Current Issue

The simplest nanostructure, a 0-D nanoparticle, is the most widely used nanostructure for hematite gas sensors, and sensors based on 0-D nanomaterials have the advantage of abundant active sites and high surface-to-volume ratios. Hjiri et al. reported a NO₂-selective sensor based on hematite nanoparticles synthesized via the hydrothermal method. For the synthesis, a solution was prepared by mixing FeCl₃ (0.1 M) with ammonia (2 mL) and heated at 200 ^oC for 10 h in a steel autoclave. The resultant material was washed several times and dried at 60 ^oC for 1 h. In the final step, the obtained powder was annealed for 2 h at 500 ^oC. The synthesized powder was then analyzed using the X-ray diffraction technique (XRD). The diffraction peaks in Fig. 3(a) show only hematite peaks, indicating that the annealing had completely changed β-FeOOH to the hematite phase. Fig. 3(b) shows the scanning electron microscopy (SEM) images of the hematite nanoparticles. As revealed in several researches [13,16,17], hematite powders prepared via the hydrothermal method are 1.5–2.5 μm in diameter owing to the agglomeration of smaller nanoparticles. Fig. 3(c) shows the transmission electron microscopy (TEM) image of the hematite nanoparticles, which shows hexagonal nanoparticles with an average crystalline size of approximately 50 nm. The high-resolution transmission electron microscopy (HRTEM) image in Fig. 3(d) indicates the high crystallinity of the grains. The fabricated hematite nanoparticles were placed on an alumina substrate to measure their gas-sensing performance. To optimize the operational temperature, response measurements to 5 ppm NO₂ were obtained for temperatures from 150 ^oC to 300 ^oC, as shown in Fig. 3(e). The hematite nanoparticles showed the best response (3.4) at 200 ^oC. For evaluating the selectivity, the authors exposed the specimen to three different gases: NO₂, CO, and CO₂. The bar chart in Fig. 3(f) shows the response values to the target gases, in addition to those toward NH₃ and methane. The response to NO₂ was higher than that to other gases because of the high affinity of NO₂ for hematite [18]. Fig. 3(g) displays the dynamic response values of hematite nanoparticles to NO₂ (in 1–5 ppm concentrations). In the measured concentration range, the hematite nanoparticles showed linear sensitivity. The response and recovery times of the hematite nanoparticles are presented in Fig. 3(h). The response time reduced from 72 s to 10 s with an increase in the target gas concentration from 1 ppm to 5 ppm. However, the recovery time increased from 8 s to 180 s in the same concentration range owing to gas adsorption kinetics [18]. This work clearly demonstrated that the gas-sensing performance of hematite nanoparticles synthesized via the hydrothermal method ensures NO₂ selectivity.

Fig. 3. 
(a) XRD pattern, (b) SEM image, (c) TEM image, (d) HRTEM image of the hematite nanoparticles, (e) Response values for 5 ppm NO₂, (f) Response values to NO₂, CO₂, CO, NH₃, and CH₄ at 200 ^oC, (g) Response values to NO₂ for 1–5 ppm concentrations at 200 ^oC, (h) Response and recovery time values of the hematite nanoparticles. Reprinted with permission from [8]. Copyright ⓒ 2019, MDPI.

