- Email: [email protected]
ZnO-nanowire size effect induced ultra-high sensing response to ppb-level H2S
Sensors and Actuators B 240 (2017) 264–272
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
ZnO-nanowire size effect induced ultra-high sensing response to ppb-level H2 S Ying Chen a , Pengcheng Xu a , Tao Xu b , Dan Zheng b , Xinxin Li a,∗ a State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China b School of Chemical and Environmental Engineering, Shanghai Institute of Technology, Shanghai 201418, China
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 10 August 2016 Accepted 19 August 2016 Available online 21 August 2016 Keywords: Nano-size sensing effect ZnO nanowires High sensitivity H2 S sensor ppb-level gas detection
a b s t r a c t Presented is nano-size effect of ZnO nanowires (NWs) on sensitivity of hydrogen sulfide (H2 S) detection in ambient atmosphere. The sensing mechanism of ZnO NWs for H2 S detection is found coming from nano-size effect that governs both the sulfuration reaction when H2 S is introduced and the following desulfuration reaction when the gas is removed from air. Based on the nano-material property of smaller dimension featuring higher chemical-activity at surface, the sensitivity of ZnO NWs to ultra-low concentration H2 S is expected to be increased by shrinking the NW diameter. In our experiment, suspended micro hotplates for chemiresistive sensing are fabricated to investigate the nano-size sensing effect. Besides conventional ZnO NWs with the diameter as 50 nm, a novel tree-branched nano-structure of ZnO NWs (with the branch diameter as 20 nm) is in situ grown onto the sensing-area of the micro hotplate. In order to investigate the sensing effect, both the 20 nm NW sensor and the 50 nm NW sensor are used to experimentally detect H2 S with concentrations in the range of ppb-ppm. The 20 nm ZnO-NWs sensor realizes resoluble sensing to 5 ppb H2 S, while the 50 nm sensor can only response to H2 S with concentration higher than 50 ppb. The well validated nano-size effect of thinner NWs featuring higher sensing response can be extended for various chemical sensing nano-materials. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Hydrogen sulfide (H2 S) is a kind of toxic, corrosive and inflammable gas. Gaseous H2 S has distinct fetor of rotten egg that is commonly formed in nature and released during the decay of organic matter. The H2 S gas is normally generated from sewage, coal mines, and petroleum/natural gas industries, and it is widely utilized in various chemical industries. When the gas concentration of H2 S is higher than 250 ppm, it may lead to death [1]. Even at lower concentration level, H2 S is harmful to human bodies. The harmful H2 S molecules can be rapidly absorbed by human’s lung and easily cause respiratory system disease, unconsciousness neurological sequelae and cardiovascular-related death [2]. Recently recognized as an important endogenous gas, H2 S can be implicated in diverse physiological and pathophysiological process, such as regulation of inflammation, brain development, blood pressure regulation and metabolism [3–9]. It has been reported that the hydrogen sulfide in human body is at ppm or sub-ppm
∗ Corresponding author. E-mail address: [email protected] (X. Li). http://dx.doi.org/10.1016/j.snb.2016.08.120 0925-4005/© 2016 Elsevier B.V. All rights reserved.
level, and it is indicated that trace-concentration detection of H2 S is helpful for disease diagnosis [10]. For accurate analysis and earlystage prognosis of diseases, H2 S detection at sub-ppm or lower concentration is highly demanded. Numerous chemical sensors have been investigated and developed, which are mainly based on metal-oxide semiconductor sensing [11,12], electrochemical detection [13] and optical measurement [14]. However, the lack of effective detection method and analytical tools for monitoring of trace-concentration H2 S has hindered the biomedical research that is correlative with the endogenous gas [15]. Therefore, it is of great importance to develop ppb-ppm level detectable H2 S sensors. Several technical approaches have been reported to improve sensing performance of H2 S sensors. At present, the H2 S sensors by using semiconducting metal-oxide nano-materials are attractive due to high sensitivity. For example, ZnO, In2 O3 and SnO2 [16–20] can specifically response to the presence of the gas by changing electronic conductivity, thereby having been widely used in various applications like industrial safety, product quality control, clinical diagnostics and home-safety alarms. Among the various investigated metal-oxide sensing materials, n-type ZnO nanowires (NWs) materials have been intensively explored and
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
widely used for gas sensing, due to their high mobility of conductive electrons, good chemical/thermal stability and low cost [21–24]. In recent years, ZnO nanostructures have been used for building chemiresistive H2 S sensors [25] and the measure range of these H2 S sensing devices generally focus in ppm level [26]. Thanks to the high surface-to-volume ratio, the nano-structured materials provide large adsorbing interface that indeed helps to increase sensitivity [27]. For instance, mesoporous thin-films or nanowire arrays have shown several folds enhanced H2 S sensing properties compared to the traditional sensors that are based on solid-state thin-film sensing materials [11,28]. However, to the best of our knowledge, it is really difficult to improve sensitivity by tens of folds or more by only using the increasing surface-to-volume method. Being an exception, our group recently reported that, when the diameter of ZnO wires is reduced to tens of nanometers, a sulfuration-desulfuration dual-stage sensing mechanism becomes to dominate and induces dramatically enhanced sensitivity. 50nm-diameter ZnO NWs can detect 50 ppb H2 S due to the dual-step reaction [29]. This new sensing mechanism can be originated from nano-size effect. When ZnO NWs are thin enough, the surface becomes active enough to be sulfurated by H2 S to form ZnS. This decreases the number of adsorbed oxygen molecule from ambience and causes thinner electron depletion layer. The sulfuration leads to drastic decrease of the ZnO resistance and enhance the sensing signal. After H2 S is removed, the metastable intermediate of ZnS will be quickly desulfurized back to ZnO by the ambient oxygen and the resistance quickly recovers to its initial state. This two-step reaction sensing mechanism is different from the widely used conventional one for detecting reductive gases in ambient air. The mechanism for conventional metal-oxide sensors is based on semiconductor resistance change which comes from electron depletion layer variation induced by surface adsorbed oxygen species [30]. When H2 S molecules react with the oxygen species, the change of depletion layer causes variation of ZnO resistance and outputs the sensing signal [17,31]. It has been reported that bulk-material of ZnO responds to H2 S with the concentration at tens of ppm level [11] while the 50 nm ZnO NWs sensor features reversible sulfuration-desulfuration sensing mechanism and outputs response to 50 ppb H2 S. Obviously, nano-size effect brings the high sensing performance. Along with size shrinkage of the nanomaterial, the confinement of bulk to surface becomes weaker and the material surface becomes more active. Compared with the sensitivity improvement caused by the increased surface-to-volume ratio, the nano-material size effect induced improvement in sensitivity is much more significant [29]. This work is just inspired by the sulfuration-desulfuration H2 S sensing mechanism for ZnO of nano-diameter. Herein, it is expected that continuously thinning the ZnO NWs induces continuously enhanced active surface and improved sensitivity. Therefore, herein we conceive to explore the ZnO nano-size effect on H2 S detection performance. For the purpose, the method for in situ grown thinner ZnO NWs than 50 nm is studied. Finally, we develop a technique to region-selectively grown tree-branched 20 nm ZnO NWs at the comb-shaped electrode region of a chemiresistive sensing micro-chip. The sensing microchip consists of a suspended micro-hotplate where the pair of comb electrodes is accommodated for the sensing nano-material growth. Finite element modeling (FEM) thermal simulation is performed to optimize the sensor structure and improve temperature uniformity across the micro-hotplate. For H2 S sensing experiment, 20 nm ZnO NWs and the conventional 50–60 nm ZnO NWs are grown, respectively. The detectable results of 5 ppb H2 S by using the 20 nm ZnO NWs well proves the regulation effect of ZnO nano-size on H2 S sensitivity.
265
2. Experimental 2.1. Design and fabrication of suspended sensing micro-hotplate Micromachined hotplates for gas sensors are advantageous in thermal isolation, low power consumption and low-cost manufacturing [32–34]. In this work, a sandwich-structured micro-hotplate comprises comb-finger electrode layer, intermediate SiO2 isolating layer and heating resistor layer. The three-layer structure is formed on a suspended silicon nitride beam-plate. Fig. 1 shows our designed micro-hotplate for gas sensor. The suspended plate is connected with four narrow supporting beams that are clamped to silicon substrate and the geometry is optimized to minimize thermal conduction to outside. The circular plate is 300 m in diameter and each support beam is 150 m in length and 20 m in width. The radial platinum resistor is for heating the plate, as platinum features very linear temperature coefficient of resistance (TCR) for easy temperature calibration. The resistance versus plate temperature is pre-calibrated in a precise oven. Above the Pt heater, a thin layer of SiO2 is employed to insulate the heater and the electrodes at top. To measure the chemiresistance change of ZnO NWs, a pair of comb-finger electrodes is densely placed above. The gap distance between the two comb-fingers is 10 m, which can be electrically bridged by the grown branching ZnO NWs into a chemiresistor. As is shown in Fig. 1(b), the ZnO NWs will in situ grown at the central sensing area, i.e., over the comb-finger electrodes. The sensor has four interconnection lines, with two for the micro-heater and the other two for chemiresistance signal readout. By using silicon micromachining technology, the chip is fabricated in single-side polished (100) silicon wafer, with the process steps sketched in Fig. 2 and described as follows: (1) Deposition 1 m-thick low-stress silicon nitride by low pressure chemical vapor deposition (LPCVD). (2) A 30 nm/300 nm Ti/Pt composite thin-film is deposited by electron-beam evaporation and patterned by lift-off process to form the heater. (3) A 400nm-thick insulating layer of SiO2 is formed using plasma enhanced (PE) CVD. The contact holes for the Pt heater are opened. Then, a 30 nm/300 nm-thick Cr/Au layer is evaporated and patterned to form the comb-finger electrodes and the wire-bonding pads of the Pt heater. (4) The SiO2 and silicon nitride layers are sequentially etched by reactive ion etching to form the shape of the suspended plate and the beams. (5) The silicon beneath the plate is etched off by isotropic etching of vapor-phase XeF2 and the micro plate is released for free-standing. After the wafer is diced, the chips are mounted on PCB and wire bonding is implemented. At this step, the chip is ready for in situ growth of ZnO NWs sensing material. It is worth pointing out that the metal materials used in the micro sensor can sustain the temperature process during synthesize of ZnO NWs. 2.2. In situ grown of ZnO NWs The ZnO NWs, with the diameter as about 50–70 nm, are synthesized from aqueous solution and in situ grown at the sensing area of the sensor by using a modified hydrothermal method [35] and shown in Fig. 3(1)–(2). The surface of the sensor is plasma pretreated (with Harrick PDC-002 equipment) for 5 min to eliminate surface contamination. Afterwards, 5 mM zinc acetate in ethanol is dropwise coated in the sensing area by using a commercial micromanipulator (Eppendorf, PatchMan NP2), with the aid of inspection under a microscope (Leica, DM4000). Later, the sensor chip is dried in air for 10 s, rinsed with ethanol and then, blow dried with a pure nitrogen stream. This coating step needs to be repeated for at least three times. Covered with the seed film of zinc acetate crystallites, the seeded chip is heated to 350 ◦ C in air for 20 min to yield a nanolayer of ZnO seeds. Then, ZnO NWs are in-situ grown at the surface of the chip by using hydrothermal method with an aqueous solu-
266
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 1. (a) Designed micro-hotplate, which consists of comb-finger electrodes, insulating layer, Pt heater and suspended plate. (b) Schematic of the suspended micro-hotplate.
