Pulse-Driven Micro Gas Sensor Fitted with ... - ACS Publications

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Pulse-Driven Micro Gas Sensor Fitted with Clustered Pd/SnO2 Nanoparticles

Koichi Suematsu*a1, Yuka Shinb, Nan Mab, Tokiharu Oyamab, Miyuki Sasakib, Masayoshi Yuasaa, Tetsuya Kida*c, and Kengo Shimanoea

a

Department of Energy and Material Sciences, Faculty of Engineering Science, Kyushu University, Kasuga, Fukuoka, 816-8580, Japan

b

Department of Molecular and Material Science, Interdisciplinary Graduate School of Engineering Science, Kyushu University, Kasuga, Fukuoka, 816-8580, Japan c

Department of Applied Chemistry and Biochemistry, Kumamoto University, Kumamoto, 860-8555, Japan

Corresponding Author Koichi Suematsu E-mail: [email protected], Fax: +81-92-925-7724 Tetsuya Kida E-mail: [email protected], Fax: +81-96-342-3679

1

Present address: Chemical and Texture Research Institute, Fukuoka Industrial Technology Center, Chikushino, Fukuoka, 818-8540, Japan

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 25

Abstract Real-time monitoring of specific gas concentrations with a compact and portable gas sensing device is required to sense potential health risk and danger from toxic gases. For such purposes, we developed an ultrasmall gas sensor device, where a micro sensing film was deposited on a micro heater integrated with electrodes fabricated by the microelectromechanical system (MEMS) technology. The developed device was operated in a pulse-heating mode to significantly reduce the heater power consumption and make the device battery-driven and portable. Using clustered Pd/SnO2 nanoparticles, we succeeded in introducing mesopores ranging from 10 to 30 nm in the micro gas sensing film (area: φ 150 µm) to detect large volatile organic compounds (VOCs). The micro sensor showed quick, stable, and high sensor responses to toluene at ppm (parts per million) concentrations at 300oC even by operating the micro heater in a pulse-heating mode where switch-on and -off cycles were repeated at one-second intervals. The high performance of the micro sensor should result from the creation of efficient diffusion paths decorated with Pd sensitizers by using the clustered Pd/SnO2 nanoparticles. Hence we demonstrate that our pulse-driven micro sensor using nanostructured oxide materials holds a promise as a battery-operable, portable gas sensing device.


2

Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION One-line detection of toxic gases in the atmosphere is highly important to avoid health risk resulting from the exposure to such gases. There are also strong demands to prevent air pollution from hazardous gases by real-time monitoring of changes in the air quality. Small and compact gas sensors are suitable for continuously monitoring the air quality and detecting a sudden change in gas concentration. Oxide semiconductor-based gas sensor, developed in 1962 1, has been extensively used for such purposes. This type of sensors is called as resistive-type gas sensors, which sense specific gases and their concentrations based on a change in the device electrical resistance. For resistive-type gas sensors, SnO2 has been recognized as the best material because of its excellent sensitivity to detect combustible gases 2-4. The recent advance in resistive-type gas sensors has been made possible with tremendous efforts in the materials development for gas sensing. To date, we have revealed that controls of surface state

5-7

, crystallite size

8,9

, and pore size

9-11

in sensing films are the keys to improving the

sensor performance. Surface state control by noble metal loading on the SnO2 surface gives rise to high sensitivity by an increase in the surface reaction activity. Crystal size control is effective in inducing a large change in the electrical resistance upon crystal interaction with gases. In particular, tuning the pore structure of gas sensing films is very critical to improve the selectivity. We proved that controlling the pore size is an efficient way to selectively detect large gas molecules such as volatile organic compounds (VOCs) 11-13. Since prolonged exposure to VOCs brings serious damages to health, the development of compact VOC sensors is important. However, the major drawback of oxide-based resistive-type gas sensors is their rather high energy consumption required to heat the device because the reversible oxygen adsorption and reactions occur only at high temperatures around 300oC. Thus, it is difficult to make this type of sensors battery-operable and portable when using a conventional printed heater chip with high power consumption.

