发表文章毕业论文201701-Sun-nanotechnology-room-temperature CO thermoelectric Gas Sensor based on Au Co3O4 Catalyst tablet

Nanotechnology PAPER Room-temperature CO Thermoelectric Gas Sensor based on Au/Co3O4 Catalyst Tablet To cite this article: L Sun et al 2017 Nanotechnology 28 075501 View the article online for updates and enhancements. Related content Liquefied petroleum gas sensor based on manganese (iii) oxide and zinc manganese (iii) oxide nanoparticles Shiva Sharma, Pratima Chauhan and Shahid Husain - One-dimensional fossil-like - Fe2O3carbon nanostructure: preparation, structural characterization and application as adsorbent for fast and selective recovery of gold ions from aqueous solution Poernomo Gunawan, Wen Xiao, Marcus Wen Hao Chua et al. - Effect of Calcination Temperature on Surface Oxygen Vacancies and Catalytic Performance Towards CO Oxidation of Co3O4 Nanoparticles Supported on SiO2 Jin-bing Li, Zhi-quan Jiang, Kun Qian et al. - This content was downloaded from IP address 32 on 08/02/2018 at 13:26 Room-temperature CO Thermoelectric Gas Sensor based on Au/Co3O4Catalyst Tablet L Sun1, W L Luan2,3, T C Wang2, W X Su1and L X Zhang1 1School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, Peoples Republic of China 2Key Laboratory of Pressure systems and Safety (MOE), School of Mechanical and Power Engineering, East China University of Science and Technology, Shanghai 200237, Peoples Republic of China E-mail: luan Received 13 October 2016, revised 25 November 2016 Accepted for publication 15 December 2016 Published 17 January 2017 Abstract A carbon monoxide (CO) thermoelectric (TE) gas sensor was fabricated by affi xing a Au/Co3O4 catalyst tablet on a TE fi lm layer. The Au/Co3O4catalyst tablet was prepared by a co- precipitation and tablet compression method and its possible catalytic mechanism was discussed by means of x-ray diffraction, fi eld emission scanning electron microscopy, high resolution transmission electron microscopy, x-ray photoelectron spectroscopy, temperature-programmed reduction of hydrogen, Fourier transform infrared spectroscopy and Brunauer-Emmett-Teller analysis. The optimal catalyst, witha Au content of 10 wt%, was obtained ata calcination temperature between 200 and 300C. The small size of the Au nanoparticles, high specifi c surface, the existence of Co3+and water-derived species contributed to high catalytic activity. Based on the optimal Au/Co3O4catalyst tablet, the CO TE gas sensor worked at room temperature and showed a response voltage signal (V) of 23 mV, high selectivity among hydrogen and methane, high stability, and a fast response timeof 106 s for 30000 ppm CO/air. In addition,a CO concentration in the range of 500030 000 ppm could obviously be detected and exhibited a linear relationship with V. The CO TE gas sensor provides a promisingoption for the detection of CO gas at room temperature. S Online supplementary data available from /NANO/28/075501/mmedia Keywords: CO thermoelectric gas sensor, Au/Co3O4catalyst tablet, room temperature, Co- precipitation (Some fi gures may appear in colour only in the online journal) 1. Introduction Carbon monoxideis a colorless, odorless, and poisonous gas at room temperature, which can be easily released from imperfect combustion of fuels or gas leaks. Even low CO concentrations can also be fatal especially in poorly ventilated indoor environments 13. Thus the use of CO sensors in indoorenvironmentshasattractedincreasedattention. Towork in anindoor environment, the CO sensor musthave asimple structure, low operating temperature, good selec- tivity, long-term stability, and fast response. Currently, both electrochemical type 4, 5 and semi-conductor type 6, 7 CO sensors are commonly commercialized; however, they still- lack selectivity, stability, and confi gurability. Thermoelectric gas sensors, composed of a catalyst and a TE fi lm layer, has attracted increasing attention due to its good selectivity, simple structure, and low cost 812. Matsumiya et al fi rst reporteda novel CO TE gas sensor witha Au/Co3O4 thin-fi lm catalyst prepared by a sputtering method and a TE SiGe layerplaying a core role 10. The sensor showed the operating temperature as 200C and an output voltage signal of 0.63 mV for 30 000 ppm CO in air. After that, more research was focused on decreasing the operating temperatureto reducethe energy consumed and to Nanotechnology Nanotechnology 28 (2017) 075501 (8pp)doi:10.1088/1361-6528/aa53f9 3 Author to whom any correspondence should be addressed. 