Fig. 2. Fabrication steps of the micro-hotplate in 3D and cross-sectional views. (1) Deposition silicon nitride on silicon wafer. (2) Pt heater is formed by e-beam evaporation and lift-off process. (3) SiO2 insulating layer is deposited by PE-CVD. Cr/Au layer is evaporated and patterned to form the comb-finger electrodes and the wire-bonding pads. (4) Suspended plate and support beams are shaped by RIE. (5) The Si beneath the plate is etched off to suspend the hotplate.
tion that contains zinc nitrate (25 mM), hexamethylenetetramine (HMTA, 25 mM) and polyethylenimine (PEI, 6 mM). The bottomup hydrothermally growth process for the NWs is carried out at 90 ◦ C for 4 h. Finally, the sensing chip with the grown ZnO NWs of 50–70 nm in diameter is taken out from the solution, washed with deionized water and dried for sensing. For in-situ synthesis of the branched ZnO-NWs in the sensing area, a two-step growth process is developed and shown in Fig. 3(1)–(4). The first step is for growing the trunks to certain length and the second step is for growing the thinner branches from the thicker tree trunks. The first-step trunk growth is similar with the process for the 50 nm ZnO NWs. After the required length of the trucks is reached, the sensor chip is heated to 350 ◦ C in air for 20 min to remove the residual HMTA and PEI. Since the ZnO trunks are vertically grown on the substrate, it is difficult to form a uniform seed-layer on the top surface of the trunks by spin- or dip-coating. In that case, the seed-layer would prefer to aggregate near the bot-
tom part of the trunks that would result in generation of undesired grass-like ZnO NWs instead of the branched nano-structure. To address this issue, highly conformal atom layer deposition (ALD) is herein employed to deposit the seed-layer [36–39]. The branch growing steps are detailed as follows. Using Zn(C2 H5 )2 and H2 O as co-precursors, 60 cycles seed-layer is constructed at 150 ◦ C by using ALD. Lift-off process is applied to remove the ZnO seed layer at undesired regions. Then, identical hydrothermal-synthesis process is repeated to grown isotropically stretched ZnO nano-branches from each trunk. Finally, the chip is taken out from the solution, washed with deionized water and dried for the further sensing applications.
2.3. Characterization Scanning electron microscopy (SEM) characterization is performed in a Hitachi S-4800 cold-field equipment for observing the
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
267
Fig. 3. Schematic illustration of the synthesis process of the ZnO NWs and nano-branched ZnO NWs.
nanostructure of ZnO NWs and the morphology of micro-hotplate. Optical microscope images are taken under a Leica DM 4000 microscope. The TCR of the Pt heater is calibrated in a temperature programmable oven. The whole sensor chip is introduced in the oven for temperature ramping experiment from 25 ◦ C to 150 ◦ C, with increment of 25 ◦ C. 2.4. Sensing experiment H2 S gas of various concentrations is generated by a lab-made gas generator which mixes the standard concentration H2 S gas and the ambient air. The standard low-concentration H2 S gas of 1000 ppm is purchased from Shenkai Gas Corporation, Shanghai, China. The pure air is connected from our lab air compressing system where the relative-humidity range is well controlled. The H2 S gas sensing experiment is carried out in a testing chamber of 20 L in volume. The sensor is sealed in the testing chamber that is connected with the gas generator for sensing experiment. A DC power source, Agilent34410A multi-meter and a personal computer are used to supply the micro heater and real-time record the chemiresistive sensing signal. 3. Results and discussion 3.1. Micro-hotplate for gas sensor Fig. 4(a) shows the top-view of the fabricated hotplate chip. As the insulating SiO2 is transparent, it is quite clear that the electrodes are located above the heater. SEM image in Fig. 4(b) shows the plate suspended with the four supporting beams. The thermal mismatch stress introduced by the composite layers of different materials is optimally designed and, as a result, the suspended hotplate is quite flat and shows good mechanical robustness in our experiment. By using the finite-element analysis tool of COMSOL, thermal analysis is implemented for the hotplate. Only conductive and convective energy losses have been taken into account since the irradiative energy losses are very small due to the not very high working temperature and the small area of the heated plate. The simulation results for the temperature distribution on top surface of the hotplate are depicted in Fig. 5(a) and (b). It can be seen that, under power supply of DC 1.6 V, the temperature of the hotplate surface is 573 K (or 300 ◦ C) at the center and the
temperature difference across the diameter of the plate is less than 5 ◦ C. Better temperature uniformity is achieved in the sensing chemiresistive sensing area. At the clamping end of the supporting beams, the temperature steeply drops to room temperature. As is shown in Fig. 5(c), the experimentally calibrated TCR value for the Pt heating resistor is 1.7 × 10−3 /◦ C and the there is a linear relationship between resistance and temperature. After calibration, a linear relationship between temperature and heating voltage is obtained that is shown in Fig. 5(d). The calibrated relationship is T = 0.939 + 157.267 V, where T is the resistance-representative temperature of the hotplate and V is the heating voltage. 3.2. In-situ growth of branched ZnO NWs As shown in the SEM images of Fig. 6(a), conventional ZnO NWs with the diameter of 50–70 nm is locally grown in the sensing area of the micro-hotplate. The ZnO NWs can be grown on top of both the gold electrode fingers and the SiO2 finger intervals. The magnified SEM image in Fig. 6(b) and the cross-sectional image in Fig. 6(c) shows that, the ZnO NWs are densely grown at the electrodes region and the adjacent NWs cross-link with each other to form the chemiresistor. Fig. 6(c) and (d) shows that the nanowires are generally 2–3 m in length and 50–67 nm in diameter. For locally growing the branched ZnO NWs, the situation of the trunks formed at the first step is similar to that of the aforementioned conventional ZnO NWs. After the second-step growth of the branches, the ZnO nano-structure formed in the sensing area of the hotplate is shown in Fig. 6(e). High-density ZnO branches are successfully grown on the previously formed ZnO trunks. As is shown in Fig. 6(f), the diameter of the nano-branches is generally 20 nm, with the deviation being less than ±5 nm. Attribute to the inter-linked dense branches, stable resistances value of the sensitive chemiresistor can be measured. It is worthy pointing out that, the 20 nm-diameter ZnO trunks previously formed at the first-step become slightly thicker during the repeated hydrothermal growth at the second-step. 3.3. Nano-size effect on sensing performance By adjusting DC voltage, the Pt resistive micro-heater can provide desired working temperature for the H2 S sensing material. In the sensing experiment, the sensing response is represented by
268
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 4. (a) Optical microscope image of the micro-hotplate integrated with a heater and a pair of comb-finger electrodes. (b) SEM image of the fabricated micro-hotplate that is suspended with a wide gap from silicon substrate.
Fig. 5. (a) Simulated temperature distribution under DC 1.6 V heating. (b) Temperature distribution across the hotplate diameter. (c) Experimentally measured Pt resistance shows linear relationship with temperature of the hotplate. The temperature coefficient of resistance (TCR) is calibrated as 0.0017/◦ C. (d) Experimentally calibrated linear relationship of between hotplate temperature and heating voltage.
the ratio of the resistance change (R) to the initial resistance in clean air (R). For the branched ZnO NWs sensor, its response to 1 ppm H2 S is measured and shown in Fig. 7, where stepped temperatures of 150–450 ◦ C is achieved by heating the hotplate (with increment of 50 ◦ C). In the relatively lower temperature range from 150 ◦ C to 300 ◦ C, the response steeply increases with the temperature increasing and, exhibits the maximum response signal at 300 ◦ C. When the temperature is further increased to over 300 ◦ C, the response decreases quickly and, the response at 350 ◦ C becomes much lower than that at 200 ◦ C. The “volcano-shape” temperature dependent behavior of the sensing response is in accordance with the theory as reported in literature [40] and the experimental phenomenon can be explained as follows. For H2 S detection by using ZnO nanowires, the activity of ZnO NWs increases along with the working temperature raising. The increased activity of ZnO NWs induces the enhanced sensing response by generating ZnS intermediate via sulfuration reaction [29]. When the sensing response
reaches the maximum value at 300 ◦ C, adsorption and desorption of H2 S molecules get to an equilibrium state. Afterwards, as the temperature increases, the adsorption of H2 S gas decreases, meanwhile, the desorption of H2 S from ZnO is enhanced. If the number of H2 S molecules involved in the above-mentioned reaction decrease (i.e. desorption of H2 S increases), it must generate less ZnS intermediate. On the other hand, as a kind of metastable intermediate, ZnS is quite unstable at high temperature because it tends to be oxidized in ambient air to form ZnO. Hence, the quantity of ZnS becomes even less at higher temperature (>300 ◦ C). At higher working temperature, the two factors mentioned-above brought less ZnS and lead to a drastic decline in the sensing response of the sensor. Based on the experimental results and the above-mentioned discussions, 300 ◦ C is the optimal working-temperature for achieving highest response of the branched ZnO NWs sensor. For comparison, the sensor with the 20 nm branch-structured ZnO and the sensor with the 50–70 nm conventional ZnO NWs are
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
269
Fig. 6. (a) Top-view SEM shows the locally grown ZnO NWs in the sensing area of the hotplate sensor. (b) Magnified SEM image shows the densely grown ZnO NWs on the hotplate. (c) Side-view of the dense ZnO NWs. (d) Further close-up view shows that the diameter of the NWs is ranged from 50 to 70 nm. (e) After the second step for growing the branches, the SEM image shows the tree-branched nano-structure, and the adjacent branches are cross-linked. (f) The diameter of the ZnO branches is generally measured as 20 nm (±5 nm).