However,

current

advances

in

microfabrication

technologies

for

microelectromechanical systems (MEMS) have enabled the fabrication of heaters in micrometer


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 25

scale. Using such minimalized heaters, it is possible to significantly reduce the power consumption needed to operate the heater and thus to induce the sensing reaction. So far, several types of gas sensors using a micro heater have been reported to detect various gases such as carbon monoxide 14-16

, ammonia

17

, hydrogen sulfide

18

, methanol

19

, ethanol

20,21

and benzene

22

. However, the

sensitivity of these micro sensors is typically low. The probable reason is the lack of materials design in fabricating micro gas sensors. Usually, downsizing the gas sensing layer makes it difficult to control the layer microstructure. Thus, the sensor design principles are critically important to develop high-performance micro sensors. In this study, we used nanostructured materials to form a gas sensing layer on micro heaters. A method that coats micro droplets containing nanostructured materials on micro heaters was employed to fabricate micro sensing films. Stable nanomaterial suspensions where nanostructured materials are homogeneously dispersed in a solvent allow for the controlled deposition of micro sensing films. Previously, we reported that Pd-loaded SnO2 clustered particles of ca. 45 nm exhibited significantly high sensitivities to VOC gases such as toluene 13. The high performance resulted from the creation of mesopores in the sensing films by controlled clustering (aggregation) of SnO2 nanoparticles, which assisted diffusion of large VOC gas molecules deep inside the porous films. Pd nanoparticles deposited on the clustered particles also contributed to the performance enhancement by boosting the reaction of surface adsorbed oxygen with VOC gases. Here, a suspension containing the clustered nanoparticles was used to introduce efficient gas diffusion paths in micro sensing films and thus to detect VOC gases. As mentioned above, the use of micro heaters drastically reduces the power consumption. For example, when using a micro heater of 100 µm2, the power consumption is reported to be approximately 26 and 15 mW for continuous heating at 450 and 300oC, respectively, which values are less than one tenth of those for conventional heaters 23. In this study, to further reduce the power consumption, we attempted to operate micro heaters in a pulse-heating mode. Repeating switch-on and -off cycles on a one-second time scale would reduce the power consumption while keeping the


4

Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


heater temperature constant. The reduction in the power consumption allows us to develop battery-driven portable gas sensors, which would open up new application fields of gas sensors.

EXPERIMENTAL SECTION Pd (0.2 mol%)-loaded SnO2 nanoparticle clusters were synthesized by a hydrothermal treatment method, as previously reported 13. Stannic acid gel was first prepared by adding a SnCl4· 5H2O aqueous solution (1M) into a NH4HCO3 solution (1M) in a drop-wise manner. A Pd[(NO2)2(NH4)2] solution was added into the obtained gel after Cl- was removed from the gel by centrifugation. Then, a suspension containing stannic acid gel and the Pd complex was hydrothermally treated at 200oC for 3 h at 10 MPa with continuous stirring at 600 rpm, producing a clear solution containing monodispersed SnO2 nanoparticles of ca. 4 nm when the solution pH was 10.6. Clustered SnO2 nanoparticles of ca. 45 nm were synthesized by decreasing the solution pH from 10.6 to 9.6 before hydrothermal treatment. The crystallite structure of the nanoparticles was analyzed by X-ray diffraction using Cu Kα radiation (XRD; Empyrean, PANalytical) and the average crystallite size was estimated by Sherrer’s equation using XRD patterns. The colloidal particle size of the obtained nanoparticles was analyzed by dynamic light scattering (DLS) analysis using a DLS spectrophotometer (LPA-3000/3100, Otsuka Electronics). The obtained nanoparticles were observed on transmission electron microscopy (TEM; TECNAI G2-F20, PHILIPS), and energy dispersive spectroscopy (EDS) analyses were carried out to confirm the presence of Pd in the nanoparticles. For TEM analyses, a suspension containing the nanoparticles was dropped onto a carbon support film on a Cu grid. Figure 1 shows the structure of the micro sensor device integrating a micro heater and Pt electrodes. The device had a hole at the center of the substrate. This unique structure allows for the precise control of the heater temperature by rapid heating and cooling due to efficient heat dissipation through the hole, which act as a heat spreader. To fabricate micro sensing films by micro