0957-4484/17/075501+08$33.00 2017 IOP Publishing LtdPrinted in the UK1 improve safety. The key to lowering the operating temper- ature is the preparation of the CO catalyst with high catalytic activity 11. Firstly, a Au/TiO2 ceramic thick-fi lm was used as the catalyst 12. The operating temperature was reduced to 150C lower than the Matsumiyas result, but the CO selectivity was poor. Later, Nishibori reporteda new CO TE gas sensor based ona Au/Co3O4catalyst prepared by mixing Au colloid and cobalt oxide powder directly 13. As a result, the operating temperature was still above 150C although the stability was enhanced. At present, Xu et al fabricateda CO TE gas sensor with a Co-Ce catalyst tablet that exhibited a lower operating temperatureof 92 C 14. Thus, it is still diffi cult to designa CO TE gas sensor towork at room temperature. Untilnow, most work to enhance catalytic activityhas beenrelated to the changein active components and catalyst supports 15, 16. However, the preparation technology of the catalysts was also an important factorin improvingcatalytic activity 17. In this paper, we tried to preparea Au/Co3O4 catalyst tablet with high room-temperature activity via a co- precipitation and tablet compression method. The preparation conditions were investigated systematically andthe optimal value was found.The catalysts obtained were characterized and the catalytic mechanism was analyzed. Based on the optimal catalyst, the CO TE gas sensor was fabricated and its properties includingits selectivity, sensitivity, and stability were studied. 2. Experimental 2.1. Catalyst preparation The tablet Au/Co3O4catalyst was prepared by the co-pre- cipitation and tablet compression method. A NH4HCO3 solution was added drop-wise into a mixed aqueous solution of Co(NO3)26H2O and HAuCl4with continuous stirring until neutral. After aging for 4h, the precipitate was fi ltered and extensively washed with deionized water. The whole precipitation and fi lter processwas carried out at room temperature. Then, the precipitate was put in an oven at 100C for 2h. Dried powder was grinded and formed into tablets with the compression pressure varied from 1 to 9MPa. Finally, it was calcined ata temperature rangeof 200 to 600C in air. 2.2. Catalyst characterization An energy dispersion spectroscopy (EDS, Falcon) was used to estimate the Au content. Field emission scanning electron microscopy (FE-SEM, S-4200) and high resolution trans- mission electron microscopy (HR-TEM, JEM2100F) were applied to observe the morphology of thecatalyst. Ultraviolet visible (UV-vis) spectra analysis was carried out on a spec- trophotometer (Cary 50, Varian). The SBET, pore volume, and pore diameter of the sampleswere measured by means of the Brunauer-Emmett-Teller (BET) method on a micrometric instrument (ASAP2010). X-ray diffraction (XRD) patterns were obtained to identify the formation of cobalt oxides using an x-ray diffractometer (D/max2550 VB/PC). X-ray photo- electron spectroscopy (XPS) was performed on a spectro- meter (ESCA 5600CI). The details of thespectra of the Au 4f and Co 2p regions were measured in the ranges 8090 eV and 775805 eV,respectively.Fouriertransforminfrared spectroscopy (FT-IR) was achieved on a spectrophotometer (JASCOFT/IR-4200)inthewavenumberrange of4000400 cm1in air. Temperature-programmed reduc- tion (TPR) was carried out in a fl ow reactor coupled with a thermal conductivity detector in a Micrometrics AutoChem II 2920 (USA). 2.3. Fabrication and measurement of the sensor The structure of the CO TE gas sensor is exhibited in fi gure 1. The TE gas sensor is composed of a TE layer and a catalyst tablet (Au/Co3O4). The TE layer (bismuth-telluride PN4 couples) was deposited onto a quartz glass substrate by magnetron sputtering 9. The catalyst was placed into the center of the TE fi lm. The concrete size of the device was as follows: 20 mm (length)20 mm (width)1mm (thick- ness) for the substrate, 8mm1mm350 nm for the TE layer, 8mm (diameter)2mm (thickness) for the Au/ Co3O4catalyst tablet. The TE gas sensor was placed in an airproof chamber. Testing gases (30 000 ppm or 12800 ppm CO/air) and air were alternatively blown through the surface of the catalysts, and the fl ow rate was concisely controlled at 100ml min1by a gas fl owmeter. The measurement was conducted at a con- stant temperature, which was provided