Fig. 7. Temperature dependent sensing response to 1 ppm H2 S gas of the same branched ZnO sensor.
both introduced to detect ppb-level H2 S. At identical temperature of 300 ◦ C under air atmosphere, the resistance value of 50–70 nm ZnO sensors is 60.7 k and the resistance value of 20 nm branched ZnO sensor is 47.3 k. Obviously, the resistance value of the 50 nm ZnO sensor is higher than the value of 20 nm ZnO sensor. According to the SEM characterization results as shown in Fig. 6, there are more dense nanowires grown in the 20 nm ZnO sensor and a branched nanostructure is formed. Hence, there are more electron conduction channels formed in the 20 nm nanowires sensor. Therefore, compared with 50–70 nm ZnO NWs, the 20 nm branched ZnO sensor exhibits a smaller chemiresistance value. During the experiment, both the 20 nm branched sensor and the 50–70 nm ZnO sensor are sequentially exposed to H2 S gas, where the gas concentration is varied from 5 ppb to 200 ppb. As is shown in Fig. 8(a), the 20 nm branched ZnO sensor exhibits recoverable responses at 300 ◦ C. For detecting H2 S of 5 ppb, the measured sensing response (R/R) is about 4%, while the noise-floor is less than 1%. To 200 ppb H2 S, the response signal is as high as 44%. In addition, it can be seen that the sensor response increases along with increasing H2 S concentration. The sensing response and H2 S concentration have a linear relationship that is plotted in Fig. 8(b). At ultra-low gas
270
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 8. (a) Sensing experimental results of the 20 nm branched ZnO sensor to trace-level H2 S gas at 300 ◦ C, with concentrations in the range of 5 ppb to 200 ppb. (b) A linear fitting line is obtained between the sensor response and H2 S concentration, which can be expressed as y = 0.002x + 0.036, herein x represents the H2 S concentration and y is the sensing response. (c) Magnified sensing curve of 20 nm sensor to 50 ppb H2 S and the fitted response curve. (d) Response to 50 ppb–200 ppb H2 S for the sensor with the conventional 50 nm ZnO NWs at 300 ◦ C.
Fig. 9. For the 20 nm branched ZnO sensor, the selectivity to twelve kinds of interfering gases is measured.
pressure (i.e. gas concentration), the experimentally obtained linear relationship can be explained based on Langmuir sorption theory [41]. Based on the results in Fig. 8(a), the limit of detection (LOD) of the 20 nm branched ZnO sensor is better than 3 ppb, which is estimated by assessing the signal noise-floor and assuming a linear relationship between the response and concentration. During the sensing experiment, H2 S gas is switched off when the output-signal almost gets stable. The time for H2 S gas-off has been optimized based on the multiple experimental confirmations and the obtained sensing results only show a negligible influence for the following discussion. Magnifying the time axis of the sensing curve, it can be clearly observed that the signal increases drastically at the initial stage but increases very slowly to achieve the final equilibrium state thereafter. Taking the measuring cycle of 20 nm sensor to 50 ppb H2 S as example, the sensing curve can be well fitted by using Langmuir theory [see Fig. 8(c)]. According to the fitted result, it will take about 19 min to reach the completely equilibrium state, which is 4.4 times longer than the practical measuring
time (4.3 min), i.e. the measuring time can be saved by more than 75%. On the other hand, the sensing response increment during the slowly increase period is quite small and can be even ignored. Based on the fitted response curve, the fitted sensing response (S) is 0.153 and the practical sensing response (S ) in this study is obtained as 0.142. Thus, it can be calculated that S /S = 92.8%. The abovementioned discussion confirms the experiment error of the sensing response is considered acceptable but the detection time can be reduced by 75%. Above all, it is reasonable to switch off H2 S gas at an opportune time during each detection cycle. For comparison, the H2 S response curves of the 50–70 nm conventional ZnO sensor are experimentally obtained at 300 ◦ C and shown in Fig. 8(d). The sensor cannot output resoluble response until H2 S concentration is increased to 50 ppb. The sensing response to 50 ppb H2 S is merely 3%. Being resoluble to 5 ppb trace-concentration H2 S, the sensor with the 20 nm branched ZnO is much more sensitive than the sensor with the conventional 50–70 nm ZnO NWs. This result can well verify the nano-size effect on the H2 S sulfuratation/desulfuration
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
sensing mechanism. Along with size shrinkage of the sensing nanomaterial, its surface should become more and more active. The high response achieved with the 20 nm branched ZnO nanowires can be mainly attributed to the nano-size induced surface effect. In addition, The ZnO nano-branches provide many nano-junctions at the branch roots located on the trunks, which possibly have minor help effect to enhance the sensitivity [42]. The reason why we grow the branched nanowires instead of conventional 20 nm nanowires to examine the size effect lie in that the very thin nanowires tend to prostrate on the surface of the sensor. The sparse prostrate-grown ZnO nanowires easily lead to instable chemiresistance. For the high-sensitivity sensor with the 20 nm branched ZnO, the selectivity is experimentally assessed at 300 ◦ C, where we select twelve kinds of interfering gases/vapors of methanol, ethanol, acetone, hexane, methanol, ethyl acetate, benzene, toluene, NH3 , CH4 , H2 and CO2 . The experimental results are compared together in Fig. 9, with the concentration values of the interfering gases denoted. According to the results, obviously, the 20 nm treebranched ZnO-nanowire H2 S sensor has a satisfactory selectivity to the interfering gases. 4. Conclusion This work experimentally reveals nano-size effect induced high sensitivity and fine resolution of ZnO-nanowires to ppb-level H2 S gas. The sulfuration/desulfuration sensing mechanism of nanoscale ZnO sensing chemiresistors is different from the widely used conventional one, and it indicates that higher sensitivity of ZnO NWs to ultra-low concentration H2 S can be achieved by size shrinking of the sensing nano-material. Locally grown on a suspended micro hotplate, a novel structure of 20 nm tree-branched ZnO NWs is used as chemiresistive sensing material to experimentally detect ppb-ppm level H2 S. In order to reveal the nano-size sensing effect, conventional 50–70 nm ZnO NWs are also prepared and examined for comparison. The results verify the nano-size sensing effect. The tree-branched 20 nm ZnO NWs sensor exhibits resoluble detection limit of 5 ppb H2 S and much higher response than the 50–70 nm ZnO NW sensor. Besides that the very sensitive H2 S sensor is achieved, the nano-size effect induced high performance sensing mechanism is promising to be applied to more chemical sensors. Conflict of interests The authors declare no conflict of interests. Acknowledgements This research is supported by MOST of China (2016YFA0200800), NSF of China (61401446, 6152781861604163, 91323304, 61321492), NSF of Shanghai (15ZR1447300), Science and Technology Commission of Shanghai (14521106100).P.C.X appreciates the financial support of the Youth Innovation Promotion Association CAS (2016213). References [1] M.D. Shirsat, M.A. Bangar, M.A. Deshusses, N.V. Myung, A. Mulchandani, Polyaniline nanowires-gold nanoparticles hybrid network based chemiresistive hydrogen sulfide sensor, Appl. Phys. Lett. 94 (2009) 083502. [2] D.C. Glass, A Review of the health-effects of hydrogen-sulfide exposure, Ann. Occup. Hyg. 34 (1990) 323–327. [3] M. Whiteman, S. Le Trionnaire, M. Chopra, B. Fox, J. Whatmore, Emerging role of hydrogen sulfide in health and disease: critical appraisal of biomarkers and pharmacological tools, Clin. Sci. 121 (2011) 459–488. [4] S.J. Choi, B.H. Jang, S.J. Lee, B.K. Min, A. Rothschild, I.D. Kim, Selective detection of acetone and hydrogen sulfide for the diagnosis of diabetes and halitosis using SnO2 nanofibers functionalized with reduced graphene oxide nanosheets, ACS. Appl. Mater. Interfaces 6 (2014) 2588–2597.