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 25

droplet injection, the clustered SnO2 nanoparticles were mixed with glycerin to form a coating paste (40 wt.%). Micro droplets of the paste were deposited on the micro heater through a capillary tube using a micro injector (Eppendorf FemroJet). Glycerin was removed by operating the micro heater at 550oC for 30 min in synthetic air. After the heat treatment, the sensing film became a doughnut-like shape (area: φ 150 µm); the film was also deposited on the back side of the substrate through the hole, as shown in Figure 1 a. The back side deposition of the film can strengthen the mechanical stability of the film by the tight contact between the film and the substrate. The surface state was observed on field emission scanning electron microscopy (FE-SEM; S-4800, HITACHI). A micro sensor fitted with bare SnO2 nanoparticles was also fabricated and tested for its gas sensing properties for comparison. The gas sensing properties of the micro sensor were measured in a conventional gas flow apparatus, as shown in figure 1 (b). The fabricated micro sensor chip was mounted on a holder and installed in the chamber of the gas flow apparatus. The device temperature was adjusted by applying appropriate voltage to the micro heater in a pulsed-heating mode, where switch-on and -off cycles were repeated at one-second intervals, as shown in Figure 1 (c). Figure 1 (d) shows the relationship between the applied heater voltage and device temperature. The device can be heated up to more than 500 oC even by applying a small voltage near 1.5 V. The linear relationship in Figure 1 (d) indicates the excellent performance of the micro heater. Sensor electrical resistances were monitored every 0.05 seconds during heater operation. As target gases, combustible gases such as hydrogen and toluene in synthetic air were used and their concentrations were controlled by mixing commercial parent gases with synthetic air. The sensor device was connected by a standard resistor in series and the voltage across the standard resistor was measured under an applied DC voltage of 4 V to evaluate the electrical resistance of the device. An electrical signal was acquired from the sensor using an electrometer (2701; Keithley Instruments). The sensor response (S) was defined as the ratio of resistance in air (Ra) to that in air containing the combustible gases (Rg) (S = Ra/Rg).


6

Page 7 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSION Microstructure analysis The XRD patterns of bare SnO2 nanoparticles and clustered Pd/SnO2 nanoparticles were shown in Figure 2 (a). The XRD patterns were matched with that of cassiterite SnO2 (JCPDS 41-1445), and no impurity phases were observed. The average crystallite sizes for the both nanoparticles were approximately 5 nm. Particle size distributions of the two nanoparticles in water were shown in supporting information (Figure S1), indicating that the average colloidal sizes of the clustered Pd/SnO2 nanoparticles and the bare SnO2 nanoparticles were 46 and 5.5 nm respectively. The difference between the colloidal and crystallite sizes of the clustered Pd/SnO2 nanoparticles is a clear indication of the clustering (aggregation) of nanoparticles, as discussed in the previous report.13 After calcination at 550oC, the crystallite size slightly increased to 8 and 9 nm for the SnO2 nanoparticles and the clustered Pd/SnO2 nanoparticles, respectively, showing their good thermal stability against particle growth. TEM analyses were also performed on the clustered Pd/SnO2 nanoparticles calcined at 550oC, as shown in Figure 2 (b). The particle sizes were approximately 10 nm, which was in good agreement with that estimated from XRD peaks. The presence of Pd on the particles was confirmed by EDS analyses. Pd was detected in five spots designated in the corresponding TEM image of the clustered Pd/SnO2 nanoparticles, as shown in Figure 2 (c) and (d), indicating a good dispersion of Pd on the SnO2 nanoparticles. Figures 3 (a) and (b) show optical microscope images of the micro sensors fitted with the bare SnO2 nanoparticles and the Pd-loaded clustered SnO2 nanoparticles, respectively. The heater and electrodes were entirely covered with a sensing film of a doughnut-like shape. SEM images of the films composed of the bare nanoparticles and the Pd-loaded clustered nanoparticles are shown in figures 3 (c) and (d), respectively. For the SnO2 nanoparticles, the particles were closely packed to form a rather dense film. In contrast, for the clustered SnO2 nanoparticles, the presence of large mesopores around 10-30 nm was clearly evident. This observation is in good agreement with a pore size distribution of a sensing film fabricated by spin coating using the clustered SnO2 nanoparticles,


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 25

as shown in Figure S2, where the peak pore size was seen at approximately 20 nm. The obtained results indicate the successful control of the microstructure in the small sensing film by using the nanostructured materials. The estimated thickness of the sensing film was approximately 10-20 µm from a cross-sectional image of the sensing film shown in Figure 3 (e).