by a stabilized tube furnace. The temperature difference (T) signal, from the hot-side to the cold-side junction, was monitored to evaluate the activity of the catalysts. The properties of the CO sensor were detected based on the response voltage signal (V) defi ned by equation (1): D=-( )VVV1 testair where Vtestand Vairare the voltage signals generated in testing gases and in air, respectively. According to the Seebeck principle, the V is closely related to the T. Acquired data were automatically collected and analyzed by a data acqui- sition/switch unit (Agilent 34970A, USA) and Labview (NI Crop., USA). Figure 1.Schematic drawing of the CO thermoelectric gas sensor. 2 Nanotechnology 28 (2017) 075501L Sun et al 3. Results and Discussion 3.1. Effect of preparation conditions Among preparation conditions, the compression pressure and the calcination temperature had an obvious infl uence on the catalytic performance of the catalyst tablet. In fi gure 2, the high value of compression pressure was found at 1MPa and the T of 34 C was obtained. The T continuously decreases with the increasein the compression pressure from 1 to 9MPa, which indicated a low compression pressure could contribute toa high T. However, when the pressure is lower than 1MPa, it was hard to form the catalyst tablet. Thus, the compression pressure of 1MPa was considered as a high value point. Further, a 10 wt% Au/Co3O4catalyst pre- pared at 1MPa was used in the investigation of the calcina- tion temperature. Figure 3 manifests the effect of the calcination temper- ature at the rangeof 200 to 600 C on the T.A steady T of 34 C is shown between 200 and 300C. After that, T reveals a sharp drop, and then over 500 C a stable trend appears. In Table S1 (see supplementary information), the SBETof theAu/Co3O4catalyst is decreased from 89 to 27 m2g1with the calcination temperature changed from 300 to 500C, which may cut the catalytic activity down. Thus, calcination below 300 C was regarded as an optimal condition. XRD patterns of the catalysts prepared at different cal- cination temperature are shown in fi gure 4. XRD peaks of the cobalt oxide are well consistent with the data of the JCPDS #42-1467 fi le of the Co3O4with the cubic phase. The peaks at 2 values of 18.9, 31.2, 36.9, 38.6, 44.8, 55.6, 59.3, and 65.2 correspond to the crystal planes of (111), (220), (311), (222), (400), (422), (511), and (440) of crystalline Co3O4, respectively. It was suggested the Co3O4phase was obtained by calcination at 300 to 500 C. With increasing the temperature from 300 to 500 C, the crystallite size of Co3O4 had a distinct variation from 31 to 57 nm calculated based on the Scherrer formula. The growth of the Co3O4could causea declinein the catalytic activity. The diffraction peaks of Au (38.2, 44.4, 64.6, 77.5) are overlapped with a part of theCo3O4 peaks, resulting in diffi culty distinguishing the Au phase from the XRD patterns. Figure 5(a) displays the FE-SEM images of Au/Co3O4, in which the fl ower-like particles of 100 nm are clearly observed. As shown in fi gure S1(see supplementary infor- mation), the surface plasmon resonance absorption wave- length of the catalysts is under 520 nm, which illustrates that Au nanoparticles possessan average size of less than 10 nm 18. Therefore, the Au nanoparticles were diffi cult to identify from theSEM images. The fl ower-like particles could be agglomerates of Co3O4nanoparticles with a small average size ofless than 40 nm. The Co3O4nanoparticles con- tinuously grew with increasing calcination temperature, which made the fl ower-like structure divide into Co3O4 nanoparticles withan average size of 60 nm (fi gure 5(b). The growth of the Co3O4nanoparticles could lead to the decrease of the catalytic activity. TEM images were used to observe the morphology of the Au nanoparticles. As displayed in the Figure 2.Dependence of temperature difference on compression pressure from 1 to 8 MPa. Figure 3.Effect of the calcination temperature increased from 200 to 600 C on the temperature difference. Figure 4.XRD patterns of Au/Co3O4calcined at 300 C (A), 400 C (B) and 500 C (C), respectively. 