271
[5] M.F. Rosario-Alomar, T. Quinones-Ruiz, D. Kurouski, V. Sereda, E.B. Ferreira, L. De Jesus-Kim, S. Hernandez-Rivera, D.V. Zagorevski, J. Lopez-Garriga, I.K. Lednev, Hydrogen sulfide inhibits amyloid formation, J. Phys. Chem. B 119 (2015) 1265–1274. [6] Y. Zhou, W. Chen, J. Zhu, W. Pei, C. Wang, L. Huang, et al., Nanoprobes: inorganic–organic hybrid nanoprobe for nir-excited imaging of hydrogen sulfide in cell cultures and inflammation in a mouse model, Small 10 (2014) 4802. [7] N.H. Kim, S.J. Choi, D.J. Yang, J. Bae, J. Park, I.D. Kim, Highly sensitive and selective hydrogen sulfide and toluene sensors using Pd functionalized WO3 nanofibers for potential diagnosis of halitosis and lung cancer, Sens. Actuators B 193 (2014) 574–581. [8] C. Szabo, Hydrogen sulphide and its therapeutic potential, Nat. Rev. Drug Discov. 6 (2007) 917–935. [9] A. Papapetropoulos, A. Pyriochou, Z. Altaany, G. Yang, A. Marazioti, Z. Zhou, M.G. Jeschke, L.K. Branski, D.N. Herndon, R. Wang, C. Szabo, Hydrogen sulfide is an endogenous stimulator of angiogenesis, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 21972–21977. [10] M. Whiteman, P.K. Moore, Hydrogen sulfide and the vasculature: a novel vasculoprotective entity and regulator of nitric oxide bioavailability? J. Cell Mol. Med. 13 (2009) 488–507. [11] P.S. Shewale, G.L. Agawane, S.W. Shin, A.V. Moholkar, J.Y. Lee, J.H. Kim, M.D. Uplane, Thickness dependent H2 S sensing properties of nanocrystalline ZnO thin films derived by advanced spray pyrolysis, Sens. Actuators B 177 (2013) 695–702. [12] Y. Zhang, Q. Xiang, J.Q. Xu, P.C. Xu, Q.Y. Pan, F. Li, Self-assemblies of Pd nanoparticles on the surfaces of single crystal ZnO nanowires for chemical sensors with enhanced performances, J. Mater. Chem. 19 (2009) 4701–4706. [13] X. Liang, Y. He, F. Liu, B. Wang, T. Zhong, B. Quan, G. Lu, Solid-state potentiometric H2 S sensor combining NASICON with Pr6 O11 -doped SnO2 electrode, Sens. Actuators B 125 (2007) 544–549. [14] A. Varga, Z. Bozóki, M. Szakáll, G. Szabó, Photoacoustic system for on-line process monitoring of hydrogen sulfide (H2 S) concentration in natural gas streams, Appl. Phys. B 85 (2006) 315–321. [15] E.A. Wintner, T.L. Deckwerth, W. Langston, A. Bengtsson, D. Leviten, P. Hill, M. A.Insko, R.Dumpit E.VandenEkart, C.F. Toombs, C. Szabo, A monobromobimane-based assay to measure the pharmacokinetic profile of reactive sulphide species in blood, Brit. J. pharmacol. 160 (2010) 941–957. [16] M.E. Franke, T.J. Koplin, U. Simon, Metal and metal oxide nanoparticles in chemiresistors: does the nanoscale matter? Small 2 (2006) 36–50. [17] N. Yamazoe, New approaches for improving semiconductor gas sensors, Sens. Actuators B 5 (1991) 7–19. [18] Y. Zhang, J.Q. Xu, Q. Xiang, H. Li, Q.Y. Pan, P.C. Xu, Brush-like hierarchical ZnO nanostructures: synthesis, photoluminescence and gas sensor properties, J. Phys. Chem. C 113 (2009) 3430–3435. [19] P.C. Xu, Z.X. Cheng, Q.Y. Pan, J.Q. Xu, Q. Xiang, W.J. Yu, Y.L. Chu, High aspect ratio In2 O3 nanowires: synthesis, mechanism and NO2 gas-sensing properties, Sens. Actuators B 130 (2008) 802–808. [20] X.L. Wang, N. Aroonyadet, Y.Z. Zhang, M. Meckenburg, X. Fang, H.T. Chen, E. Goo, C.W. Zhou, Aligned epitaxial SnO2 nanowires on sapphire: growth and device applications, Nano. Lett. 14 (2014) 3014–3022. [21] Y.L. Zhang, Y.X. Yang, H.X. Han, M. Yang, L. Wang, Y.N. Zhang, Z.X. Jiang, C. Li, Ultra-deep desulfurization via reactive adsorption on Ni/ZnO: The effect of ZnO particle size on the adsorption performance, Appl. Catal. B 119 (2012) 13–19. [22] J. Zhang, S.R. Wang, Y. Wang, M.J. Xu, H.J. Xia, S.M. Zhang, P.W. Huang, X.Z. Guo, S.H. Wu, ZnO hollow spheres: preparation, characterization, and gas sensing properties, Sens. Actuators B 139 (2009) 411–417. [23] B.C. Cheng, Y.H. Xiao, G.S. Wu, L.D. Zhang, Controlled growth and properties of one-dimensional ZnO nanostructures with Ce as activator/dopant, Adv. Funct. Mater. 14 (2004) 913–919. [24] T. Seiyama, A. Kato, K. Fujiishi, M. Nagatani, A new detector for gaseous components using semiconductive thin films, Anal. Chem. 34 (1962) 1502–1503. [25] I.I. Novochinskii, C.S. Song, X.L. Ma, X.S. Liu, L. Shore, J. Lampert, R.J. Farrauto, Low-temperature H2 S removal from steam-containing gas mixtures with ZnO for fuel cell application. 1. ZnO particles and extrudates, Energ. Fuel. 18 (2004) 576–583. [26] S.K. Pandey, K.-H. Kim, K.-T. Tang, A review of sensor-based methods for monitoring hydrogen sulfide, TRAC-Trends Anal. Chem. 32 (2012) 87–99. [27] Q. Wan, Q.H. Li, Y.J. Chen, T.H. Wang, X.L. He, J.P. Li, C.L. Lin, Fabrication and ethanol sensing characteristics of ZnO nanowire gas sensors, Appl. Phys. Lett. 84 (2004) 3654–3656. [28] Z. Fan, D. Wang, P.-C. Chang, W.-Y. Tseng, J.G. Lu, ZnO nanowire field-effect transistor and oxygen sensing property, Appl. Phys. Lett. 85 (2004) 5923–5925. [29] H.Y. Huang, P.C. Xu, D. Zheng, C.Z. Chen, X.X. Li, Sulfuration-desulfuration reaction sensing effect of intrinsic ZnO nanowires for high-performance H2 S detection, J. Mater. Chem. A 3 (2015) 6330–6339. [30] J. Huang, Q. Wan, Gas sensors based on semiconducting metal oxide one-dimensional nanostructures, Sensors 9 (2009) 9903–9924. [31] S.R. Morrison, Selectivity in semiconductor gas sensors, Sens. Actuators 12 (1987) 425–440. [32] J.C. Belmonte, J. Puigcorbe, J. Arbiol, A. Vila, J.R. Morante, N. Sabate, I. Gracia, C. Cane, High-temperature low-power performing micromachined suspended
272
[33]
[34]
[35]
[36]
[37] [38] [39]
[40] [41] [42]
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272 micro-hotplate for gas sensing applications, Sens. Actuators B 114 (2006) 826–835. S. Semancik, R.E. Cavicchi, M.C. Wheeler, J.E. Tiffany, G.E. Poirier, R.M. Walton, J.S. Suehle, B. Panchapakesan, D.L. Devoe, Microhotplate platforms for chemical sensor research, Sens. Actuators B 77 (2001) 579–591. C. Ducso, E. Vazsonyi, M. Adam, I. Szabo, I. Barsony, J.G.E. Gardeniers, A. Vandenberg, Porous silicon bulk micromachining for thermally isolated membrane formation, Sens. Actuators A 60 (1997) 235–239. B.S. Choi, Y.W. Lee, S.W. Kang, J.W. Hong, J. Kim, I. Park, S.W. Han, Multimetallic alloy nanotubes with nanoporous framework, ACS Nano 6 (2012) 5659–5667. X.L. Li, C. Li, Y. Zhang, D.P. Chu, W.I. Milne, H.J. Fan, Atomic layer deposition of ZnO on multi-walled carbon nanotubes and its use for synthesis of CNT-ZnO heterostructures, Nanoscale Res. Lett. 5 (2010) 1836–1840. S.Y. Pung, K.L. Choy, X. Hou, C.X. Shan, Preferential growth of ZnO thin films by the atomic layer deposition technique, Nanotechnology 19 (2008) 435609. J.F. Conley, L. Stecker, Y. Ono, Directed integration of ZnO nanobridge devices on a Si substrate, Appl. Phys. Lett. 87 (2005) 223114. J.S. Na, B. Gong, G. Scarel, G.N. Parsons, Surface polarity shielding and hierarchical ZnO nano-architectures produced using sequential hydrothermal crystal synthesis and thin film atomic layer deposition, ACS Nano 3 (2009) 3191–3199. N. Yamazoe, G. Sakai, K. Shimanoe, Oxide semiconductor gas sensors, Catal. Surv. Asia 7 (2003) 63–75. P. Atkins, J. de Paula, Atkins’ Physical Chemistry, 7th ed., Oxford University Press, Oxford UK, 2007. M.W. Ahn, K.S. Park, J.H. Heo, J.G. Park, D.W. Kim, K.J. Choi, J.H. Lee, S.H. Hong, Gas sensing properties of defect-controlled ZnO-nanowire gas sensor, Appl. Phys. Lett. 93 (2008) 263103.
Biographies Ying Chen received her B.S. degree (2005) in material science from Fudan University, and her Ph.D. degree (2010) in Microelectronics and Solid State Electronics from Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. She is currently an assistant professor at State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. Her present research interests include microcantilever chemical sensors and lab-on-chip. Pengcheng Xu received his B.S. degree from Zhengzhou Institute of Light Industry, M.S. degree from Shanghai University and Ph. D. degree from University of Chinese Academy of Sciences. Now he is an assistant professor at the State Key Laboratory of
Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. His present research covers advanced sensing materials (like nanowires, hyperbranched polymers and nanoporous materials), micro/nanofluidic chips and MEMS/NEMS-based chemical sensors. Since 2016, he has been appointed as a member of Youth Innovation Promotion Association CAS. Tao Xu received his B.S. degree (2014) in Chemical Engineering from Nanchang University. Presently he is pursuing his Master degree in Shanghai Institute of Technology. His research is focused on chemical sensing materials and gas sensors. Dan Zheng received B.S. degree in Physical Chemistry from Zhejiang University, Hangzhou, China, in 1987. She received Master degree in Chemical Engineering from Shenyang University of Chemical Engineering, Shenyang, China, in 1999. She received PhD degree in Physical Chemistry of Material from Shanghai Institute of Microsystem and Information Technology, Chinese academy of Sciences in 2009. Dr. Zheng is now a professor in School of Chemical and Environmental Engineering, Shanghai Institute of Technology, China. Her research interest lies in micro fuel-cells, electro-chemistry of nanomaterials and sensing nanomaterials for bio/chemical detection. Xinxin Li received B.S. degree from Tsinghua University, Beijing and Ph.D. degree from Fudan University, Shanghai. Thereafter, he sequentially worked in Hong Kong University of Science and Technology as a Research Associate, in Nanyang Technological University, Singapore as a Research Fellow and, then, joined Tohoku University, Japan, as a Lecturer (COE fellowship). From 2001 to now, he has been a professor and now serves as the Director of the State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. He has also served as Adjunct Professor in Fudan University, Shanghai Jiaotong University, Dalian University of Technology, Shanghai Tech University and Suzhou University. From 2009–2013, he had served as Consultant Professor for World Class University Program of Korean in Chonnam National University, Korea. He was granted the National Science Fund for Distinguished Young Scholar in 2007. His Ph.D. student was awarded National Excellent 100 Ph.D. Dissertation in 2009. Prof. Li’s research interest includes micro/nano sensors and MEMS/NEMS. He has invented about 100 patents and published more than 300 papers in referred journals and conferences (including about 170 SCI journal papers). He ever served as TPC member for the conferences of IEEE MEMS, Transducers and IEEE Sensors. He is now the editorial member for Journal of Micromechanics and Microengineering and the International Steering Committee member for Transducers.