Electrical resistance measurements Figures 4 (a-e) and 5 (a-e) show the electric resistances in synthetic air, 200 ppm hydrogen, and 50 ppm toluene at 300-500oC for the micro sensors using the bare SnO2 nanoparticles and the clustered Pd/SnO2 nanoparticles, respectively, as a function of time (cycle number). The figures only show the resistances at the first five (1-5) and the last five (196-200) cycles for the clarity. The electrical resistances did not change before and after the heater was switched off for one second. This trend was seen at all temperatures, suggesting that the pulse-heating operation successfully kept the heater temperature constant. On the other hand, the resistance in air slightly increased with time at initial cycles. The kinetics of adsorption and desorption of oxygen on SnO2 dictates how fast the electrical resistance in air reaches a steady value. Slower oxygen adsorption on the SnO2 surface likely caused the electrical resistance to gradually change during initial cycles. Nevertheless, the resistance reached steady values within 100 seconds (50 cycles). Therefore, once the sensor signal is stabilized after 100 seconds for the startup, the sensor is ready for use in continuous gas monitoring. The resistances at every cycle for the micro sensors using the bare SnO2 nanoparticles and the clustered Pd/SnO2 nanoparticles, respectively, are also shown in supporting information (Figures S3 and S4). For the clarity, the heater-off region is omitted in these figures. The resistance values continuously changed, but were stabilized after approximately 100 seconds without any large fluctuation, showing stable responses to oxygen, hydrogen, and toluene at 350-500oC in a pulse-heating mode. Steady state values of the resistance in air for the bare SnO2 nanoparticles and the clustered


8

Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Pd/SnO2 nanoparticles at 300oC were in the range of 104 and 105 Ω, respectively , which values were about one order lower than that for conventional thick film sensors 24. In general, films composed of nanostructured materials exhibit high electrical resistances due to their porous structures resulting from the small sizes and anisotropic morphologies of the constituent materials. It is sometimes difficult to measure high electrical resistances with a simple electrical measurement system that is suitable for portable devices. Thus, the lower electrical resistance of our sensor is advantageous for device miniaturization. The small gap size between the measuring electrodes (3 µm) should be the reason for the lower electrical resistances of the micro sensing film despite its porous structure. Similarly, the resistance in hydrogen at 300 and 350oC also showed a gradual decrease in the electrical resistance, as shown in Figures 4 (a, b) and 5 (a, b). Previously, we analyzed the response and recovery speeds of SnO2-based gas sensors using a high speed gas switching system and suggested that the desorption of water formed by the surface combustion of hydrogen was rather slow

25

. The observed gradual decrease in the electrical resistance means the slow combustion of

hydrogen during initial cycles at lower temperatures. Thus, it is possible that remaining adsorbed water impeded the reaction of hydrogen with adsorbed oxygen. As a result, it took a relatively longer time for the electrical resistance to reach a steady state where the hydrogen combustion and oxygen re-adsorption are in steady state. After some cycles, the whole sensing film was warmed up enough to promote water desorption and thus the hydrogen combustion smoothly occurred without interference of adsorbed water. On the other hand, the resistance at higher than 500oC showed a rapid decrease at initial cycles and reached a steady state, as shown in Figures 4 (e) and 5 (e). This should be due to rapid desorption of water at high temperature. In contrast, the resistance in toluene showed steady values without significant changes from the first cycle. Ideally, the combustion of toluene on the SnO2 surface produces water and carbon dioxide. However, the complete oxidation of toluene is probably difficult and partly oxidized by-products may be produced. As compared with the high affinity of water on the SnO2 surface 26, it is possible that such by-products do not strongly bind to the SnO2 surface. We also reported that the


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

desorption of carbon dioxide from the SnO2 surface is faster than that of water

Page 10 of 25

27

. In addition, the

amounts of combustion products should be smaller because the combustion rate of toluene is slower than that of hydrogen. Thus, in contrast to the case for hydrogen, the reaction of toluene with adsorbed oxygen was not largely influenced by adsorbed species on the SnO2 surface, leading to the constant output of the electrical resistance, as measured. The lower concentration of toluene (50 ppm) than that of hydrogen (200 ppm) also accounts for the less effect of the combustion products on the electrical resistance.