3 Nanotechnology 28 (2017) 075501L Sun et al inset of fi gure 5(a), Au nanoparticles are homogeneously distributed on the support and their average size is 34nm in diameter. With the increasein calcination temperature, the average size of the Au nanoparticles grows to 8nm, exhibited in the set of fi gure 5(b). Generally, Au with a smaller size could provide better performance 19. Thus, the high activity of the Au/Co3O4catalyst tablet could be attributed to the small size of the Au and Co3O4particles. XPS measurement was performed to examine the che- mical species of the catalyst. Figure 6 exhibits the XPS spectra of the Au 4f and Co 2p regions on the Au/Co3O4 catalyst calcined at 300, 400, and 500 C, respectively. In fi gure 6(a), the fi rst doublet with peaks at 83.8 and 87.5eV is the characteristic of Au0, and the peaks at 84.6 and 88.6eV are assigned to Au+20. The Au+species are decreased with the increase of the calcination temperature. Only Au0is observed at the calcination temperature of 500 C. It was considered that the existence of Au0 and Au+with an optimal ratio could lead to high catalytic activity. Thus, the large reduction of Au+could be one of the reasons for the decreasein the catalytic activity. The peaks of Au3+are located at 86.3 and 89.9eV, which cannot appear at the XPS spectra. It indicated Au3+was reduced completely to Au+ and Au0during the calcination process. The Co2p3/2at 779.30.3eV is indicative of Co3+, whereas the Co2p3/2at 780.9 eV corresponds to Co2+ . In fi gure 6(b), the Co2p3/2is shifted from 779.6 to 780.4 eV with the calcination temper- ature enhanced from 300 to 500 C, which illustrated the increase of Co2+. Furthermore, the spin-orbit splitting of the Co 2p peaks (E) was also used to analyze the oxidation state of cobalt. Generally, the value of E is 16.0, 15.0, and 15.2 eV corresponding to the CoO, Co2O3, and Co3O4, respectively 21, 22. In fi gure 6(b), the E is varied from 15.2 to 15.5 eV with the increasing calcination temperature, which confi rms an increase of Co2+again 22.Co2+has no ability to reduce CO gas. The appearance of Co2+with the increase of the calcination temperature should also be responsible for the decreasein the catalytic activity. Figure 7 presents the FT-IR spectra of the tablet catalysts calcined at different temperatures. Two bands are shown in Figure 5.FE-SEM images of Au/Co3O4calcined at 300 C (a) and 500 C (b), respectively. The inset is the HR-TEM image of (a). Figure 6.XPS spectra of Au 4 f (a) and Co 2p (b) regions on the Au/Co3O4catalyst calcined at 300 C (A), 400 C (B) and 500 C (C), respectively. 4 Nanotechnology 28 (2017) 075501L Sun et al the spectra, which can be respectively ascribed to thes- tretching vibration of water-derived species (O-H of absorbed water or surface hydroxyls) in a 32003500 cm1region and their deformation vibration at 1637 cm1. The FT-IR result suggested that there were a large amount of water-derived species on the surface of the catalysts calcined at 300 and 400C, and an obvious reduction in the absorbance intensity for the catalysts calcined at 500 C. Combined withfi gure 3, it could be found that the decrease of water-derived species could weaken the catalytic activity. The function of the water- derived species has been illustrated in the study of Au/CeO2 catalysts 23. The water-derived species absorbed on the support could promote the molecular oxygen adsorption and activity. And they had a strong synergetic action with Au+ species. The form of Au+-OH could help the decomposition of the stable carbonate species generated from the absorption of CO on Au+species. 3.2. Effect of Au on the catalytic activity The effect of the Au content on the activity of Au/Co3O4 catalyst was investigated based on the catalyst tablet prepared at optimized conditions (1MPa, 200C). Table 1 shows the T is directly dependent on the Au content measured by EDS. The T keeps rising until the Au content reaches to 10wt%, which indicates Au could be the active site of the Au/Co3O4catalyst. The value of T dis- plays a little change when the Au content was between 10 and 15 wt%. After that the T exhibits a decline trend with the Au content enhanced from 15 to 30wt%. According to pre- vious reports, the particle size of Au got bigger with the increasein the Au content 13, 24. Thus, the decrease of the catalytic activity could be related to the increase in the size of the Au nanoparticles. In order to further study how the Au content affects the T, the SBETvalues of the catalysts with different Au con- tents are also presented in table 1. The highest SBETvalue of Au/Co3O4catalyst is obtained atan Au content of 10 wt%; however, a furtherincrease in the Au content results in a lower SBETvalue. The trend i