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
ZnO-nanowire size effect induced ultra-high sensing response to ppb-level H2 S Ying Chen a , Pengcheng Xu a , Tao Xu b , Dan Zheng b , Xinxin Li a,∗ a State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China b School of Chemical and Environmental Engineering, Shanghai Institute of Technology, Shanghai 201418, China
a r t i c l e
i n f o
Article history: Received 3 May 2016 Received in revised form 10 August 2016 Accepted 19 August 2016 Available online 21 August 2016 Keywords: Nano-size sensing effect ZnO nanowires High sensitivity H2 S sensor ppb-level gas detection
a b s t r a c t Presented is nano-size effect of ZnO nanowires (NWs) on sensitivity of hydrogen sulfide (H2 S) detection in ambient atmosphere. The sensing mechanism of ZnO NWs for H2 S detection is found coming from nano-size effect that governs both the sulfuration reaction when H2 S is introduced and the following desulfuration reaction when the gas is removed from air. Based on the nano-material property of smaller dimension featuring higher chemical-activity at surface, the sensitivity of ZnO NWs to ultra-low concentration H2 S is expected to be increased by shrinking the NW diameter. In our experiment, suspended micro hotplates for chemiresistive sensing are fabricated to investigate the nano-size sensing effect. Besides conventional ZnO NWs with the diameter as 50 nm, a novel tree-branched nano-structure of ZnO NWs (with the branch diameter as 20 nm) is in situ grown onto the sensing-area of the micro hotplate. In order to investigate the sensing effect, both the 20 nm NW sensor and the 50 nm NW sensor are used to experimentally detect H2 S with concentrations in the range of ppb-ppm. The 20 nm ZnO-NWs sensor realizes resoluble sensing to 5 ppb H2 S, while the 50 nm sensor can only response to H2 S with concentration higher than 50 ppb. The well validated nano-size effect of thinner NWs featuring higher sensing response can be extended for various chemical sensing nano-materials. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Hydrogen sulfide (H2 S) is a kind of toxic, corrosive and inflammable gas. Gaseous H2 S has distinct fetor of rotten egg that is commonly formed in nature and released during the decay of organic matter. The H2 S gas is normally generated from sewage, coal mines, and petroleum/natural gas industries, and it is widely utilized in various chemical industries. When the gas concentration of H2 S is higher than 250 ppm, it may lead to death [1]. Even at lower concentration level, H2 S is harmful to human bodies. The harmful H2 S molecules can be rapidly absorbed by human’s lung and easily cause respiratory system disease, unconsciousness neurological sequelae and cardiovascular-related death [2]. Recently recognized as an important endogenous gas, H2 S can be implicated in diverse physiological and pathophysiological process, such as regulation of inflammation, brain development, blood pressure regulation and metabolism [3–9]. It has been reported that the hydrogen sulfide in human body is at ppm or sub-ppm
∗ Corresponding author. E-mail address: [email protected] (X. Li). http://dx.doi.org/10.1016/j.snb.2016.08.120 0925-4005/© 2016 Elsevier B.V. All rights reserved.
level, and it is indicated that trace-concentration detection of H2 S is helpful for disease diagnosis [10]. For accurate analysis and earlystage prognosis of diseases, H2 S detection at sub-ppm or lower concentration is highly demanded. Numerous chemical sensors have been investigated and developed, which are mainly based on metal-oxide semiconductor sensing [11,12], electrochemical detection [13] and optical measurement [14]. However, the lack of effective detection method and analytical tools for monitoring of trace-concentration H2 S has hindered the biomedical research that is correlative with the endogenous gas [15]. Therefore, it is of great importance to develop ppb-ppm level detectable H2 S sensors. Several technical approaches have been reported to improve sensing performance of H2 S sensors. At present, the H2 S sensors by using semiconducting metal-oxide nano-materials are attractive due to high sensitivity. For example, ZnO, In2 O3 and SnO2 [16–20] can specifically response to the presence of the gas by changing electronic conductivity, thereby having been widely used in various applications like industrial safety, product quality control, clinical diagnostics and home-safety alarms. Among the various investigated metal-oxide sensing materials, n-type ZnO nanowires (NWs) materials have been intensively explored and
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
widely used for gas sensing, due to their high mobility of conductive electrons, good chemical/thermal stability and low cost [21–24]. In recent years, ZnO nanostructures have been used for building chemiresistive H2 S sensors [25] and the measure range of these H2 S sensing devices generally focus in ppm level [26]. Thanks to the high surface-to-volume ratio, the nano-structured materials provide large adsorbing interface that indeed helps to increase sensitivity [27]. For instance, mesoporous thin-films or nanowire arrays have shown several folds enhanced H2 S sensing properties compared to the traditional sensors that are based on solid-state thin-film sensing materials [11,28]. However, to the best of our knowledge, it is really difficult to improve sensitivity by tens of folds or more by only using the increasing surface-to-volume method. Being an exception, our group recently reported that, when the diameter of ZnO wires is reduced to tens of nanometers, a sulfuration-desulfuration dual-stage sensing mechanism becomes to dominate and induces dramatically enhanced sensitivity. 50nm-diameter ZnO NWs can detect 50 ppb H2 S due to the dual-step reaction [29]. This new sensing mechanism can be originated from nano-size effect. When ZnO NWs are thin enough, the surface becomes active enough to be sulfurated by H2 S to form ZnS. This decreases the number of adsorbed oxygen molecule from ambience and causes thinner electron depletion layer. The sulfuration leads to drastic decrease of the ZnO resistance and enhance the sensing signal. After H2 S is removed, the metastable intermediate of ZnS will be quickly desulfurized back to ZnO by the ambient oxygen and the resistance quickly recovers to its initial state. This two-step reaction sensing mechanism is different from the widely used conventional one for detecting reductive gases in ambient air. The mechanism for conventional metal-oxide sensors is based on semiconductor resistance change which comes from electron depletion layer variation induced by surface adsorbed oxygen species [30]. When H2 S molecules react with the oxygen species, the change of depletion layer causes variation of ZnO resistance and outputs the sensing signal [17,31]. It has been reported that bulk-material of ZnO responds to H2 S with the concentration at tens of ppm level [11] while the 50 nm ZnO NWs sensor features reversible sulfuration-desulfuration sensing mechanism and outputs response to 50 ppb H2 S. Obviously, nano-size effect brings the high sensing performance. Along with size shrinkage of the nanomaterial, the confinement of bulk to surface becomes weaker and the material surface becomes more active. Compared with the sensitivity improvement caused by the increased surface-to-volume ratio, the nano-material size effect induced improvement in sensitivity is much more significant [29]. This work is just inspired by the sulfuration-desulfuration H2 S sensing mechanism for ZnO of nano-diameter. Herein, it is expected that continuously thinning the ZnO NWs induces continuously enhanced active surface and improved sensitivity. Therefore, herein we conceive to explore the ZnO nano-size effect on H2 S detection performance. For the purpose, the method for in situ grown thinner ZnO NWs than 50 nm is studied. Finally, we develop a technique to region-selectively grown tree-branched 20 nm ZnO NWs at the comb-shaped electrode region of a chemiresistive sensing micro-chip. The sensing microchip consists of a suspended micro-hotplate where the pair of comb electrodes is accommodated for the sensing nano-material growth. Finite element modeling (FEM) thermal simulation is performed to optimize the sensor structure and improve temperature uniformity across the micro-hotplate. For H2 S sensing experiment, 20 nm ZnO NWs and the conventional 50–60 nm ZnO NWs are grown, respectively. The detectable results of 5 ppb H2 S by using the 20 nm ZnO NWs well proves the regulation effect of ZnO nano-size on H2 S sensitivity.
265
2. Experimental 2.1. Design and fabrication of suspended sensing micro-hotplate Micromachined hotplates for gas sensors are advantageous in thermal isolation, low power consumption and low-cost manufacturing [32–34]. In this work, a sandwich-structured micro-hotplate comprises comb-finger electrode layer, intermediate SiO2 isolating layer and heating resistor layer. The three-layer structure is formed on a suspended silicon nitride beam-plate. Fig. 1 shows our designed micro-hotplate for gas sensor. The suspended plate is connected with four narrow supporting beams that are clamped to silicon substrate and the geometry is optimized to minimize thermal conduction to outside. The circular plate is 300 m in diameter and each support beam is 150 m in length and 20 m in width. The radial platinum resistor is for heating the plate, as platinum features very linear temperature coefficient of resistance (TCR) for easy temperature calibration. The resistance versus plate temperature is pre-calibrated in a precise oven. Above the Pt heater, a thin layer of SiO2 is employed to insulate the heater and the electrodes at top. To measure the chemiresistance change of ZnO NWs, a pair of comb-finger electrodes is densely placed above. The gap distance between the two comb-fingers is 10 m, which can be electrically bridged by the grown branching ZnO NWs into a chemiresistor. As is shown in Fig. 1(b), the ZnO NWs will in situ grown at the central sensing area, i.e., over the comb-finger electrodes. The sensor has four interconnection lines, with two for the micro-heater and the other two for chemiresistance signal readout. By using silicon micromachining technology, the chip is fabricated in single-side polished (100) silicon wafer, with the process steps sketched in Fig. 2 and described as follows: (1) Deposition 1 m-thick low-stress silicon nitride by low pressure chemical vapor deposition (LPCVD). (2) A 30 nm/300 nm Ti/Pt composite thin-film is deposited by electron-beam evaporation and patterned by lift-off process to form the heater. (3) A 400nm-thick insulating layer of SiO2 is formed using plasma enhanced (PE) CVD. The contact holes for the Pt heater are opened. Then, a 30 nm/300 nm-thick Cr/Au layer is evaporated and patterned to form the comb-finger electrodes and the wire-bonding pads of the Pt heater. (4) The SiO2 and silicon nitride layers are sequentially etched by reactive ion etching to form the shape of the suspended plate and the beams. (5) The silicon beneath the plate is etched off by isotropic etching of vapor-phase XeF2 and the micro plate is released for free-standing. After the wafer is diced, the chips are mounted on PCB and wire bonding is implemented. At this step, the chip is ready for in situ growth of ZnO NWs sensing material. It is worth pointing out that the metal materials used in the micro sensor can sustain the temperature process during synthesize of ZnO NWs. 2.2. In situ grown of ZnO NWs The ZnO NWs, with the diameter as about 50–70 nm, are synthesized from aqueous solution and in situ grown at the sensing area of the sensor by using a modified hydrothermal method [35] and shown in Fig. 3(1)–(2). The surface of the sensor is plasma pretreated (with Harrick PDC-002 equipment) for 5 min to eliminate surface contamination. Afterwards, 5 mM zinc acetate in ethanol is dropwise coated in the sensing area by using a commercial micromanipulator (Eppendorf, PatchMan NP2), with the aid of inspection under a microscope (Leica, DM4000). Later, the sensor chip is dried in air for 10 s, rinsed with ethanol and then, blow dried with a pure nitrogen stream. This coating step needs to be repeated for at least three times. Covered with the seed film of zinc acetate crystallites, the seeded chip is heated to 350 ◦ C in air for 20 min to yield a nanolayer of ZnO seeds. Then, ZnO NWs are in-situ grown at the surface of the chip by using hydrothermal method with an aqueous solu-
266
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 1. (a) Designed micro-hotplate, which consists of comb-finger electrodes, insulating layer, Pt heater and suspended plate. (b) Schematic of the suspended micro-hotplate.
