 μm gaps
can be found. These characterization results confirm a uniform coating of PIB on the FBAR sensor.
Figure 3. Illustration of the dual-mode gas sensor. (a,b) present the optical microscope images of the
Si-NR FET array and the scanning electron microscopy (SEM) image of an FBAR configuration; (c)
and (d) describe the I
ds
-V
g
characteristic of the Si-NR FET sensor and the quality-factor curve of the
FBAR sensor.
Figure 4. (a) FTIR spectroscopy result and (b) AFM image of the PIB layer coated on the FBAR sensor.
3.2. Si-NR FET Results
In the dual-mode gas sensor, the Si-NR FET sensor was incorporated to provide polarity
information for the analytes. We first examined the influence of polarity on the FET response using
ethanol and hexane as test samples. Figure 5 shows the real-time detection results of polar ethanol
(dipole moment = 1.85) and nonpolar hexane (dipole moment = 0) at the same concentration with a
Si-NR FET device. With the introduction of ethanol at 1 min, the drain current of the Si-NR FET sensor
significantly increases. In contrast, the drain current does not change even with a continuous flow of
hexane. The results confirm the polarity-sensitivity of the Si-NR FET sensor. For a Si-NR FET device,
the response of the device depends on carrier concentration inside the Si-NR, which can be directly
modulated by electrostatic interaction between a polar vapor molecule and the device. The long
response time of the Si-NR FET sensor to ethanol is attributed to the chemical reaction kinetics
Figure 4.
(
a
) FTIR spectroscopy result and (
b
) AFM image of the PIB layer coated on the FBAR sensor.
3.2. Si-NR FET Results
In the dual-mode gas sensor, the Si-NR FET sensor was incorporated to provide polarity
information for the analytes. We first examined the influence of polarity on the FET response using
ethanol and hexane as test samples. Figure 5shows the real-time detection results of polar ethanol
(dipole moment = 1.85) and nonpolar hexane (dipole moment = 0) at the same concentration with a
Si-NR FET device. With the introduction of ethanol at 1 min, the drain current of the Si-NR FET sensor
significantly increases. In contrast, the drain current does not change even with a continuous flow of
hexane. The results confirm the polarity-sensitivity of the Si-NR FET sensor. For a Si-NR FET device,
Sensors 2018,18, 343 6 of 12
the response of the device depends on carrier concentration inside the Si-NR, which can be directly
modulated by electrostatic interaction between a polar vapor molecule and the device. The long
response time of the Si-NR FET sensor to ethanol is attributed to the chemical reaction kinetics between
molecules and oxygen ions adsorbed on the surface of the nanomaterials [
48
]. In the following tests,
the response time was shortened to 1 min, which still met the demand of detection.
Sensors 2018, 18, x FOR PEER REVIEW 6 of 12
between molecules and oxygen ions adsorbed on the surface of the nanomaterials [48]. In the
following tests, the response time was shortened to 1 min, which still met the demand of detection.
Figure 5. Real-time responses of a Si-NR FET device to ethanol and hexane at P/P
0
= 0.1.
We then simulated CFC-113 and HCFC-141b using ChemBio3D with the GAMESS interface [49].
In general, the properties of CFC-113 and HCFC-141b are found to be quite similar except for polarity.
Figure 6 shows three-dimensional (3D) molecular structures of the two freons with their dipole
moments. The simulation results indicate that HCFC-141b has higher polarity due to the partially
halogenated structure, while CFC-113 has negligible polarity due to the nearly symmetric
halogenated groups.
Figure 6. Three-dimensional (3D) molecular structure of (a) CFC-113 and (b) HCFC-141b. The
magenta arrow indicates the orientation and value of the dipole moment for each structure based on
calculations using ChemBio3D with the GAMESS interface.
Figure 7a,b show the real-time responses of the Si-NR FET sensor to CFC-113 and HCFC-141b
vapors at different concentrations. The y-axis scale is kept the same between the two figures for
comparison. It can be seen that the measured current (I
ds
) significantly increases once the sensor is
exposed to HCFC-141b, while for CFC-113 the current keeps nearly constant during the entire
sampling period. As discussed previously, the poor response of the Si-NR FET sensor to CFC-113 can
be attributed to the weak polarity of the analyte. In contrast, a stronger polarity of HCFC-141b results
in a much more obvious response from the Si-NR FET sensor (Figure 7c). These results suggest that
CFC-113 and HCFC-141b can be effectively distinguished with the Si-NR FET sensor.
-101234567891011
0.00
0.02
0.04
0.06
ΔI
ds
(μA)
Time (min)
Ethanol
Hexane
Figure 5. Real-time responses of a Si-NR FET device to ethanol and hexane at P/P0= 0.1.
We then simulated CFC-113 and HCFC-141b using ChemBio3D with the GAMESS interface [
49
].
In general, the properties of CFC-113 and HCFC-141b are found to be quite similar except for
polarity. Figure 6shows three-dimensional (3D) molecular structures of the two freons with their
dipole moments. The simulation results indicate that HCFC-141b has higher polarity due to the
partially halogenated structure, while CFC-113 has negligible polarity due to the nearly symmetric
halogenated groups.