Sensor response analysis Response transients of the micro sensor to hydrogen and toluene at 300-500oC are shown in Figures 6 (a-e) and 7 (a-e), respectively. The figures only show the sensor responses at the first five (1-5) and the last five (196-200) cycles for the clarity. The responses at every cycle at 300-500oC are also shown in supporting information (Figures S5 and S6). The sensor responses of the clustered Pd/SnO2 nanoparticles are compared with those of the bare SnO2 nanoparticles. The sensor response to hydrogen showed a gradual increase with time at initial cycles at lower temperatures. The same behaviors were observed for the two devices using the clustered Pd/SnO2 nanoparticles and the bare SnO2 nanoparticles. The gradual increase during first few cycles should be due to the two effects discussed above; slow oxygen adsorption and slow water desorption lead to gradual changes in the electrical resistances in air and hydrogen, respectively, until the device was warmed up. However, such a slow response to hydrogen disappeared at 500oC, as shown in Figure S5 (e), probably because of enhanced oxygen adsorption and water desorption. Pd-loaded clustered nanoparticles exhibited higher sensor responses to hydrogen than the bare nanoparticles. It is well accepted that Pd loading can enhance the sensor response to combustible gases by chemical sensitization and electrical sensitization effects 28. The result clearly indicates the active role of dispersed Pd nanoparticles as an efficient sensitizer even in the micro sensing film. The sensor responses to toluene were stable from the first cycle although a slight decrease


10

Page 11 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was observed at the beginning of the switch-on cycle. This stable response behavior is due to efficient surface reactions of toluene on the SnO2 surface without severe interference from combustion products. Notably, the sensor response of the Pd-loaded clustered nanoparticles to toluene was about twenty times higher than that of the bare nanoparticles at 300oC. According to our previous report

13

, the mesopore structure (10-30 nm) decorated with Pd nanoparticles should

promote the efficient diffusion and reactions of toluene deep inside the film, resulting in the higher sensor response of the Pd/SnO2 clustered nanoparticles. Hence we successfully demonstrate that our material design is also applicable to micro gas sensing films. The operating temperature dependences of the sensor responses to hydrogen for the devices using the clustered Pd/SnO2 nanoparticles and the SnO2 nanoparticles are shown in Figure 8 (a). For calculation of the sensor responses, steady state values of electrical resistances were used. All the temperatures, the Pd-loaded clustered nanoparticles showed larger sensor responses to hydrogen, indicating the efficient activation of the sensing film by Pd loading. The highest response was obtained at 450oC in both cases, showing a volcano-type temperature dependence, which can theoretically be interpreted in terms of kinetics of combustion reaction and gas diffusion, as follows 29

. Increasing the operating temperature accelerates combustion reactions and gas penetration into a

sensing film, which results in an increase in the sensor response. However, when further increasing the operating temperature, the combustion reactions preferentially should occur on the surface of a sensing film due to thermally activated catalytic action of constituent oxide particles, which impedes gas penetration deep inside the sensing layer. As a result, the sensor response decreases with a further increasing temperature. At an optimum temperature, rates of the combustion reaction and gas diffusion are well balanced to give the best sensor response. These tendencies have often been seen for resistive-type sensors 9,13. Figure 8 (b) shows the temperature dependence of the sensor responses to toluene. All the temperatures, the Pd-loaded clustered nanoparticles showed larger sensor responses to toluene than the bare nanoparticles. The pore size distribution and SEM measurements revealed that the sensing


