Abstract

Development and/or design of new materials and/or structures for effective gas sensor applications with fast response and high sensitivity, selectivity, and stability are very important issues in the gas sensor technology. This critical review introduces our recent progress in the development of meso-/nanoporous semiconducting metal oxides and their applications to gas sensors. First, the basic concepts of resistive gas sensors and the recent synthesis of meso-/nanoporous metal oxides for gas sensor applications are introduced. The advantages of meso-/nanoporous metal oxides are also presented, taking into account the crystallinity and ordered/disordered porous structures. Second, the synthesis methods of meso-/nanoporous metal oxides including the soft-template, hard-template, and temple-free methods are introduced, in which the advantages and disadvantages of each synthetic method are figured out. Third, the applications of meso-/nanoporous metal oxides as gas sensors are presented. The gas nanosensors are designed based on meso-/nanoporous metal oxides for effective detection of toxic gases. The sensitivity, selectivity, and stability of the meso-/nanoporous gas nanosensors are also discussed. Finally, some conclusions and an outlook are presented.

1. Introduction

Air pollution caused by toxic, flammable, and explosive gases, such as CO, H2S, NH3, NO2, CH4, C3H8, and H2, is one of the critical factors that contribute to global warming, climate change, and harm to human health [110]. Development and fabrication of a device for early detection and/or alarm of certain flammable, explosive, and toxic gases are extremely necessary. For this purpose, gas sensors have been invented and developed toward tract detection and/or concentration monitoring of such pollution gases [3, 4]. Gas sensors are devices that detect and/or measure the track or concentration of analytic gaseous agents. However, the sensing device has difficulty in measuring or screening the analytic gas molecules directly; hence, the measured signals are usually converted into the change/variation in physical and/or chemical quantities, such as temperature, conductivity, frequency, capacitance, color, or pressure. According to the working principle of devices and/or the analytic species, gas sensors can be classified into different types. Based on (i) the working principle, the kinds of gas sensors include the capacitance, solid electrolyte, acoustic wave, and resistive types; (ii) analytic agents can process volatile organic compounds (VOCs), explosive or toxic gas sensor classification. Scheme 1(a) illustrates a general gas sensor, in which the adsorption of gaseous molecules to the sensing materials leads to the change in readout signals. Generally, any change in physical or chemical properties of materials upon gaseous molecule exposures can also be used as significant response of gas sensors. In practical applications, physical or chemical changes are usually converted into measureable or electrical signals through a transducer for easy measurements [5, 6]. The human nose is an example of a sensitive gas sensor that can detect odor gases, such as H2S, NH3, and VOCs. However, it cannot detect certain odorless gas such as CO, CO2, and H2. Despite the different gas sensors that have been developed, such as optical sensors, electrochemical sensors, calorimetric sensors, and resistive sensors, the most common and simplest gas sensor remains to be the resistive type [110]. The operation of this type of gas sensor is based on the change in electrical conductance (resistance) of the sensing materials upon gaseous molecule exposures. However, the low sensitivity, slow recovery, poor selectivity, and high working temperature of the bulk, thin, or thick films metal oxide-based gas sensors limit their potential applications. Our recent studies have been dedicated to the development of novel nanostructures for high performance gas sensors with high sensitivity, fast response/recovery time, and good selectivity [36].