Sensors 2018, 18, x FOR PEER REVIEW 6 of 12
between molecules and oxygen ions adsorbed on the surface of the nanomaterials [48]. In the
following tests, the response time was shortened to 1 min, which still met the demand of detection.
Figure 5. Real-time responses of a Si-NR FET device to ethanol and hexane at P/P
0
= 0.1.
We then simulated CFC-113 and HCFC-141b using ChemBio3D with the GAMESS interface [49].
In general, the properties of CFC-113 and HCFC-141b are found to be quite similar except for polarity.
Figure 6 shows three-dimensional (3D) molecular structures of the two freons with their dipole
moments. The simulation results indicate that HCFC-141b has higher polarity due to the partially
halogenated structure, while CFC-113 has negligible polarity due to the nearly symmetric
halogenated groups.
Figure 6. Three-dimensional (3D) molecular structure of (a) CFC-113 and (b) HCFC-141b. The
magenta arrow indicates the orientation and value of the dipole moment for each structure based on
calculations using ChemBio3D with the GAMESS interface.
Figure 7a,b show the real-time responses of the Si-NR FET sensor to CFC-113 and HCFC-141b
vapors at different concentrations. The y-axis scale is kept the same between the two figures for
comparison. It can be seen that the measured current (I
ds
) significantly increases once the sensor is
exposed to HCFC-141b, while for CFC-113 the current keeps nearly constant during the entire
sampling period. As discussed previously, the poor response of the Si-NR FET sensor to CFC-113 can
be attributed to the weak polarity of the analyte. In contrast, a stronger polarity of HCFC-141b results
in a much more obvious response from the Si-NR FET sensor (Figure 7c). These results suggest that
CFC-113 and HCFC-141b can be effectively distinguished with the Si-NR FET sensor.
-101234567891011
0.00
0.02
0.04
0.06
ΔI
ds
(μA)
Time (min)
Ethanol
Hexane
Figure 6.
Three-dimensional (3D) molecular structure of (
a
) CFC-113 and (
b
) HCFC-141b. The magenta
arrow indicates the orientation and value of the dipole moment for each structure based on calculations
using ChemBio3D with the GAMESS interface.
Figure 7a,b show the real-time responses of the Si-NR FET sensor to CFC-113 and HCFC-141b
vapors at different concentrations. The y-axis scale is kept the same between the two figures for
comparison. It can be seen that the measured current (I
ds
) significantly increases once the sensor
is exposed to HCFC-141b, while for CFC-113 the current keeps nearly constant during the entire
Sensors 2018,18, 343 7 of 12
sampling period. As discussed previously, the poor response of the Si-NR FET sensor to CFC-113 can
be attributed to the weak polarity of the analyte. In contrast, a stronger polarity of HCFC-141b results
in a much more obvious response from the Si-NR FET sensor (Figure 7c). These results suggest that
CFC-113 and HCFC-141b can be effectively distinguished with the Si-NR FET sensor.
Sensors 2018, 18, x FOR PEER REVIEW 7 of 12
Figure 7. Plots of real-time responses of the Si-NR FET sensor to (a) CFC-113 and (b) HCFC-141b in
the same scale at different concentrations (responses of the Si-NR FET sensor to CFC-113 in the
appropriate scale are shown in the insert); (c) Ids shifts of the Si-NR FET sensor at different
concentration ranges.
3.3. FBAR Results
In order to obtain quantitative concentration information for freons regardless of polarity, we
applied a PIB-coated FBAR for CFC-113 and HCFC-141b detection. Figure 8a,b respectively show the
real-time responses of FBAR to CFC-113 and HCFC-141b vapor from P/P0 = 0.1 to 0.6. As can be seen
from comparing the figures, HCFC-141b shows larger frequency shifts than CFC-113. To further
understand this phenomena, we investigated the adsorption isotherms by fitting data with the
Brunauer–Emmett–Teller (BET) equation [50], a typical model of multilayer gas physisorptions:
() ()
2
00
0
m
P
P
1C
P
P
2C1
C
P
P
v
v
−−
−+
=
where v, vm, P, and P0 are the adsorption capacity, the monolayer adsorption capacity, the vapor
pressures, and the saturated vapor pressure of freons, respectively, and C is the adsorption energy
constant. The excellent fit of the BET equation to the frequency data shown in in Figure 8c indicates
that freons molecules are mainly adsorbed on the surface of PIB by multilayer molecule stacking at
high concentration [51].
0.00.51.01.52.02.5
0.0
0.5
1.0
1.5
2.0
Δ
I
ds
(μA)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
0.00.51.01.52.02.5
0.0
0.5
1.0
1.5
2.0
Δ
I
ds
(μA)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
012
0.00
0.01
0.1 0.2 0.3 0.4 0.5 0.6
0.0
0.5
1.0
1.5
2.0
CFC-113
HCFC-141b
Δ
I
ds
(μA)
Concentration (P/P
0
)
(a)(b)
(c)
Figure 7.