Fig. 2. Fabrication steps of the micro-hotplate in 3D and cross-sectional views. (1) Deposition silicon nitride on silicon wafer. (2) Pt heater is formed by e-beam evaporation and lift-off process. (3) SiO2 insulating layer is deposited by PE-CVD. Cr/Au layer is evaporated and patterned to form the comb-finger electrodes and the wire-bonding pads. (4) Suspended plate and support beams are shaped by RIE. (5) The Si beneath the plate is etched off to suspend the hotplate.
tion that contains zinc nitrate (25 mM), hexamethylenetetramine (HMTA, 25 mM) and polyethylenimine (PEI, 6 mM). The bottomup hydrothermally growth process for the NWs is carried out at 90 ◦ C for 4 h. Finally, the sensing chip with the grown ZnO NWs of 50–70 nm in diameter is taken out from the solution, washed with deionized water and dried for sensing. For in-situ synthesis of the branched ZnO-NWs in the sensing area, a two-step growth process is developed and shown in Fig. 3(1)–(4). The first step is for growing the trunks to certain length and the second step is for growing the thinner branches from the thicker tree trunks. The first-step trunk growth is similar with the process for the 50 nm ZnO NWs. After the required length of the trucks is reached, the sensor chip is heated to 350 ◦ C in air for 20 min to remove the residual HMTA and PEI. Since the ZnO trunks are vertically grown on the substrate, it is difficult to form a uniform seed-layer on the top surface of the trunks by spin- or dip-coating. In that case, the seed-layer would prefer to aggregate near the bot-
tom part of the trunks that would result in generation of undesired grass-like ZnO NWs instead of the branched nano-structure. To address this issue, highly conformal atom layer deposition (ALD) is herein employed to deposit the seed-layer [36–39]. The branch growing steps are detailed as follows. Using Zn(C2 H5 )2 and H2 O as co-precursors, 60 cycles seed-layer is constructed at 150 ◦ C by using ALD. Lift-off process is applied to remove the ZnO seed layer at undesired regions. Then, identical hydrothermal-synthesis process is repeated to grown isotropically stretched ZnO nano-branches from each trunk. Finally, the chip is taken out from the solution, washed with deionized water and dried for the further sensing applications.
2.3. Characterization Scanning electron microscopy (SEM) characterization is performed in a Hitachi S-4800 cold-field equipment for observing the
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
267
Fig. 3. Schematic illustration of the synthesis process of the ZnO NWs and nano-branched ZnO NWs.
nanostructure of ZnO NWs and the morphology of micro-hotplate. Optical microscope images are taken under a Leica DM 4000 microscope. The TCR of the Pt heater is calibrated in a temperature programmable oven. The whole sensor chip is introduced in the oven for temperature ramping experiment from 25 ◦ C to 150 ◦ C, with increment of 25 ◦ C. 2.4. Sensing experiment H2 S gas of various concentrations is generated by a lab-made gas generator which mixes the standard concentration H2 S gas and the ambient air. The standard low-concentration H2 S gas of 1000 ppm is purchased from Shenkai Gas Corporation, Shanghai, China. The pure air is connected from our lab air compressing system where the relative-humidity range is well controlled. The H2 S gas sensing experiment is carried out in a testing chamber of 20 L in volume. The sensor is sealed in the testing chamber that is connected with the gas generator for sensing experiment. A DC power source, Agilent34410A multi-meter and a personal computer are used to supply the micro heater and real-time record the chemiresistive sensing signal. 3. Results and discussion 3.1. Micro-hotplate for gas sensor Fig. 4(a) shows the top-view of the fabricated hotplate chip. As the insulating SiO2 is transparent, it is quite clear that the electrodes are located above the heater. SEM image in Fig. 4(b) shows the plate suspended with the four supporting beams. The thermal mismatch stress introduced by the composite layers of different materials is optimally designed and, as a result, the suspended hotplate is quite flat and shows good mechanical robustness in our experiment. By using the finite-element analysis tool of COMSOL, thermal analysis is implemented for the hotplate. Only conductive and convective energy losses have been taken into account since the irradiative energy losses are very small due to the not very high working temperature and the small area of the heated plate. The simulation results for the temperature distribution on top surface of the hotplate are depicted in Fig. 5(a) and (b). It can be seen that, under power supply of DC 1.6 V, the temperature of the hotplate surface is 573 K (or 300 ◦ C) at the center and the
temperature difference across the diameter of the plate is less than 5 ◦ C. Better temperature uniformity is achieved in the sensing chemiresistive sensing area. At the clamping end of the supporting beams, the temperature steeply drops to room temperature. As is shown in Fig. 5(c), the experimentally calibrated TCR value for the Pt heating resistor is 1.7 × 10−3 /◦ C and the there is a linear relationship between resistance and temperature. After calibration, a linear relationship between temperature and heating voltage is obtained that is shown in Fig. 5(d). The calibrated relationship is T = 0.939 + 157.267 V, where T is the resistance-representative temperature of the hotplate and V is the heating voltage. 3.2. In-situ growth of branched ZnO NWs As shown in the SEM images of Fig. 6(a), conventional ZnO NWs with the diameter of 50–70 nm is locally grown in the sensing area of the micro-hotplate. The ZnO NWs can be grown on top of both the gold electrode fingers and the SiO2 finger intervals. The magnified SEM image in Fig. 6(b) and the cross-sectional image in Fig. 6(c) shows that, the ZnO NWs are densely grown at the electrodes region and the adjacent NWs cross-link with each other to form the chemiresistor. Fig. 6(c) and (d) shows that the nanowires are generally 2–3 m in length and 50–67 nm in diameter. For locally growing the branched ZnO NWs, the situation of the trunks formed at the first step is similar to that of the aforementioned conventional ZnO NWs. After the second-step growth of the branches, the ZnO nano-structure formed in the sensing area of the hotplate is shown in Fig. 6(e). High-density ZnO branches are successfully grown on the previously formed ZnO trunks. As is shown in Fig. 6(f), the diameter of the nano-branches is generally 20 nm, with the deviation being less than ±5 nm. Attribute to the inter-linked dense branches, stable resistances value of the sensitive chemiresistor can be measured. It is worthy pointing out that, the 20 nm-diameter ZnO trunks previously formed at the first-step become slightly thicker during the repeated hydrothermal growth at the second-step. 3.3. Nano-size effect on sensing performance By adjusting DC voltage, the Pt resistive micro-heater can provide desired working temperature for the H2 S sensing material. In the sensing experiment, the sensing response is represented by
268
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 4. (a) Optical microscope image of the micro-hotplate integrated with a heater and a pair of comb-finger electrodes. (b) SEM image of the fabricated micro-hotplate that is suspended with a wide gap from silicon substrate.
Fig. 5. (a) Simulated temperature distribution under DC 1.6 V heating. (b) Temperature distribution across the hotplate diameter. (c) Experimentally measured Pt resistance shows linear relationship with temperature of the hotplate. The temperature coefficient of resistance (TCR) is calibrated as 0.0017/◦ C. (d) Experimentally calibrated linear relationship of between hotplate temperature and heating voltage.
the ratio of the resistance change (R) to the initial resistance in clean air (R). For the branched ZnO NWs sensor, its response to 1 ppm H2 S is measured and shown in Fig. 7, where stepped temperatures of 150–450 ◦ C is achieved by heating the hotplate (with increment of 50 ◦ C). In the relatively lower temperature range from 150 ◦ C to 300 ◦ C, the response steeply increases with the temperature increasing and, exhibits the maximum response signal at 300 ◦ C. When the temperature is further increased to over 300 ◦ C, the response decreases quickly and, the response at 350 ◦ C becomes much lower than that at 200 ◦ C. The “volcano-shape” temperature dependent behavior of the sensing response is in accordance with the theory as reported in literature [40] and the experimental phenomenon can be explained as follows. For H2 S detection by using ZnO nanowires, the activity of ZnO NWs increases along with the working temperature raising. The increased activity of ZnO NWs induces the enhanced sensing response by generating ZnS intermediate via sulfuration reaction [29]. When the sensing response
reaches the maximum value at 300 ◦ C, adsorption and desorption of H2 S molecules get to an equilibrium state. Afterwards, as the temperature increases, the adsorption of H2 S gas decreases, meanwhile, the desorption of H2 S from ZnO is enhanced. If the number of H2 S molecules involved in the above-mentioned reaction decrease (i.e. desorption of H2 S increases), it must generate less ZnS intermediate. On the other hand, as a kind of metastable intermediate, ZnS is quite unstable at high temperature because it tends to be oxidized in ambient air to form ZnO. Hence, the quantity of ZnS becomes even less at higher temperature (>300 ◦ C). At higher working temperature, the two factors mentioned-above brought less ZnS and lead to a drastic decline in the sensing response of the sensor. Based on the experimental results and the above-mentioned discussions, 300 ◦ C is the optimal working-temperature for achieving highest response of the branched ZnO NWs sensor. For comparison, the sensor with the 20 nm branch-structured ZnO and the sensor with the 50–70 nm conventional ZnO NWs are
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
269
Fig. 6. (a) Top-view SEM shows the locally grown ZnO NWs in the sensing area of the hotplate sensor. (b) Magnified SEM image shows the densely grown ZnO NWs on the hotplate. (c) Side-view of the dense ZnO NWs. (d) Further close-up view shows that the diameter of the NWs is ranged from 50 to 70 nm. (e) After the second step for growing the branches, the SEM image shows the tree-branched nano-structure, and the adjacent branches are cross-linked. (f) The diameter of the ZnO branches is generally measured as 20 nm (±5 nm).