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

films had mesopores ranging from 10 to 30 nm, where gases diffuse on the basis of the Knudsen diffusion mechanism. In such a case, increasing the pore size can increase the Knudsen diffusion coefficient of toluene. Thus, the introduction of larger mesopores in the sensing film by controlled clustering of nanoparticles is responsible for the improved sensor response to toluene. For the clustered nanoparticles, the highest sensor response was obtained at 350oC, which was much lower than that for the bare nanoparticles that showed the maximum sensor response at 450oC. However, a drop in the sensor response to toluene was observed at 400oC. The reason for this is unclear yet. Nevertheless, the observed lowering of the maximum response temperature can be explained by increased rates of combustion reactions and gas diffusion, which were caused by the Pd loading and the increase in pore size, respectively. Previously, we reported that the clustered Pd/SnO2 nanoparticles showed larger sensor responses to toluene than CO and H2

13

. The reason for this was clearly explained in terms of

efficient toluene diffusion in the porous films fabricated by the clustered nanoparticles. We also reported that a porous sensing film fabricated by TiO2 nanotubes showed enhanced sensor responses to toluene and ethanol than H2

20

. The importance of porosity control in gas sensing films has also

been reported in our paper where porous films fabricated by larger particles were used to detect larger molecules such as H2S 9. Thus, according to our previous findings, we believe that the present micro sensing film made by the clustered SnO2 nanoparticles should also exhibit good sensor responses and selectivity to other VOCs such as ethanol and formaldehyde. The micro sensor using the clustered Pd/SnO2 nanoparticles responded well to toluene (50 ppm), reaching S = 25 at 350oC. However, the obtained values are much less than those obtained for conventional thin film-type gas sensors (S = 3,000 at 300oC) 13. Gas sensing films are assembly of fine constituent particles that show a change in the electrical resistance in response to combustible gases. The output of the device sensor response is obtained by the integration of sensor response for each particle. Accordingly, for micro devices with a small sensing area, the sensor response is intrinsically lower because of the smaller number of constituent particles. Although this effect has


12

Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


not yet been experimentally verified, further materials development and optimization of the film microstructure would overcome such a drawback by enhancing the utility factor of the sensing film, thereby improving the performance of micro gas sensors.

CONCLUSIONS We have developed a pulse-driven micro sensor fitted with clustered Pd/SnO2 nanoparticles to detect VOCs gases. Few droplets of a nanoparticle suspension were deposited on a micro heater integrated with micro electrodes to form a micro sensing film (area: φ 150 µm). SEM analysis revealed that the micro sensing film clearly had mesopores ranging from 10 to 30 nm, which provides efficient diffusion paths for large VOC molecules. The micro sensor responded well to hydrogen and toluene even though the heater was driven in a pulse-heating mode. In particular, the sensor quickly responded to toluene within 0.1 seconds when the heater was switched on, suggesting its combustion reaction and diffusion efficiently occur in the micro film with a controlled microstructure. The highest sensor response to toluene was obtained at 350oC; the maximum temperature was effectively shifted from 450 to 350oC by loading Pd on the constituent particles. The obtained results suggest that our pulse-driven micro sensor holds a promise as a battery-operable, portable gas sensor. Using the miniaturized sensor device equipped with the ultrasmall heater, we were able to use a relatively small volume chamber (115 cm3) where the device was installed. Accordingly, the reduction in the volume of the measurement system allowed for rapid replacement of the gas atmosphere in the chamber at a very short time, which enabled the analysis of kinetics of surface reactions on the SnO2 surface on the one-second time scale, as revealed in this study. We believe that our micro gas sensor would also be applicable to the analysis of the reaction kinetics on other heterogeneous gas/solid systems.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 25

Acknowledgments This work was supported by a Grant-in-Aid for Scientific Research (B) (No. 22350064) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and the Nippon Sheet Glass Foundation for Materials Science and Engineering. We also thank Figaro Engineering Inc. for providing the micro heaters integrated with electrodes.

Supporting Information Available DLS analysis and pore size distribution of clustered Pd/SnO2 nanoparticles; time dependent electrical resistance and sensor response to hydrogen and toluene for the micro sensors. This material is available free of charge via the Internet at http://pubs.acs.org.