Plots of real-time responses of the Si-NR FET sensor to (
a
) CFC-113 and (
b
) HCFC-141b in the
same scale at different concentrations (responses of the Si-NR FET sensor to CFC-113 in the appropriate
scale are shown in the insert); (c) Ids shifts of the Si-NR FET sensor at different concentration ranges.
3.3. FBAR Results
In order to obtain quantitative concentration information for freons regardless of polarity,
we applied a PIB-coated FBAR for CFC-113 and HCFC-141b detection. Figure 8a,b respectively
show the real-time responses of FBAR to CFC-113 and HCFC-141b vapor from P/P
0
= 0.1 to 0.6.
As can be seen from comparing the figures, HCFC-141b shows larger frequency shifts than CFC-113.
To further understand this phenomena, we investigated the adsorption isotherms by fitting data with
the Brunauer–Emmett–Teller (BET) equation [50], a typical model of multilayer gas physisorptions:
v=vmP
P0C
1+(C−2)P
P0−(C−1)P
P02
where v, v
m
, P, and P
0
are the adsorption capacity, the monolayer adsorption capacity, the vapor
pressures, and the saturated vapor pressure of freons, respectively, and C is the adsorption energy
constant. The excellent fit of the BET equation to the frequency data shown in in Figure 8c indicates
that freons molecules are mainly adsorbed on the surface of PIB by multilayer molecule stacking at
high concentration [51].
Sensors 2018,18, 343 8 of 12
Sensors 2018, 18, x FOR PEER REVIEW 8 of 12
Figure 8. Plots of real-time responses of the FBAR sensor to (a) CFC-113 and (b) HCFC-141b at
different concentrations from P/P0 = 0.1 to 0.6; (c) BET fitting of adsorption isotherms of freons on the
PIB layer.
To investigate the LOD of this method, the same sensor was used to detect a lower concentration
range of CFC-113 and HCFC-141b (here we used the ppm horizontal axis instead of P/P0). Figure 9a,b
respectively show the real-time responses of the FBAR sensor to CFC-113 and HCFC-141b vapors at
1 to 20 ppm. It is clear that the relationship between concentration and frequency change is linear
within this concentration range (Figure 9c). This linearity is caused by the unsaturated adsorption of
freons in the PIB polymer, and thus this adsorption can be modeled as monolayer adsorption. Owing
to the high adsorption capacity of the polymer coating and high operation frequency of the device,
the detection limit, typically calculated as three times the baseline noise level response, is estimated
to be around 4 ppm.
Figure 9. Plots of real-time responses of the FBAR sensor to (a) CFC-113 and (b) HCFC-141b at
concentrations from 1 to 20 ppm; (c) Linear fits of responses of the FBAR sensor to freons at low
concentration ranges.
(a)(b)
(c)
0.0 0.5 1.0 1.5 2.0 2.5
-2000
-1500
-1000
-500
0
Δ
F (kHz)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
0.0 0.5 1.0 1.5 2.0 2.5
-600
-400
-200
0
Δ
F (kHz)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
0.1 0.2 0.3 0.4 0.5 0.6
0
500
1000
1500
2000
−Δ
F (kHz)
Concentration (P/P0)
CFC-113
HCFC-141b
0.0 0.5 1.0 1.5 2.0 2.5
-10
-8
-6
-4
-2
0
2
Δ
F (kHz)
Time (min)
1 ppm
2 ppm
4 ppm
8 ppm
12 ppm
16 ppm
20 ppm
(a)(b)
0 2 4 6 8 10 12 14 16 18 20 22
0
2
4
6
−Δ
F (kHz)
Concentration (
pp
m)
CFC-113
HCFC-141b
(c)
0.0 0.5 1.0 1.5 2.0 2.5
-10
-8
-6
-4
-2
0
2
Δ
F (kHz)
Time (min)
1 ppm
2 ppm
4 ppm
8 ppm
12 ppm
16 ppm
20 ppm
Figure 8.
Plots of real-time responses of the FBAR sensor to (
a
) CFC-113 and (
b
) HCFC-141b at different
concentrations from P/P
0
= 0.1 to 0.6; (
c
) BET fitting of adsorption isotherms of freons on the PIB layer.
To investigate the LOD of this method, the same sensor was used to detect a lower concentration
range of CFC-113 and HCFC-141b (here we used the ppm horizontal axis instead of P/P
0
). Figure 9a,b
respectively show the real-time responses of the FBAR sensor to CFC-113 and HCFC-141b vapors at 1 to
20 ppm. It is clear that the relationship between concentration and frequency change is linear within
this concentration range (Figure 9c). This linearity is caused by the unsaturated adsorption of freons in
the PIB polymer, and thus this adsorption can be modeled as monolayer adsorption. Owing to the high
adsorption capacity of the polymer coating and high operation frequency of the device, the detection
limit, typically calculated as three times the baseline noise level response, is estimated to be around
4 ppm.
Sensors 2018, 18, x FOR PEER REVIEW 8 of 12
Figure 8. Plots of real-time responses of the FBAR sensor to (a) CFC-113 and (b) HCFC-141b at
different concentrations from P/P0 = 0.1 to 0.6; (c) BET fitting of adsorption isotherms of freons on the
PIB layer.