Fig. 7. Temperature dependent sensing response to 1 ppm H2 S gas of the same branched ZnO sensor.
both introduced to detect ppb-level H2 S. At identical temperature of 300 ◦ C under air atmosphere, the resistance value of 50–70 nm ZnO sensors is 60.7 k and the resistance value of 20 nm branched ZnO sensor is 47.3 k. Obviously, the resistance value of the 50 nm ZnO sensor is higher than the value of 20 nm ZnO sensor. According to the SEM characterization results as shown in Fig. 6, there are more dense nanowires grown in the 20 nm ZnO sensor and a branched nanostructure is formed. Hence, there are more electron conduction channels formed in the 20 nm nanowires sensor. Therefore, compared with 50–70 nm ZnO NWs, the 20 nm branched ZnO sensor exhibits a smaller chemiresistance value. During the experiment, both the 20 nm branched sensor and the 50–70 nm ZnO sensor are sequentially exposed to H2 S gas, where the gas concentration is varied from 5 ppb to 200 ppb. As is shown in Fig. 8(a), the 20 nm branched ZnO sensor exhibits recoverable responses at 300 ◦ C. For detecting H2 S of 5 ppb, the measured sensing response (R/R) is about 4%, while the noise-floor is less than 1%. To 200 ppb H2 S, the response signal is as high as 44%. In addition, it can be seen that the sensor response increases along with increasing H2 S concentration. The sensing response and H2 S concentration have a linear relationship that is plotted in Fig. 8(b). At ultra-low gas
270
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
Fig. 8. (a) Sensing experimental results of the 20 nm branched ZnO sensor to trace-level H2 S gas at 300 ◦ C, with concentrations in the range of 5 ppb to 200 ppb. (b) A linear fitting line is obtained between the sensor response and H2 S concentration, which can be expressed as y = 0.002x + 0.036, herein x represents the H2 S concentration and y is the sensing response. (c) Magnified sensing curve of 20 nm sensor to 50 ppb H2 S and the fitted response curve. (d) Response to 50 ppb–200 ppb H2 S for the sensor with the conventional 50 nm ZnO NWs at 300 ◦ C.
Fig. 9. For the 20 nm branched ZnO sensor, the selectivity to twelve kinds of interfering gases is measured.
pressure (i.e. gas concentration), the experimentally obtained linear relationship can be explained based on Langmuir sorption theory [41]. Based on the results in Fig. 8(a), the limit of detection (LOD) of the 20 nm branched ZnO sensor is better than 3 ppb, which is estimated by assessing the signal noise-floor and assuming a linear relationship between the response and concentration. During the sensing experiment, H2 S gas is switched off when the output-signal almost gets stable. The time for H2 S gas-off has been optimized based on the multiple experimental confirmations and the obtained sensing results only show a negligible influence for the following discussion. Magnifying the time axis of the sensing curve, it can be clearly observed that the signal increases drastically at the initial stage but increases very slowly to achieve the final equilibrium state thereafter. Taking the measuring cycle of 20 nm sensor to 50 ppb H2 S as example, the sensing curve can be well fitted by using Langmuir theory [see Fig. 8(c)]. According to the fitted result, it will take about 19 min to reach the completely equilibrium state, which is 4.4 times longer than the practical measuring
time (4.3 min), i.e. the measuring time can be saved by more than 75%. On the other hand, the sensing response increment during the slowly increase period is quite small and can be even ignored. Based on the fitted response curve, the fitted sensing response (S) is 0.153 and the practical sensing response (S ) in this study is obtained as 0.142. Thus, it can be calculated that S /S = 92.8%. The abovementioned discussion confirms the experiment error of the sensing response is considered acceptable but the detection time can be reduced by 75%. Above all, it is reasonable to switch off H2 S gas at an opportune time during each detection cycle. For comparison, the H2 S response curves of the 50–70 nm conventional ZnO sensor are experimentally obtained at 300 ◦ C and shown in Fig. 8(d). The sensor cannot output resoluble response until H2 S concentration is increased to 50 ppb. The sensing response to 50 ppb H2 S is merely 3%. Being resoluble to 5 ppb trace-concentration H2 S, the sensor with the 20 nm branched ZnO is much more sensitive than the sensor with the conventional 50–70 nm ZnO NWs. This result can well verify the nano-size effect on the H2 S sulfuratation/desulfuration
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272
sensing mechanism. Along with size shrinkage of the sensing nanomaterial, its surface should become more and more active. The high response achieved with the 20 nm branched ZnO nanowires can be mainly attributed to the nano-size induced surface effect. In addition, The ZnO nano-branches provide many nano-junctions at the branch roots located on the trunks, which possibly have minor help effect to enhance the sensitivity [42]. The reason why we grow the branched nanowires instead of conventional 20 nm nanowires to examine the size effect lie in that the very thin nanowires tend to prostrate on the surface of the sensor. The sparse prostrate-grown ZnO nanowires easily lead to instable chemiresistance. For the high-sensitivity sensor with the 20 nm branched ZnO, the selectivity is experimentally assessed at 300 ◦ C, where we select twelve kinds of interfering gases/vapors of methanol, ethanol, acetone, hexane, methanol, ethyl acetate, benzene, toluene, NH3 , CH4 , H2 and CO2 . The experimental results are compared together in Fig. 9, with the concentration values of the interfering gases denoted. According to the results, obviously, the 20 nm treebranched ZnO-nanowire H2 S sensor has a satisfactory selectivity to the interfering gases. 4. Conclusion This work experimentally reveals nano-size effect induced high sensitivity and fine resolution of ZnO-nanowires to ppb-level H2 S gas. The sulfuration/desulfuration sensing mechanism of nanoscale ZnO sensing chemiresistors is different from the widely used conventional one, and it indicates that higher sensitivity of ZnO NWs to ultra-low concentration H2 S can be achieved by size shrinking of the sensing nano-material. Locally grown on a suspended micro hotplate, a novel structure of 20 nm tree-branched ZnO NWs is used as chemiresistive sensing material to experimentally detect ppb-ppm level H2 S. In order to reveal the nano-size sensing effect, conventional 50–70 nm ZnO NWs are also prepared and examined for comparison. The results verify the nano-size sensing effect. The tree-branched 20 nm ZnO NWs sensor exhibits resoluble detection limit of 5 ppb H2 S and much higher response than the 50–70 nm ZnO NW sensor. Besides that the very sensitive H2 S sensor is achieved, the nano-size effect induced high performance sensing mechanism is promising to be applied to more chemical sensors. Conflict of interests The authors declare no conflict of interests. Acknowledgements This research is supported by MOST of China (2016YFA0200800), NSF of China (61401446, 6152781861604163, 91323304, 61321492), NSF of Shanghai (15ZR1447300), Science and Technology Commission of Shanghai (14521106100).P.C.X appreciates the financial support of the Youth Innovation Promotion Association CAS (2016213). References [1] M.D. Shirsat, M.A. Bangar, M.A. Deshusses, N.V. Myung, A. Mulchandani, Polyaniline nanowires-gold nanoparticles hybrid network based chemiresistive hydrogen sulfide sensor, Appl. Phys. Lett. 94 (2009) 083502. [2] D.C. Glass, A Review of the health-effects of hydrogen-sulfide exposure, Ann. Occup. Hyg. 34 (1990) 323–327. [3] M. Whiteman, S. Le Trionnaire, M. Chopra, B. Fox, J. Whatmore, Emerging role of hydrogen sulfide in health and disease: critical appraisal of biomarkers and pharmacological tools, Clin. Sci. 121 (2011) 459–488. [4] S.J. Choi, B.H. Jang, S.J. Lee, B.K. Min, A. Rothschild, I.D. Kim, Selective detection of acetone and hydrogen sulfide for the diagnosis of diabetes and halitosis using SnO2 nanofibers functionalized with reduced graphene oxide nanosheets, ACS. Appl. Mater. Interfaces 6 (2014) 2588–2597.