Metal Oxide-Based Resistive Gas Sensors. In the early 1960s, Seiyama and Taguchi introduced the gas sensors that are operated based on variation in their electrical resistance (conductance) upon a chemical reaction and/or adsorption between the analytical gas species and the surface of metal oxide semiconducting layers [1, 2]. Since then, investigations on resistive gas sensors have received a great deal of attention because of their low cost, simple completion, online-monitoring, and good reliability for real-time control systems, as well as their diverse practical applications in environmental monitoring, transportation, security, defense, space missions, energy, agriculture, and medicine [124]. Various metal oxide semiconductors with different geometrical structures, such as nanoplates, thin film, nanoparticles, nanorods, nanotubes [7], nanofibers [8], nanowires, and hollow spheres [9, 10], have been developed for gas sensor applications. Both p-type and n-type semiconductors [316] have been applied for detection of different gases, such as C2H5OH [17], NO, NH3 [18], CO [19], H2S [20], O3 [21], and NO2 [2224]. Wide bandgap metal oxide semiconductors such as ZnO, SnO2, WO3, In2O3, and CuO are commonly used as sensing materials in resistive gas sensors because of their high sensitivity to different gases [127]. Recently, reduction in the size of the device, miniaturization of production expense, and improvement of sensor performances for rapid response and high sensitivity, selectivity, stability, and feasibility have gained significant interests in the field of gas sensor technology. Moreover, development and exploration of new materials, structures, and geometries for effective gas sensor applications are of extreme interests [25, 26]. Gao et al. prepared the ZnO materials of different morphologies such as nanorod arrays, nanoribbon bundles, nanosheets, nanocubes, and nanoparticles [27]. They also used the ZnO nanorods and hollow spheres for ethanol sensors, where the porous structure of hollow spheres showed a better gas sensing characteristic compared with the nanorods [26]. Studies on the synthesis and application of meso-/nanoporous materials to gas sensors are also increasing [2224]. This highlight review focuses on mesoporous metal oxides, from synthesis to effective gas sensor applications. The use of mesoporous metal oxides for gas sensor applications has some advantages in enhancing sensor performance through total exposure of sensing sites to the analytic gases.

Meso-/Nanoporous Metal Oxides Based Resistive Gas Sensors. Nanostructures of meso-/nanoporous semiconducting metal oxides with large specific surface area are ideal materials for improving gas sensing performances by enhancing sensing sites and total exposures to analytical gases [2224]. According to their working principles, the responses of metal oxide-based gas sensors are dependent on various parameters, such as (i) the density and mobility of the main carriers, (ii) surface modification, (iii) grain size effects, and (iv) specific surface area and surface chemical properties of the materials [2831]. The two former parameters are controlled by the type of sensing materials (free electron in n-type and hole in p-type semiconductors) and the doping elements [3, 4, 20]. By contrast, the latter two (iii and iv) are dependent on the morphologies, shape, and size of materials and can be controlled through the fabrication of nanostructured meso-/nanoporous materials [9, 10, 2224]. Essentially, nanostructured meso-/nanoporous materials with nanosize crystals and superior specific surface area could accelerate the gaseous adsorption/desorption processes during gas sensing measurements because such processes occur mainly on the surface of the sensing materials [31]. Therefore, the use of nanostructured meso-/nanoporous materials to improve gas sensor performances is of advantage [32]. In recent years, investigation on the synthesis and application of meso-/nanoporous semiconducting metal oxides to gas sensors has gained increasing attention [33]. The number of reports on the mesoporous gas sensors has increased exponentially since 2004, as shown in Figure 1 (source: web of science core collection).

Taking into account the synthesis processes, meso-/nanoporous metal oxides can be fabricated by soft- or hard-template methods. The soft-template method [34] uses polymer surfactants as structure guide agents in controlling the mesoporous structures, whereas the hard-template [35] applies preformed meso-/nanoporous materials as structure guide. Normally, the soft-template is a direct-synthesis, whereas the hard-template is often undirected. To date, both soft- and hard-template methods have been applied to synthesis of mesoporous metal oxides and their composites for gas sensor applications, despite their advantages and disadvantages compared with other methods [3444].

2. Synthesis of Meso-/Nanoporous Metal Oxides

2.1. Soft-Template Synthesis of Meso-/Nanoporous Metal Oxides

The most commonly used method for direct-synthesis of mesoporous metal oxides is the use of polymer surfactants as soft-templates [31]. This method utilizes the structure of the polymers as a structure-direct agent for fabrication of mesoporous metal oxides, which is very effective, specifically for synthesis of highly ordered mesoporous silica. Figure 2 illustrates the soft-template synthesis of ordered mesoporous metal oxides. Typical synthesis involves the wet chemical processes (sol-gel and hydrothermal) for self-assembly of surfactant and metal precursors to form the hybrid of metal oxides and polymers, followed by template removal.