To investigate the LOD of this method, the same sensor was used to detect a lower concentration
range of CFC-113 and HCFC-141b (here we used the ppm horizontal axis instead of P/P0). Figure 9a,b
respectively show the real-time responses of the FBAR sensor to CFC-113 and HCFC-141b vapors at
1 to 20 ppm. It is clear that the relationship between concentration and frequency change is linear
within this concentration range (Figure 9c). This linearity is caused by the unsaturated adsorption of
freons in the PIB polymer, and thus this adsorption can be modeled as monolayer adsorption. Owing
to the high adsorption capacity of the polymer coating and high operation frequency of the device,
the detection limit, typically calculated as three times the baseline noise level response, is estimated
to be around 4 ppm.
Figure 9. Plots of real-time responses of the FBAR sensor to (a) CFC-113 and (b) HCFC-141b at
concentrations from 1 to 20 ppm; (c) Linear fits of responses of the FBAR sensor to freons at low
concentration ranges.
(a)(b)
(c)
0.0 0.5 1.0 1.5 2.0 2.5
-2000
-1500
-1000
-500
0
Δ
F (kHz)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
0.0 0.5 1.0 1.5 2.0 2.5
-600
-400
-200
0
Δ
F (kHz)
Time (min)
0.1
0.2
0.3
0.4
0.5
0.6
0.1 0.2 0.3 0.4 0.5 0.6
0
500
1000
1500
2000
−Δ
F (kHz)
Concentration (P/P0)
CFC-113
HCFC-141b
0.0 0.5 1.0 1.5 2.0 2.5
-10
-8
-6
-4
-2
0
2
Δ
F (kHz)
Time (min)
1 ppm
2 ppm
4 ppm
8 ppm
12 ppm
16 ppm
20 ppm
(a)(b)
0 2 4 6 8 10 12 14 16 18 20 22
0
2
4
6
−Δ
F (kHz)
Concentration (
pp
m)
CFC-113
HCFC-141b
(c)
0.0 0.5 1.0 1.5 2.0 2.5
-10
-8
-6
-4
-2
0
2
Δ
F (kHz)
Time (min)
1 ppm
2 ppm
4 ppm
8 ppm
12 ppm
16 ppm
20 ppm
Figure 9.
Plots of real-time responses of the FBAR sensor to (
a
) CFC-113 and (
b
) HCFC-141b at
concentrations from 1 to 20 ppm; (
c
) Linear fits of responses of the FBAR sensor to freons at low
concentration ranges.
Sensors 2018,18, 343 9 of 12
It is worthwhile to point out that the frequency response to HCFC-141b at high concentration
is larger than that of CFC-113 and this is opposite to the trend at low concentration. This is likely
due to the fact that the freons form multilayer adsorptions on the sensor’s surface when exposed to
high concentrations. In comparison, the adsorption of freons in the PIB polymer adopts a monolayer
adsorption model at low concentration. In the monolayer adsorption case, the interaction of freons
with PIB dominates, with both molecules showing similar behaviors, while for the multilayer stacking
case, the interactions of freons with themselves dominate, with CFC-113 and HCFC-141b showing a
large difference.
3.4. Mixture Analysis with the Dual-Mode Gas Sensor
To further investigate the performance of the dual-mode gas sensor, here we demonstrate the
qualitative and quantitative analysis of freons mixtures with the integrated complementary sensor
array. Analytes in this study were prepared by bubbling nitrogen gas through liquid mixtures of
CFC-113 and HCFC-141b with volume ratios of 3:1, 1:1, and 1:3. Figure 10 displays the results of
CFC-113, HCFC-141b, and their mixtures measured by the dual-mode gas sensor. As shown in
Figure 10a, the current response of the Si-NR FET sensor shows a strong positive correlation with the
HCFC-141b proportion in the freons mixture, which indicates that the component of freon analyte
can be directly discriminated with the device. Once the component is discriminated, concentration
detection is achieved with FBAR-based mass detection. As shown in Figure 10b, BET correlations
can be found for all vapors, confirming the excellent quantitative detection capacity of the FBAR
sensor. The combination of complementary miniaturized sensors provides the dual-mode gas sensor
with potential to become an integrated sensing system. Additionally, the novel gas sensor also holds
potential for the analysis of other dual-component gas mixtures composed of polar and nonpolar
vapor in the future.
Sensors 2018, 18, x FOR PEER REVIEW 9 of 12
It is worthwhile to point out that the frequency response to HCFC-141b at high concentration is
larger than that of CFC-113 and this is opposite to the trend at low concentration. This is likely due
to the fact that the freons form multilayer adsorptions on the sensor’s surface when exposed to high
concentrations. In comparison, the adsorption of freons in the PIB polymer adopts a monolayer
adsorption model at low concentration. In the monolayer adsorption case, the interaction of freons
with PIB dominates, with both molecules showing similar behaviors, while for the multilayer
stacking case, the interactions of freons with themselves dominate, with CFC-113 and HCFC-141b
showing a large difference.