14

Page 15 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1)

Seiyama, T.; Kato, A.; Fujiishi, K.; Nagatani, A. Anal. Chem. 1962, 34, 1502-1503.

(2)

Yamazoe, N. Sens. Actuator B 2005, 108, 2-14.

(3)

McAleer, J.F.; Moseley, P.T.; Norris, J.O. W.; Williams, D.E. J. Chem. Soc., Faraday Trans. 1 1987, 83, 1323-1346.

(4)

Tanaka, K.; Morimoto, T.; Sonoda, S.; Matsuura, S.; Moriya, K.; Egashira, M. Sens. Actuators B 1991, 3, 247-253.

(5)

Yuasa, M.; Masaki, T.; Kida, T.; Shimanoe, K.; Yamazoe, N. Sens. Actuator B 2009, 136, 99-104.

(6)

Yuasa, M.; Kida, T.; Shimanoe, K. ACS Appl. Mater. Interfaces 2012, 4, 4231-4236.

(7)

Hua, Z.; Yuasa, M.; Kida T.; Yamazoe, N.; Shimanoe, K. Thin Solid Films 2013, 548, 677-682.

(8)

Xu, C.; Tamaki, J.; Miura, N.; Yamazoe, N. Sens. Actuators B 1991, 3, 147-155.

(9)

Kida, T.; Fujiyama, S.; Suematsu, K.; Yuasa, M.; Shimanoe, K. J. Phys. Chem. C 2014, 117, 17574-17582.

(10)

Kida, T.; Nishiyama, A.; Hua, Z.; Suematsu, K.; Yuasa, M.; Shimanoe, K. Langmuir 2014,

30, 2571-2579. (11)

Seo, M.-H.; Yuasa, M.; Kida, T.; Huh, J.-S.; Yamazoe, N.; Shimanoe, K. Sens. Actuator B

2011, 154, 251-256. (12)

Seo, M.-H.; Yuasa, M.; Kida, T.; Kanmura, Y.; Huh, J.-S.; Yamazoe, N.; Shimanoe, K. J.

Ceram. Soc. Jpn. 2011, 119, 884-889. (13)

Suematsu, K.; Shin, Y.; Hua, Z.; Yoshida, K.; Yuasa, M.; Kida, T.; Shimanoe, K. ACS Appl.

Mater. Interfaces 2014, 6, 5319-5326. (14)

Graf, M.; Barrettino, D.; Taschini, S.; Hagleitner, C.; Hierlemann, A.; Baltes, H. Anal. Chem.

2004, 76, 4437-4445.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 25

(15)

Lee, C.-Y.; Chang, C.-C.; Lo. Y.-M. Sensors 2010, 10, 10701-10713.

(16)

Triantafyllopoulou, R.; Tsamis, C. Phys. Stat. Sol. 2008, 11, 2643-2646.

(17)

Meier, D.C.; Semancik, S.; Button, B.; Strelcov, E.; Kolmakov, A. Appl. Phys. Lett. 2007, 91,

063118. (18)

Gong, J.; Chen, Q.; Lian, M.-R.; Liu, N.-C.; Stevenson, R.G.; Adami, F. Sens. Actuator B

2006, 114, 32-39. (19)

Martines, C.J.; Hockey, B.; Montgomery, C.B.; Semancik, S. Langmuir 2005, 21,

7937-7944. (20)

Kida, T.; Seo, M.-H.; Suematsu, K.; Yuasa, M.; Kanmura, Y.; Shimanoe, K. Appl. Phys.

Express 2013, 6, 047201. (21)

Wadn, Q.; Li, H.; Chen, J.; Wang, T.H.; He, X.L.; Li, J.P. Appl. Phys. Lett. 2004, 84, 3654.

(22)

Ghaddab, B.; Berger, F.; Sanchez, J.B.; Menini, P.; Mavon, C.; Yoboue, P.; Potin, V. Sens.