However, the as-synthesized materials were amorphous phases or hybrid of metal oxides and polymers. After calcination at high temperature, the amorphous phases became semicrystalline or crystalline. However, the porous structure was distorted, leading to lower specific surface area [34]. Yang et al. reported on the generalized direct syntheses of large-pore mesoporous metal oxides, including TiO2, ZrO2, Al2O3, Nb2O5, Ta2O5, WO3, HfO2, and SnO2, and mixed oxides SiAlO3.5, SiTiO4, ZrTiO4, Al2TiO5, and ZrW2O8 [34]. The syntheses used amphiphilic poly (alkylene oxide) block copolymers as soft-template structure-directing agents in nonaqueous solutions to organize the network-forming metal oxide species. The results exhibited ordered mesoporous oxides containing nanocrystalline domains within thick amorphous walls. Cheng et al. reported on the synthesis of mesoporous tungsten oxide thin film using triblock copolymer P123 as structure-directing agent [36]. The synthesis involved the sol-gel processes for thin film formation, solvent extraction, and/or calcination to remove the copolymer template. The mesoporous metal oxides fabricated by direct-synthesis method have also been applied to gas sensors fabrication. Sun et al. reported on the synthesis of mesoporous α-Fe2O3 nanostructures for gas sensor applications, in which the α-Fe2O3 was synthesized by the soft-template synthesis method using the triblock copolymer F127 surfactant [37]. The synthesized α-Fe2O3 material has a disordered mesoporous structure with a specific surface area of 128 m2 g−1 and a pore size of about 7 nm. The gas sensing properties of synthesized mesoporous α-Fe2O3 were investigated for detection of flammable, toxic, and corrosive gases, such as ethanol, acetone, gasoline, heptane, formaldehyde, acetic acid, 1-butanol, and 2-propanol. Our recent work reported on the synthesis of mesoporous WO3 through an instant direct-template method [38]. The F108 polymer surfactant was utilized as soft-template for the synthesis, similar to the conventional synthesis of mesoporous silica. However, the polymer template was converted into carbon by annealing in an inert gas (N2) at high temperature to maintain the porous structure of materials before the removing of the soft-template by calcination in the air. The synthesized WO3 materials have good porosity and high crystallinity, but disordered porous structures (Figure 3). The soft-template directed syntheses of meso-/nanoporous metal oxides have been investigated for gas sensors applications; however, the disordered porous structures and low crystallinity of the prepared materials have limitations and drawbacks that should be addressed to improve the gas sensing performances of rapid response, high sensitivity, and long-term stability [31, 34].

2.2. Hard-Template Synthesis of Meso-/Nanoporous Metal Oxides

Hard-template synthesis of nanostructured materials has been well known since the introduction of the synthesis of nanomaterials (nanotubes and nanowires) by using anodic alumina membrane as templates [39]. Since the success fabrication of ordered mesoporous silica, the hard-template method has been developed for the synthesis of crystalline ordered mesoporous metal oxides [4044]. Figure 4 shows a diagram for the hard-template synthesis of ordered mesoporous metal oxides. Typically, the hard-template synthesis includes the (i) fabrication of mesoporous structure materials (silica or carbon) as hard-templates by the conventional soft-template direct-synthesis using copolymer surfactants, the (ii) filling of metal oxide precursors into the nanopores of hard-templates, the (iii) calcination that converts the metal precursors to metal oxides, and the (iv) selective etching of the hard-templates. In some cases, the filling and calcination processes (ii and iii) were repeated several times to ensure the sufficient filling of metal oxides into the nanopores of templates [43]. The common and popular hard-templates used for the synthesis of semiconducting metal oxides mesoporous are the highly ordered silica and carbon [35, 4348]. The fabrication method is the post synthesis, where the mesoporous template is fabricated first, and then the metal precursor is filled in the pores to generate the desired oxide. This synthesis is sometimes called nanocasting [35, 4547] or replication method [48].