3.4. Mixture Analysis with the Dual-Mode Gas Sensor
To further investigate the performance of the dual-mode gas sensor, here we demonstrate the
qualitative and quantitative analysis of freons mixtures with the integrated complementary sensor
array. Analytes in this study were prepared by bubbling nitrogen gas through liquid mixtures of
CFC-113 and HCFC-141b with volume ratios of 3:1, 1:1, and 1:3. Figure 10 displays the results of CFC-
113, HCFC-141b, and their mixtures measured by the dual-mode gas sensor. As shown in Figure 10a,
the current response of the Si-NR FET sensor shows a strong positive correlation with the HCFC-141b
proportion in the freons mixture, which indicates that the component of freon analyte can be directly
discriminated with the device. Once the component is discriminated, concentration detection is
achieved with FBAR-based mass detection. As shown in Figure 10b, BET correlations can be found
for all vapors, confirming the excellent quantitative detection capacity of the FBAR sensor. The
combination of complementary miniaturized sensors provides the dual-mode gas sensor with
potential to become an integrated sensing system. Additionally, the novel gas sensor also holds
potential for the analysis of other dual-component gas mixtures composed of polar and nonpolar
vapor in the future.
Figure 10. Freons mixtures detection by the dual-mode gas sensor. Three-dimensional bar plot of (a)
current responses of the Si-NR FET sensor and (b) frequency responses of the FBAR sensor to CFC-
113, HCFC-141b, and their three mixtures at different concentration ranges.
4. Conclusions
In summary, a novel dual-mode gas sensor composed of a charge-sensitive Si-NR FET sensor
and a mass-sensitive FBAR sensor was developed. Owing to the different sensing mechanisms, these
two types of sensors offer complementary information regarding the analytes. Such system was
successfully applied for the detection of CFC-113 and HCFC-141b. Because of the nonpolar property
of CFC-113, it shows negligible current response on the Si-NR FET sensor, while the Si-NR FET sensor
is very sensitive to the polar HCFC-141b. Thus, these two types of freon molecules can be effectively
distinguished with the gas sensor. In addition, the polymer-coated FBAR sensor shows excellent
quantitative detection capacity of the freon molecules with a ppm-level LOD. Mixtures of CFC-113
(a)(b)
0.1
0.2
0.3
0.4
0.5
0.6
0.0
0.5
1.0
1.5
2.0
CFC-113
3:1 Mixture
1:1 Mixture
1:3 Mixture
HCFC-141b
Δ
I
ds
(μA)
Concentration (P/P
0
)
0.1
0.2
0.3
0.4
0.5
0.6
0
500
1000
1500
2000
CFC-113
3:1 Mixture
1:1 Mixture
1:3 Mixture
HCFC-141b
−Δ
F (kHz)
Concentration (P/P
0
)
Figure 10.
Freons mixtures detection by the dual-mode gas sensor. Three-dimensional bar plot of
(
a
) current responses of the Si-NR FET sensor and (
b
) frequency responses of the FBAR sensor to
CFC-113, HCFC-141b, and their three mixtures at different concentration ranges.
4. Conclusions
In summary, a novel dual-mode gas sensor composed of a charge-sensitive Si-NR FET sensor
and a mass-sensitive FBAR sensor was developed. Owing to the different sensing mechanisms,
these two types of sensors offer complementary information regarding the analytes. Such system was
successfully applied for the detection of CFC-113 and HCFC-141b. Because of the nonpolar property
of CFC-113, it shows negligible current response on the Si-NR FET sensor, while the Si-NR FET sensor
is very sensitive to the polar HCFC-141b. Thus, these two types of freon molecules can be effectively
distinguished with the gas sensor. In addition, the polymer-coated FBAR sensor shows excellent
Sensors 2018,18, 343 10 of 12
quantitative detection capacity of the freon molecules with a ppm-level LOD. Mixtures of CFC-113
and HCFC-141b have been successfully analyzed with the dual-mode gas sensor. Once the component
is discriminated by the Si-NR FET sensor, concentration information can be measured by the FBAR
sensor. In conclusion, the sensing system can discriminate CFC-113 and HCFC-141b and calculate
the concentration of each component by simple math without using a complex mathematic model.
Therefore, it provides a much more effective method to distinguish the two components with different
polarities than a sensing system based on one type of sensor, e.g., a conventional FBAR-based electronic
nose. Though freon is used as a test compound to demonstrate the polarity differential, the sensing
performance of the system can be further improved by arraying multiple Si-NR FETs as well as FBARs
and coating the sensors with multiple chemicals (e.g., molecularly imprinted polymers [
52
]). Owing to
the small size and CMOS-compatibility of the two types of sensors, the dual-mode sensing system is a
potential candidate as a portable integrated analysis system for the analysis of gas mixtures.
Acknowledgments:
The authors gratefully acknowledge financial support from the Natural Science Foundation
of China (NSFC No. 61674114), Tianjin Applied Basic Research and Advanced Technology (14JCYBJC41500),
and the 111 Project (B07014). The authors thank Luye Mu and Mark A. Reed from Yale University for the help
with the Si-NR FETs and helpful discussions.