Actuator B 2011, 152, 68-72. (23)

Yoshioka, K.; Tanihira, T.; Shinnishi, K.; Kaneyasu, K. Chem. Sens. 2007, 23, suppl. B

16-18 [in Japanese]. (24)

Ma, N.; Suematsu, K.; Yuasa, M.; Kida, T.; Shimanoe, K. ACS Appl. Mater. Interfaces 2015,

7, 5863-5869. (25)

Kida, T.; Kuroiwa, T.; Yuasa, M.; Shimanoe K.; Yamazoe, N. Sens. Actuator B 2008, 134,

928-933. (26)

Yamazoe, N.; Fuchigami, J.; Kishikawa, M.; Seiyama, T. Surf. Sci. 1979, 86, 335-344.

(27)

Tamaki, J.; Nagaishi, M.; Teraoka Y.; Miura, N.; Yamazoe, N. Surf. Sci. 1989, 221, 183-196.

(28)

Yamazoe, N. Sens Actuator B 1991, 5, 7-19.

(29)

Yamazoe, N.; Sakai, G.; Shimanoe, K. Catal. Surv. Asia 2003, 7, 63-75.


16

Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1

(a) Schematic images of the micro sensor device before and after sensing film deposition

and (b) a gas flow apparatus with an electrical measurement system. (c) An operation profile of the micro heater in a pulse-heating mode. (d) Relationship between the heater temperature and the applied voltage.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2

Page 18 of 25

(a) XRD patterns of the bare SnO2 nanoparticles and the Pd/SnO2 nanoparticles with or

without calcination. TEM images of (b) the clustered Pd/SnO2 nanoparticles calcined at 550oC for 30 min. (c) EDS spectra obtained at the five spots designated in (d) the corresponding TEM image of the Pd/SnO2 clustered nanoparticles deposited on a carbon support film on a Cu grid.


18

Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3

Optical micrographs of the micro sensor devices fitted with (a) the SnO2 nanoparticles and

(b) the clustered Pd/SnO2 nanoparticles. Surface SEM images of the gas sensing films made of (c) the SnO2 nanoparticles and (d) the clustered Pd/SnO2 nanoparticles. (e) Cross-sectional SEM image of the sensing film made of the clustered Pd/SnO2 nanoparticles.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4

Page 20 of 25

Time dependence of electrical resistances in (●) synthetic air, (●) 200 ppm hydrogen, and

(●) 50 ppm toluene at (a) 300oC (applied heater voltage: 0.86 V), (b) 350oC (0.99 V), (c) 400oC (1.12 V), (d) 450oC (1.24 V), and (e) 500oC (1.37 V) for the device fitted with the bare SnO2 nanoparticles. The device was operated in a pulse-heating mode. The figure only shows the resistances at the first five (1-5) and the last five (196-200) cycles for the clarity.


20

Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5

Time dependence of electrical resistances in (●) synthetic air, (●) 200 ppm hydrogen, and

(●) 50 ppm toluene at (a) 300, (b) 350, (c) 400, (d) 450, and (e) 500oC for the device fitted with the clustered Pd/SnO2 nanoparticles. The figure only shows the resistances at the first five (1-5) and the last five (196-200) cycles for the clarity.


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6

Page 22 of 25

Response transients to 200 ppm hydrogen at (a) 300, (b) 350, (c) 400, (d) 450, and (e)

o

500 C for the devices fitted with (●) the clustered Pd/SnO2 nanoparticles and (●) the bare SnO2 nanoparticles. The devices were operated in a pulse-heating mode. The figures only show the sensor responses at the first five (1-5) and the last five (196-200) cycles for the clarity.


22

Page 23 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7

Response transients to 50 ppm toluene at (a) 300, (b) 350, (c) 400, (d) 450, and (e) 500oC

for the devices fitted with (●) the clustered Pd/SnO2 nanoparticles and (●) the bare SnO2 nanoparticles. The figures only show the sensor responses at the first five (1-5) and the last five (196-200) cycles for the clarity.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 8

Page 24 of 25

Temperature dependence of sensor responses to (a) 200 ppm hydrogen and (b) 50 ppm

toluene for the devices fitted with (▲) the clustered Pd/SnO2 nanoparticles and (●) the bare SnO2 nanoparticles.


24

Page 25 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


for TOC only


25

Pulse-Driven Micro Gas Sensor Fitted with ... - ACS Publications

Recommend Documents