Hard-template method has been applied effectively to the synthesis of various mesoporous semiconducting metal oxides such as ZnO [43], Cr2O3 [49], CeO2 [50], Co3O4 [4547], MgO [51], In2O3 [52, 53], Fe3O4 [54], and WO3−x [55]. Schüth et al. [56, 57] reported on the hard-template synthesis of nanostructured porous Co3O4 using two-dimensional (2D) hexagonal SBA-15 and three-dimensional (3D) cubic KIT-6 silica as templates. We also used mesoporous silica monoliths as a hard-template for the synthesis of crystalline tungsten oxide, and the data are reported in Figure 5. Highly ordered mesoporous silica monoliths were synthesized by an instant direct-templating method. Figure 5 shows the TEM images of mesoporous silica template and tungsten oxide obtained by hard-template. The synthesis involved the dispersion of mesoporous silica templates in ethanol solution of tungsten hexachloride, with stirring for 1 h at room temperature, followed by evaporation of the ethanol, and calcination at high temperature. These processes were repeated twice to enhance the loading amount of tungsten precursors. The silica template was removed by leaching with sodium hydroxide aqueous solution. The obtained mesoporous WO3 samples have specific surface area of 70–153 m2/g and pore sizes of 3–7 nm. As demonstrated, the hard-template method is effective for the synthesis of crystalline ordered mesoporous metal oxides. However, hard-template synthesis is both costly and time-consuming because the synthesis requires long preparation and multiple processes, including the fabrication of porous templates, the incubation of metal precursors in the nanopores of templates, conversion of metal precursors into metal oxides, and selectively etching the templates [35, 4357]. In addition, the method has some drawbacks, which include the difficulty in loading the amount of metal oxides into the pore channels of mesoporous silica because of possible blockage by the metal oxides at the mouths of the pore channels during the incubation and/or filling into the pore channels of mesoporous silica, which prevents the further filling process. In turn, the covering of the pore channels led to the coating of metal oxides on the outside surface of the templates particles as the bulk dense samples, but not on the mesoporous structures. Furthermore, the hard-template method also faces challenges in the selective etching of the templates, specifically when the desired synthesis materials can interact with the templates to form other phases; in addition, oxide phase and/or the desired synthesis materials can easily be dissolved in etching solution.

2.3. Template-Free Synthesis of Meso-/Nanoporous Metal Oxides

Synthesis of porous metal oxides without using templates has gained interests in recent years. Several methods such as sol-gel process [58], hydrothermal hot-press [59], anodic anodization [60], and electrodeposition [61] have also been developed for the synthesis of porous metal oxides. The anodic anodization method enables the synthesis of highly ordered mesoporous metal oxides, such as Al2O3 and TiO2, but the products are usually of amorphous phases. In addition, the ordered Al2O3 fabricated by anodic anodization method is not suitable for the resistive gas sensor application because of its insulating behaviors. The template-free methods exhibited advantages by being simple, inexpensive, and scalable technique for the synthesis of meso-/nanoporous metal oxides [62]. Figure 6 shows a diagram process for the template-free synthesis of meso-/nanoporous metal. The synthesis generally involves the fabrication of intermediate phase of metal oxides, such as metal hydroxide, or metal carbonate and the conversion of intermediate phases into meso-/nanoporous metal oxides [63]. In our recent work, we introduced the synthesis of mesoporous NiO nanosheets by template-free method [64]. This material showed good sensing characteristics to highly toxic NO2 gas and potentially for large scale fabrication of sensors.