Author Contributions:
X.D. and Y.C. conceived and supervised the project. Z.H. and X.W. designed and
performed the experiments. H.Q. and W.P. fabricated the FBAR device. Z.H. and Y.C. wrote the manuscript. X.D.
and H.Q. commented on the manuscript. All authors discussed the results.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Arshak, K.; Moore, E.; Lyons, G.M.; Harris, J.; Clifford, S. A review of gas sensors employed in electronic
nose applications. Sens. Rev. 2004,24, 181–198. [CrossRef]
2.
Muñoz, R.; Sivret, E.C.; Parcsi, G.; Lebrero, R.; Wang, X.; Suffet, I.H.; Stuetz, R.M. Monitoring techniques for
odour abatement assessment. Water Res. 2010,44, 5129–5149. [CrossRef] [PubMed]
3.
Fanget, S.; Hentz, S.; Puget, P.; Arcamone, J.; Matheron, M.; Colinet, E.; Andreucci, P.; Duraffourg, L.;
Myers, E.; Roukes, M.L. Gas sensors based on gravimetric detection—A review. Sens. Actuators B Chem.
2011,160, 804–821. [CrossRef]
4.
Stetter, J.R.; Penrose, W.R.; Yao, S. Sensors, chemical sensors, electrochemical sensors, and ecs. J. Electrochem. Soc.
2003,150, S11–S16. [CrossRef]
5.
Gebicki, J. Application of electrochemical sensors and sensor matrixes for measurement of odorous chemical
compounds. TrAC Trends Anal. Chem. 2016,77, 1–13. [CrossRef]
6.
Rakow, N.A.; Suslick, K.S. A colorimetric sensor array for odour visualization. Nature
2000
,406, 710–713.
[CrossRef] [PubMed]
7.
Pereira, J.; Porto-Figueira, P.; Cavaco, C.; Taunk, K.; Rapole, S.; Dhakne, R.; Nagarajaram, H.; Câmara, J.S.
Breath analysis as a potential and non-invasive frontier in disease diagnosis: An overview. Metabolites
2015
,
5, 3–55. [CrossRef] [PubMed]
8.
Guentner, A.T.; Koren, V.; Chikkadi, K.; Righettoni, M.; Pratsinis, S.E. E-nose sensing of low-ppb
formaldehyde in gas mixtures at high relative humidity for breath screening of lung cancer? ACS Sens.
2016
,
1, 528–535. [CrossRef]
9.
Cynkar, W.; Dambergs, R.; Smith, P.; Cozzolino, D. Classification of tempranillo wines according
to geographic origin: Combination of mass spectrometry based electronic nose and chemometrics.
Anal. Chim. Acta 2010,660, 227–231. [CrossRef] [PubMed]
10.
Sankaran, S.; Panigrahi, S.; Mallik, S. Olfactory receptor based piezoelectric biosensors for detection of
alcohols related to food safety applications. Sens. Actuators B Chem. 2011,155, 8–18. [CrossRef]
11.
Compagnone, D.; Fusella, G.C.; Carlo, D.M.; Pittia, P.; Martinelli, E.; Tortora, L.; Paolesse, R.; Natale, D.C.
Gold nanoparticles-peptide based gas sensor arrays for the detection of foodaromas. Biosens. Bioelectron.
2013,42, 618–625. [CrossRef] [PubMed]
12.
Hansen, T.; Petersen, M.A.; Byrne, D.V. Sensory based quality control utilising an electronic nose and GC-MS
analyses to predict end-product quality from raw materials. Meat Sci.
2005
,69, 621–634. [CrossRef] [PubMed]
Sensors 2018,18, 343 11 of 12
13.
Varghese, O.K.; Gong, D.; Dreschel, W.R.; Ong, K.G.; Grimes, C.A. Ammonia detection using nanoporous
alumina resistive and surface acoustic wave sensors. Sens. Actuators B Chem. 2003,94, 27–35. [CrossRef]
14.
Bachar, N.; Liberman, L.; Muallem, F.; Feng, X.; Mullen, K.; Haick, H. Sensor arrays based on polycyclic
aromatic hydrocarbons: Chemiresistors versus quartz-crystal microbalance. ACS Appl. Mater. Interfaces
2013
,
5, 11641–11653. [CrossRef] [PubMed]
15.
Chen, Y.; Zhang, H.; Feng, Z.; Zhang, H.; Zhang, R.; Yu, Y.; Tao, J.; Zhao, H.; Guo, W.; Pang, W.; et al.
Chemiresistive and gravimetric dual-mode gas sensor toward target recognition and differentiation.
ACS Appl. Mater. Interfaces 2016,8, 21742–21749. [CrossRef] [PubMed]
16.
Paska, Y.; Stelzner, T.; Christiansen, S.; Haick, H. Enhanced sensing of nonpolar volatile organic compounds
by silicon nanowire field effect transistors. ACS Nano 2011,5, 5620–5626. [CrossRef] [PubMed]
17.
Wang, B.; Haick, H. Effect of functional groups on the sensing properties of silicon nanowires toward volatile
compounds. ACS Appl. Mater. Interfaces 2013,5, 2289–2299. [CrossRef] [PubMed]
18.