271
[5] M.F. Rosario-Alomar, T. Quinones-Ruiz, D. Kurouski, V. Sereda, E.B. Ferreira, L. De Jesus-Kim, S. Hernandez-Rivera, D.V. Zagorevski, J. Lopez-Garriga, I.K. Lednev, Hydrogen sulfide inhibits amyloid formation, J. Phys. Chem. B 119 (2015) 1265–1274. [6] Y. Zhou, W. Chen, J. Zhu, W. Pei, C. Wang, L. Huang, et al., Nanoprobes: inorganic–organic hybrid nanoprobe for nir-excited imaging of hydrogen sulfide in cell cultures and inflammation in a mouse model, Small 10 (2014) 4802. [7] N.H. Kim, S.J. Choi, D.J. Yang, J. Bae, J. Park, I.D. Kim, Highly sensitive and selective hydrogen sulfide and toluene sensors using Pd functionalized WO3 nanofibers for potential diagnosis of halitosis and lung cancer, Sens. Actuators B 193 (2014) 574–581. [8] C. Szabo, Hydrogen sulphide and its therapeutic potential, Nat. Rev. Drug Discov. 6 (2007) 917–935. [9] A. Papapetropoulos, A. Pyriochou, Z. Altaany, G. Yang, A. Marazioti, Z. Zhou, M.G. Jeschke, L.K. Branski, D.N. Herndon, R. Wang, C. Szabo, Hydrogen sulfide is an endogenous stimulator of angiogenesis, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 21972–21977. [10] M. Whiteman, P.K. Moore, Hydrogen sulfide and the vasculature: a novel vasculoprotective entity and regulator of nitric oxide bioavailability? J. Cell Mol. Med. 13 (2009) 488–507. [11] P.S. Shewale, G.L. Agawane, S.W. Shin, A.V. Moholkar, J.Y. Lee, J.H. Kim, M.D. Uplane, Thickness dependent H2 S sensing properties of nanocrystalline ZnO thin films derived by advanced spray pyrolysis, Sens. Actuators B 177 (2013) 695–702. [12] Y. Zhang, Q. Xiang, J.Q. Xu, P.C. Xu, Q.Y. Pan, F. Li, Self-assemblies of Pd nanoparticles on the surfaces of single crystal ZnO nanowires for chemical sensors with enhanced performances, J. Mater. Chem. 19 (2009) 4701–4706. [13] X. Liang, Y. He, F. Liu, B. Wang, T. Zhong, B. Quan, G. Lu, Solid-state potentiometric H2 S sensor combining NASICON with Pr6 O11 -doped SnO2 electrode, Sens. Actuators B 125 (2007) 544–549. [14] A. Varga, Z. Bozóki, M. Szakáll, G. Szabó, Photoacoustic system for on-line process monitoring of hydrogen sulfide (H2 S) concentration in natural gas streams, Appl. Phys. B 85 (2006) 315–321. [15] E.A. Wintner, T.L. Deckwerth, W. Langston, A. Bengtsson, D. Leviten, P. Hill, M. A.Insko, R.Dumpit E.VandenEkart, C.F. Toombs, C. Szabo, A monobromobimane-based assay to measure the pharmacokinetic profile of reactive sulphide species in blood, Brit. J. pharmacol. 160 (2010) 941–957. [16] M.E. Franke, T.J. Koplin, U. Simon, Metal and metal oxide nanoparticles in chemiresistors: does the nanoscale matter? Small 2 (2006) 36–50. [17] N. Yamazoe, New approaches for improving semiconductor gas sensors, Sens. Actuators B 5 (1991) 7–19. [18] Y. Zhang, J.Q. Xu, Q. Xiang, H. Li, Q.Y. Pan, P.C. Xu, Brush-like hierarchical ZnO nanostructures: synthesis, photoluminescence and gas sensor properties, J. Phys. Chem. C 113 (2009) 3430–3435. [19] P.C. Xu, Z.X. Cheng, Q.Y. Pan, J.Q. Xu, Q. Xiang, W.J. Yu, Y.L. Chu, High aspect ratio In2 O3 nanowires: synthesis, mechanism and NO2 gas-sensing properties, Sens. Actuators B 130 (2008) 802–808. [20] X.L. Wang, N. Aroonyadet, Y.Z. Zhang, M. Meckenburg, X. Fang, H.T. Chen, E. Goo, C.W. Zhou, Aligned epitaxial SnO2 nanowires on sapphire: growth and device applications, Nano. Lett. 14 (2014) 3014–3022. [21] Y.L. Zhang, Y.X. Yang, H.X. Han, M. Yang, L. Wang, Y.N. Zhang, Z.X. Jiang, C. Li, Ultra-deep desulfurization via reactive adsorption on Ni/ZnO: The effect of ZnO particle size on the adsorption performance, Appl. Catal. B 119 (2012) 13–19. [22] J. Zhang, S.R. Wang, Y. Wang, M.J. Xu, H.J. Xia, S.M. Zhang, P.W. Huang, X.Z. Guo, S.H. Wu, ZnO hollow spheres: preparation, characterization, and gas sensing properties, Sens. Actuators B 139 (2009) 411–417. [23] B.C. Cheng, Y.H. Xiao, G.S. Wu, L.D. Zhang, Controlled growth and properties of one-dimensional ZnO nanostructures with Ce as activator/dopant, Adv. Funct. Mater. 14 (2004) 913–919. [24] T. Seiyama, A. Kato, K. Fujiishi, M. Nagatani, A new detector for gaseous components using semiconductive thin films, Anal. Chem. 34 (1962) 1502–1503. [25] I.I. Novochinskii, C.S. Song, X.L. Ma, X.S. Liu, L. Shore, J. Lampert, R.J. Farrauto, Low-temperature H2 S removal from steam-containing gas mixtures with ZnO for fuel cell application. 1. ZnO particles and extrudates, Energ. Fuel. 18 (2004) 576–583. [26] S.K. Pandey, K.-H. Kim, K.-T. Tang, A review of sensor-based methods for monitoring hydrogen sulfide, TRAC-Trends Anal. Chem. 32 (2012) 87–99. [27] Q. Wan, Q.H. Li, Y.J. Chen, T.H. Wang, X.L. He, J.P. Li, C.L. Lin, Fabrication and ethanol sensing characteristics of ZnO nanowire gas sensors, Appl. Phys. Lett. 84 (2004) 3654–3656. [28] Z. Fan, D. Wang, P.-C. Chang, W.-Y. Tseng, J.G. Lu, ZnO nanowire field-effect transistor and oxygen sensing property, Appl. Phys. Lett. 85 (2004) 5923–5925. [29] H.Y. Huang, P.C. Xu, D. Zheng, C.Z. Chen, X.X. Li, Sulfuration-desulfuration reaction sensing effect of intrinsic ZnO nanowires for high-performance H2 S detection, J. Mater. Chem. A 3 (2015) 6330–6339. [30] J. Huang, Q. Wan, Gas sensors based on semiconducting metal oxide one-dimensional nanostructures, Sensors 9 (2009) 9903–9924. [31] S.R. Morrison, Selectivity in semiconductor gas sensors, Sens. Actuators 12 (1987) 425–440. [32] J.C. Belmonte, J. Puigcorbe, J. Arbiol, A. Vila, J.R. Morante, N. Sabate, I. Gracia, C. Cane, High-temperature low-power performing micromachined suspended
272
[33]
[34]
[35]
[36]
[37] [38] [39]
[40] [41] [42]
Y. Chen et al. / Sensors and Actuators B 240 (2017) 264–272 micro-hotplate for gas sensing applications, Sens. Actuators B 114 (2006) 826–835. S. Semancik, R.E. Cavicchi, M.C. Wheeler, J.E. Tiffany, G.E. Poirier, R.M. Walton, J.S. Suehle, B. Panchapakesan, D.L. Devoe, Microhotplate platforms for chemical sensor research, Sens. Actuators B 77 (2001) 579–591. C. Ducso, E. Vazsonyi, M. Adam, I. Szabo, I. Barsony, J.G.E. Gardeniers, A. Vandenberg, Porous silicon bulk micromachining for thermally isolated membrane formation, Sens. Actuators A 60 (1997) 235–239. B.S. Choi, Y.W. Lee, S.W. Kang, J.W. Hong, J. Kim, I. Park, S.W. Han, Multimetallic alloy nanotubes with nanoporous framework, ACS Nano 6 (2012) 5659–5667. X.L. Li, C. Li, Y. Zhang, D.P. Chu, W.I. Milne, H.J. Fan, Atomic layer deposition of ZnO on multi-walled carbon nanotubes and its use for synthesis of CNT-ZnO heterostructures, Nanoscale Res. Lett. 5 (2010) 1836–1840. S.Y. Pung, K.L. Choy, X. Hou, C.X. Shan, Preferential growth of ZnO thin films by the atomic layer deposition technique, Nanotechnology 19 (2008) 435609. J.F. Conley, L. Stecker, Y. Ono, Directed integration of ZnO nanobridge devices on a Si substrate, Appl. Phys. Lett. 87 (2005) 223114. J.S. Na, B. Gong, G. Scarel, G.N. Parsons, Surface polarity shielding and hierarchical ZnO nano-architectures produced using sequential hydrothermal crystal synthesis and thin film atomic layer deposition, ACS Nano 3 (2009) 3191–3199. N. Yamazoe, G. Sakai, K. Shimanoe, Oxide semiconductor gas sensors, Catal. Surv. Asia 7 (2003) 63–75. P. Atkins, J. de Paula, Atkins’ Physical Chemistry, 7th ed., Oxford University Press, Oxford UK, 2007. M.W. Ahn, K.S. Park, J.H. Heo, J.G. Park, D.W. Kim, K.J. Choi, J.H. Lee, S.H. Hong, Gas sensing properties of defect-controlled ZnO-nanowire gas sensor, Appl. Phys. Lett. 93 (2008) 263103.
Biographies Ying Chen received her B.S. degree (2005) in material science from Fudan University, and her Ph.D. degree (2010) in Microelectronics and Solid State Electronics from Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. She is currently an assistant professor at State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. Her present research interests include microcantilever chemical sensors and lab-on-chip. Pengcheng Xu received his B.S. degree from Zhengzhou Institute of Light Industry, M.S. degree from Shanghai University and Ph. D. degree from University of Chinese Academy of Sciences. Now he is an assistant professor at the State Key Laboratory of
Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. His present research covers advanced sensing materials (like nanowires, hyperbranched polymers and nanoporous materials), micro/nanofluidic chips and MEMS/NEMS-based chemical sensors. Since 2016, he has been appointed as a member of Youth Innovation Promotion Association CAS. Tao Xu received his B.S. degree (2014) in Chemical Engineering from Nanchang University. Presently he is pursuing his Master degree in Shanghai Institute of Technology. His research is focused on chemical sensing materials and gas sensors. Dan Zheng received B.S. degree in Physical Chemistry from Zhejiang University, Hangzhou, China, in 1987. She received Master degree in Chemical Engineering from Shenyang University of Chemical Engineering, Shenyang, China, in 1999. She received PhD degree in Physical Chemistry of Material from Shanghai Institute of Microsystem and Information Technology, Chinese academy of Sciences in 2009. Dr. Zheng is now a professor in School of Chemical and Environmental Engineering, Shanghai Institute of Technology, China. Her research interest lies in micro fuel-cells, electro-chemistry of nanomaterials and sensing nanomaterials for bio/chemical detection. Xinxin Li received B.S. degree from Tsinghua University, Beijing and Ph.D. degree from Fudan University, Shanghai. Thereafter, he sequentially worked in Hong Kong University of Science and Technology as a Research Associate, in Nanyang Technological University, Singapore as a Research Fellow and, then, joined Tohoku University, Japan, as a Lecturer (COE fellowship). From 2001 to now, he has been a professor and now serves as the Director of the State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences. He has also served as Adjunct Professor in Fudan University, Shanghai Jiaotong University, Dalian University of Technology, Shanghai Tech University and Suzhou University. From 2009–2013, he had served as Consultant Professor for World Class University Program of Korean in Chonnam National University, Korea. He was granted the National Science Fund for Distinguished Young Scholar in 2007. His Ph.D. student was awarded National Excellent 100 Ph.D. Dissertation in 2009. Prof. Li’s research interest includes micro/nano sensors and MEMS/NEMS. He has invented about 100 patents and published more than 300 papers in referred journals and conferences (including about 170 SCI journal papers). He ever served as TPC member for the conferences of IEEE MEMS, Transducers and IEEE Sensors. He is now the editorial member for Journal of Micromechanics and Microengineering and the International Steering Committee member for Transducers.