Different meso-/nanoporous metal oxides were synthesized successfully by using template-free method. Figure 7 shows the SEM and TEM images of mesoporous NiO nanosheets synthesized by template-free hydrothermal method. First, the hexagonal Ni(OH)2 nanosheets were prepared by hydrothermal method without using any surfactant or structure-directing agent. Thereafter, the Ni(OH)2 was converted into mesoporous NiO nanosheets by thermal oxidation [64]. The mesoporous NiO nanosheets are highly crystalline with tunable pore size just by simply varying the synthesis conditions. The hydrothermal method could also fabricate other mesoporous metal oxides such as In2O3 [65], Fe2O3 [66], and Co3O4 [63, 67]. The mesoporous metal oxides are of excellent materials for application in gas sensing fields [6874].

2.4. Meso-/Nanoporous Composite of Semiconducting Metal Oxide and Silica

The use of mesoporous metal oxide composites for gas sensor applications has gained interest because of its enhanced sensitivity and selectivity [75]. The most popular mesoporous metal oxide composites used for gas sensors application are based on mesoporous silica because of the easy synthesis and control of the mesoporous structures. In addition, the incorporation of metal oxide in the mesoporous silica leads to an increase in the sensitivity and selectivity of materials [75, 76]. Recently, we reported on the direct-synthesis of high ordered silica and metal oxide nanocomposites (HOM/MO) for gas sensor application. The synthesis involved an instant, one-pot, and direct-template method using Brij 56 (C16EO10) surfactant as soft-template. In typical synthesis, 0.815 g Brij 56 was dissolved in 1.63 g (~0.013 mol) tetramethoxysilane (TMOS) in a round balloon flask (300 mL in volume) and agitated at 60°C in a water bath for ~1 min to obtain a well-homogenized mixture. Subsequently, the predissolved metal chloride (SnCl22H2O, ZnCl2, NiCl2, CuCl22H2O, or FeCl24H2O) in 0.815 g acidified aqueous solution HCl/H2O (pH 1.3) was added to the mixture. The exothermic hydrolysis and condensation of TMOS occurred rapidly. The samples were dried under vacuum using a rotary evaporator to obtain a gel-like material at 45°C for ~5 min. The mass ratio of Brij56 : TMOS : HCl/H2O was 1 : 2 : 1. The amount of adding metal chloride was calculated according to the atomic ratio of metal to silicon ( = M/Si), varying from 0.11 to 4.00. The Brij56 soft-template was removed by calcination at 500°C for 8 h to obtain the HOM/MO nanocomposite monoliths [77]. Figure 8(a) shows the SEM, STEM, and TEM images of the HOM/SnO2 nanocomposites. Its particles are of nanometer to micrometer dimensions. The EDS results indicate the presence of C, O, Sn, and Si (Figure 8(a), inset); however, there was no detectable signal of the Cl in the EDS spectra of MO/HOM. The presence of C was due to contamination, whereas O, Sn, and Si originated from the sample. The atomic composition was 1.0% C, 66.0% O, 13.1% Sn, and 19.9% Si, which is consistent with the calculation from the precursors (). Figure 8(b) shows a bright field STEM image of an HOM/SnO2 nanocomposite monolith with a size of ~350 nm. The SnO2 nanocrystals (dark dots, 5 nm average size) were distributed homogenously in the matrix of the mesoporous silica. No aggregation of SnO2 particles can be observed in the STEM image. The distribution of the nanocrystalline SnO2 in the matrix of mesoporous silica prevented the grain growth of nanocrystals and increased the long-term stability of sensors made from the nanocomposites. Figures 13(c) and 13(d) show the HRTEM images of the HOM/SnO2 () nanocomposites. Uniform cylindrical pore channels ran through the monolith (Figure 8(c)), which confirmed its high-order structure. The wall thickness and the pore size averaged 3.5 and 3.2 nm, respectively (Figure 8(d)). The HRTEM images also indicate the presence of SnO2 nanoparticles (dark dots), which had an average diameter of about 5 nm and were distributed homogenously in the matrix of the mesoporous silica (bright region). The higher-magnification HRTEM image (Figure 8(d), inset) of the dark dot confirmed that the SnO2 nanoparticle was a single crystal. The lattice fringes were clearly observed; the distance between each was found to be 0.33 nm, which corresponded to the (110) interplanar spacing of tetragonal SnO2.