Wang, H.; Sun, Z.; Lu, Q.; Zeng, F.; Su, D. One-Pot Synthesis of (Au Nanorod)-(Metal Sulfide) Core-Shell
Nanostructures with Enhanced Gas-Sensing Property. Small 2012,8, 1167–1172. [CrossRef] [PubMed]
19.
Majhi, S.; Rai, P.; Raj, S.; Chon, B.-S.; Park, K.-K.; Yu, Y.-T. Effect of Au Nanorods on Potential Barrier
Modulation in Morphologically Controlled Au@Cu
2
O Core-Shell Nanoreactors for Gas Sensor Applications.
ACS Appl. Mater. Interfaces 2014,6, 7491–7497. [CrossRef] [PubMed]
20.
Chen, R.J.; Franklin, N.R.; Kong, J.; Cao, J.; Tombler, T.W.; Zhang, Y.; Dai, H. Molecular photodesorption
from single-walled carbon nanotubes. Appl. Phys. Lett. 2001,79, 2258–2260. [CrossRef]
21.
Nakhleh, M.K.; Amal, H.; Jeries, R.; Broza, Y.Y.; Aboud, M.; Gharra, A.; Ivgi, H.; Khatib, S.; Badarneh, S.;
Har-Shai, L.; et al. Diagnosis and classification of 17 diseases from 1404 subjects via pattern analysis of
exhaled molecules. ACS Nano 2017,11, 112–125. [CrossRef] [PubMed]
22.
Yu, J.; Hong, M.; Wang, L.; Huang, K.; Guo, Y. Synthesis and gas-sensing properties of P-type Na
0.44
MnO
2
nanoribbons. Mater. Lett. 2016,164, 440–443. [CrossRef]
23.
Clavaguera, S.; Carella, A.; Caillier, L.; Celle, C.; Pécaut, J.; Lenfant, S.; Vuillaume, D.; Simonato, J.P.
Sub-ppm detection of nerve agents using chemically functionalized silicon nanoribbon field-effect transistors.
Angew. Chem. Int. Ed. 2010,49, 4063–4066. [CrossRef] [PubMed]
24.
Ippolito, S.J.; Trinchi, A.; Powell, D.A.; Wlodarski, W. Acoustic wave gas and vapor sensors. In Solid State
Gas Sensing; Springer: Berlin, Germany, 2009; pp. 1–44.
25.
Benetti, M.; Cannatà, D.; Pietrantonio, D.F.; Foglietti, V.; Verona, E. Microbalance chemical sensor based on
thin-film bulk acoustic wave resonators. Appl. Phys. Lett. 2005,87, 173504. [CrossRef]
26.
Liu, W.; Wang, J.; Yu, Y.; Chang, Y.; Tang, N.; Qu, H.; Wang, Y.; Pang, W.; Zhang, H.; Zhang, D.; et al.
Tuning the resonant frequency of resonators using molecular surface self-assembly approach. ACS Appl.
Mater. Interfaces 2015,7, 950–958. [CrossRef] [PubMed]
27.
Lu, Y.; Chang, Y.; Tang, N.; Qu, H.; Liu, J.; Pang, W.; Zhang, H.; Zhang, D.; Duan, X. Detection of volatile
organic compounds using microfabricated resonator array functionalized with supramolecular monolayers.
ACS Appl. Mater. Interfaces 2015,7, 17893–17903. [CrossRef] [PubMed]
28.
Chang, Y.; Tang, N.; Qu, H.; Liu, J.; Zhang, D.; Zhang, H.; Pang, W.; Duan, X. Detection of volatile organic
compounds by self-assembled monolayer coated sensor array with concentration-independent fingerprints.
Sci. Rep. 2016,6, 23970. [CrossRef] [PubMed]
29.
Liu, W.; Qu, H.; Hu, J.; Pang, W.; Zhang, H.; Duan, X. A highly sensitive humidity sensor based on
ultrahigh-frequency microelectromechanical resonator coated with nano-assembled polyelectrolyte thin
films. Micromachines 2017,8, 116. [CrossRef]
30.
Lin, R.-C.; Chen, Y.-C.; Chang, W.-T.; Cheng, C.-C.; Kao, K.-S. Highly sensitive mass sensor using film bulk
acoustic resonator. Sens. Actuators A Phys. 2008,147, 425–429. [CrossRef]
31.
Nirschl, M.; Schreiter, M.; Vörös, J. Comparison of fbar and qcm-d sensitivity dependence on adlayer
thickness and viscosity. Sens. Actuators A Phys. 2011,165, 415–421. [CrossRef]
32.
Wallington, T.J.; Schneider, W.F.; Worsnop, D.R.; Nielsen, O.J.; Sehested, J.; Debruyn, W.J.; Shorter, J.A.
The environmental impact of CFC replacements—HFCs and HCFCs. Environ. Sci. Technol.
1994
,28,
320A–326A. [PubMed]
33.
Tsai, W.-T. A review of environmental hazards and adsorption recovery of cleaning solvent
hydrochlorofluorocarbons (HCFCs). J. Loss Prev. Process Indust. 2002,15, 147–157. [CrossRef]
Sensors 2018,18, 343 12 of 12
34.