3. Meso-/Nanoporous Metal Oxides for Gas Sensors

3.1. Design of Meso-/Nanoporous Metal Oxide-Based Gas Sensors

To improve the gas sensor performances by enhancing the total exposure volume of sensing materials to analytical gases, we recently designed the gas nanosensors, as shown in Figure 9. The gas nanosensors involved the integrated electrodes deposited on a thermally oxidized silicon substrate. To this integrated substrate, the meso-/nanoporous metal oxides with different geometrical designs of monoliths, mesocages, hollow spheres, nanosheets, nanorods, and nanowires were sprayed or screen printing deposited to act as sensing layers. The meso-/nanoporous metal oxide sensing materials have numerous meso-/nanopores and large specific BET surface area, which enable the analytical gas molecules to be easily adsorbed on the total volume of the sensing layers, resulting in rapid response time and supervisor sensitivity [71]. The selectivity of gas nanosensors can be improved by selecting proper sensing metal oxides and/or using the nanocomposites of metal oxides.

Figure 10 shows the SEM images of meso-/nanoporous metal oxide-based gas nanosensors fabricated by a thick film technique [38]. This technique enables the controlled fabrication of inexpensive and scalable gas nanosensors, whereas up to hundreds of gas sensing devices can be fabricated on a 4-inch silicon wafer. The interdigitated Pt/Ti electrodes were deposited onto a thermally oxidized silicon substrate by a sputtering system, using a conventional lithography technique. The electrode contained 18 pairs of fingers, each 800 μm long and 20 μm wide, respectively. The meso-/nanoporous metal oxides were deposited between and/or over the electrode fingers and acted as conducting and sensing layers for gas adsorption. Using this synthetic technique, ordered and disordered meso-/nanoporous metal oxides can be deposited, and no distortion of material structures occurred during processing, as revealed by the SEM images (inset, Figure 10). Details about the gas sensing characteristics of the devices were reported in [38].

3.2. Sensitivity of Meso-/Nanoporous Metal Oxide-Based Gas Sensors

Sensitivity is one of the most important parameters of the sensors in practical applications. Higher sensitivity makes the sensor better because it allows the detection of the lower concentration of analytic gas. We have used different meso-/nanoporous metal oxides for gas sensing applications. The meso-/nanoporous p-type Co3O4 nanorods were used for highly sensitive VOC gas sensor applications [67]. Different meso-/nanoporous metal oxide semiconductors can be used for different sensors. Figure 11 presents the response of the meso-/nanoporous WO3 to NO2; in the inset is the TEM image of the meso-/nanoporous WO3. The sensor’s response to 1, 2.5, and 5 ppm concentrations of NO2 gas at 150°C was 850%, 6903%, and 21155%, respectively. The values decreased to 123%, 966%, and 2893%, and 3%, 38%, and 136% at temperatures of 200 and 300°C, respectively. Comparing the sensor response obtained in this work with the values of the porous WO3 nanorods, or the solvothermally synthesized W18O49 nanorods, the mesoporous tungsten oxide nanoplates exhibited much higher sensitivity [38]. The high sensor response obtained in this study was possible because of the large specific surface area and the small crystalline size of materials; the larger surface area provides a larger sensing site for NO2 adsorption, thus enhancing the sensor’s response. The linear dependence on sensor’s response as a function of acetone concentration in measured range is one of its advantages in practical application because of the easy design in the readout signal circuit.

3.3. Selectivity of Meso-/Nanoporous Metal Oxide-Based Gas Sensors

Selectivity is another most important parameter in the practical application of gas sensors. Generally, the selectivity of metal oxide-based resistive gas sensors can be achieved by using some additive metals. Figure 12(a) shows the effects of the doping level in the HOM/SnO2 nanocomposites on their VOC sensing. The sensor’s response to benzene increased from 55% to 1,050% when the Sn doping content increased from 10% () to 80% (), whereas the response to ethanol appeared to be independent of the Sn content. Sensor’s response to acetone decreased with higher Sn content within 10% to 60% but further increased at 80% Sn [77]. The effect of Zn doping on the response of the HOM/ZnO monolith sensors is presented in Figure 12(b). The sensors still showed the highest response to acetone. However, the sample doped with 40% Zn () showed the lowest response to acetone, although it exhibited the highest response to benzene and ethanol compared with the other samples. From the experimental data, the enhancement of the sensor’s sensitivity to benzene is suggested; the use of high doping level of Sn or Zn is also preferred. However, in both cases, the sensors exhibited extreme selectivity for the detection of acetone, suggesting a method to improve the selectivity of the materials.

3.4. Stability of Meso-/Nanoporous Metal Oxide-Based Gas Sensors

In practical applications, good transient and long-term stability of the gas sensor is expected for the reusability of the devices. The stability of sensors is dependent on the thermal and chemical stability of the sensing materials upon gas sensing measurements. Generally, the gas sensors operate at high temperature of about 200–400°C. At such high operating temperature, the stability of the metal oxide-based sensors tends to decrease as a result of the grain growth during measurement. In addition, the chemical interaction between analytic gas and sensing materials to form a new phase destroys the stability of the sensors [29, 30]. The amorphous or polycrystalline metal oxides have poor stability because they are not thermally stable as a result of the crystallization and grain growth of materials during sensor operation. The crystallization and grain growth of materials lead to a shift of base line resistance and also decrease the sensitivity and stability of device. Oppositely, the high crystallinity of materials such as a single crystal is very stable under sensor operation condition and results in a better stability [78]. By using meso-/nanoporous metal oxides with high crystallinity, a gas sensor with very high stability can be fabricated. Figure 13 shows the stability of the single crystal meso-/nanoporous ZnO nanorod sensor after several cycles of exposure to NO2 and back to dry air [79]. The sensor showed very good stability, a stable signal, a high response, and high recovery. Sensor’s response did not decay after prolonged storage, even after eight cycles of switching on/off from dry air to gas and back to dry air. This is the result of the high crystallinity of the fabricated ZnO, where the lattice fringes can be seen clearly in the HRTEM image (inset of Figure 13) with an interplanar spacing of 0.52 nm. The HRTEM image also indicates that the ZnO nanorod is single crystalline and free of defects, with preferred growth direction along the -axis of the hexagonal ZnO (JCPDS, 36-1451) [80]. The high crystallinity of the synthesized meso-/nanoporous metal oxides prevented the grain growth during sensor operation at high temperature and enhanced the long-term stability of the sensors [78].

4. Conclusions and Outlook

We have briefly reviewed the recent research on the synthesis of mesoporous metal oxides for gas sensor applications. The general synthesis methods of soft-template, hard-template, and, specifically, template-free have been introduced for fabrication of meso-/nanoporous semiconducting metal oxides and their nanocomposites. The advantages and disadvantages of each synthesis method have been figured out. The mesoporous structures play important roles in the application of semiconducting metal oxides as gas sensors, in which the large specific surface areas with the ability to control the grain size, pore size, and pore architecture are of advantage for the enhancement of gas sensor performances. Specifically, the synthesis of nanomaterials, such as nanoparticles, nanorods, and nanowires contents of meso-/nanoporous structures, would be a great advantage for fabrication of advanced gas nanosensors for different applications. Furthermore, integration of low power consumption gas sensors into smart phones for human heath diagnostic is a perspective. The meso-/nanoporous metal oxides are advantageous to the limited application to resistive type gas sensors, but also for optical sensors, electrochemical sensors, and calorimetric sensors.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgment

This research was funded by the Vietnam National Foundation for Science and Technology Development (Nafosted, 103.02-2014.06).