Mangani, F.; Maione, M.; Lattanzi, L.; Arduini, J. A gas chromatographic-mass spectrometric method for trace
analysis of chlorofluorocarbons and their replacement compounds in atmospheric samples. Chromatographia
2000,51, 325–330. [CrossRef]
35.
Mangani, F.; Maione, M.; Lattanzi, L.; Arduini, J. Atmospheric measurements of the halogenated
hydrocarbons involved in global change phenomena. Atmos. Environ. 2000,34, 5303–5309. [CrossRef]
36.
Brown, A.T.; Chipperfield, M.P.; Boone, C.; Wilson, C.; Walker, K.A.; Bernath, P.F. Trends in atmospheric
halogen containing gases since 2004. J. Quant. Spectrosc. Radiat. Transf. 2011,112, 2552–2566. [CrossRef]
37.
Bris, K.; McDowell, J.; Strong, K. Measurements of the infrared absorption cross-sections of HCFC-141b
(CH3CFCL2). J. Quant. Spectrosc. Radiat. Transf. 2012,113, 1913–1919. [CrossRef]
38.
Belton, P.S.; Saffa, A.M.; Wilson, R.H.; Ince, A.D. The potential of fourier transform infrared spectroscopy for
the analysis of confectionery products. Food Chem. 1988,28, 53–61. [CrossRef]
39.
Castillo, S.; Mattila, I.; Miettinen, J.; Orešiˇc, M.; Hyötyläinen, T. Data analysis tool for comprehensive
two-dimensional gas chromatography/time-of-flight mass spectrometry. Anal. Chem.
2011
,83, 3058–3067.
[CrossRef] [PubMed]
40.
Ul-Haq, Z.; Ali, M.; Batool, S.; Tariq, S.; Qayyum, Z. Emission quantification of refrigerant CFCs, HCFCs
and HFCs in megacity Lahore (Pakistan) and contributed ODPs and GWPs. J. Earth Syst. Sci.
2016
,125,
1273–1284. [CrossRef]
41.
Abbasi, S.A.; Abbasi, T.; Abbasi, S.A.; Abbasi, T. Ozone Hole; Springer-Verlag NY Inc.: New York, NY, USA,
2017; pp. 37–50.
42.
Nguyen, H.; El-Safty, S.A. Meso- and macroporous Co
3
O
4
nanorods for effective voc gas sensors. J. Phys.
Chem. C 2011,115, 8466–8474. [CrossRef]
43.
Duan, X.; Li, Y.; Rajan, N.K.; Routenberg, D.A.; Modis, Y.; Reed, M.A. Quantification of the affinities and
kinetics of protein interactions using silicon nanowire biosensors. Nat. Nanotechnol.
2012
,7, 401–407.
[CrossRef] [PubMed]
44.
Sauerbrey, G. Use of vibrating quartz for thin film weighing and microweighing. Z. Phys.
1959
,155, 206–222.
[CrossRef]
45.
Pang, W.; Zhao, H.; Kim, E.; Zhang, H.; Yu, H.; Hu, X. Piezoelectric microelectromechanical resonant sensors
for chemical and biological detection. Lab Chip 2011,12, 29–44. [CrossRef] [PubMed]
46.
Ghaffari, S.; Ng, E.; Ahn, C.; Yang, Y.; Wang, S.; Hong, V.A.; Kenny, T.W. Accurate modeling of quality factor
behavior of complex silicon mems resonators. J. Microelectromech. Syst. 2015,24, 276–288. [CrossRef]
47.
Hwang, M.-J.; Shim, W.; Ryu, D.-W.; Moon, H. Low-pressure adsorption isotherms of aromatic compounds
on polyisobutylene gel measured on a quartz crystal microbalance. J. Chem. Eng. Data
2012
,57, 701–707.
[CrossRef]
48.
Chen, Y.-J.; Zhu, C.-L.; Xiao, G. Ethanol sensing characteristics of ambient temperature sonochemically
synthesized zno nanotubes. Sens. Actuators B Chem. 2008,129, 639–642. [CrossRef]
49.
Schmidt, M.W.; Baldridge, K.K.; Boatz, J.A.; Elbert, S.T.; Gordon, M.S.; Jensen, J.H.; Koseki, S.; Matsunaga, N.;
Nguyen, K.A.; Su, S. General atomic and molecular electronic structure system. J. Comput. Chem.
1993
,14,
1347–1363. [CrossRef]
50.
Brunauer, S.; Emmett, P.H.; Teller, E. Adsorption of gases in multimolecular layers. J. Am. Chem. Soc.
1938
,
60, 309–319. [CrossRef]
51.
Seri-Levy, A.; Avnir, D. The brunauer-emmett-teller equation and the effects of lateral interactions.
A simulation study. Langmuir 1993,9, 2523–2529. [CrossRef]
52.
Iqbal, N.; Mustafa, G.; Rehman, A.; Biedermann, A.; Najafi, B.; Lieberzeit, P.A.; Dickert, F.L. QCM-arrays for
sensing terpenes in fresh and dried herbs via bio-mimetic mip layers. Sensors
2010
,10, 6361–6376. [CrossRef]
[PubMed]
©
2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