1-D nanostructures include nanorods, nanowires, nanofibers, and nanotubes. Various methods such as thermal oxidation, e-beam evaporation, chemical vapor deposition, and anodic oxidation are used to synthesize these nanostructures [9,19–22]. Compared to 0-D nanostructures, 1-D nanostructures show better transduction of the surface phenomenon into the electrical resistance change of the sensor. Kim et al. reported a vertically ordered hematite nanotube array using the anodic oxidation method. An Fe film was deposited on a patterned Pt electrode-SiO₂/Si substrate of 500 nm thickness using an electron beam evaporator. After the deposition, anodization was conducted in an ethylene glycol solution containing ammonium fluoride and water with a Pt cathode and Fe film anode, as shown in Fig. 4(a). Anodization for 3 min resulted in films composed of 1.2 μm hematite nanotubes. Finally, the amorphous hematite nanotube layer was annealed at 550 ^oC for 5 h and turned into a 1 μm-thick crystalline hematite film. Fig. 4(b) shows the SEM images of the hematite nanotube array. The cross-sectional image confirms that the Fe film was completely converted into a hematite nanotube array. Moreover, the XRD patterns display no metallic Fe peak for the as-anodized nanotube array, as visible in Fig. 4(d). Furthermore, the hematite nanotube array after annealing shows sharp hematite phase peaks without secondary phases. The TEM images in Fig. 4(c) clearly show that the vertically ordered nanotube array has uniform diameters. The selected-area electron diffraction (SAED) pattern shows the polycrystalline characteristics of the nanotube array after annealing. After the fabrication procedure, the dense, planar hematite film and hematite nanotube array were exposed to 50 ppm acetone at 350 °C. Both specimens responded rapidly to acetone with a decrease in resistance. However, the hematite nanotube array showed an approximately 10 times higher response compared to the dense, planar hematite film (Fig. 4(e)). To test the selectivity of the hematite nanotube arrays to acetone, responses to various target gases were measured, including ethanol, CO, H₂, C₆H₆, C₇H₈, NH₃, and CH₄. Evidently, the hematite nanotube array showed the highest sensitivity to acetone (84) [23,24]. The response ratio of acetone to the second most sensitive gas, ethanol, i.e., S_acetone/S_ethanol, was higher than 2.1. The responses of the dense-planar hematite film to these target gases were also measured, and the results are plotted in Fig. 4(f). Compared to reference dense, planar hematite films, the hematite nanotube array showed higher sensitivity and selectivity toward acetone gas. In addition to good sensitivity, the hematite nanotube array also exhibited a short response time of approximately 3 s in 90% responses, as shown in Fig. 4(g), which is much shorter than that demonstrated by the dense-planar hematite film (28 s). From Fig. 4(h), the responses of the hematite nanotube array to 1–5 ppm concentrations of acetone are 10.28, 13.21, 15.6, 17.7, and 20.1, respectively. From these response values, the theoretical detection limit for acetone is calculated to be approximately 1.71 ppb. Furthermore, according to Fig. 4(i), the response values for hematite nanorods indicate long-term stability, with sustained response values for 48 days. These results suggest that the introduction of 1-D nanotube structures promises extremely high sensitivity, great selectivity, and long-term stability to hematite-based gas sensors.

Fig. 4. 
(a) Schematic showing the anodizing method to create a hematite nanotube array on Pt interdigitated electrode (IDE)-patterned SiO₂/Si substrate, (b) Plain-view and cross-sectional SEM images of a hematite nanotube array, (c) Cross-sectional TEM images, SAED pattern with indexing, HRTEM image of the wall thickness of a hematite nanotube, (d) Glancing angle XRD patterns of an Fe/Pt/SiO₂/Si structure before anodization and hematite nanotube arrays before and after annealing at 550 ^oC for 5 h, (e) Dynamic sensing transient curve of a dense, planar hematite film and hematite nanotube array; sensing transients to various gases for the hematite nanotube array; and response values of the dense, planar hematite film and hematite nanotube array to various gases at 350 ^oC, (f) Responses of the hematite nanotube array to 50 ppm acetone and ethanol, (g) Dynamic acetone sensing transient of hematite nanotube array at 350 ^oC. Inset: 90% response time (t_response) of the hematite nanotube array, (h) Theoretical detection limit (DL) of hematite nanotube array to acetone. Inset: response curve of the hematite nanotube array to 1−5 ppm acetone at 350 ^oC, (i) Changes in the response of hematite nanotube array to 50 ppm acetone for 48 days. Inset: response to 13 consecutive pulses of 10−50 ppm acetone after 48 days. Reprinted with permission from [9]. Copyright ⓒ 2014, American Chemical Society.

2-D nanostructures represented by graphene have a high aspect ratio; therefore, they have numerous surface sites to react with gaseous molecules. Ma et al. reported the gas-sensing properties and mechanism of network structures assembled from hematite nanosheets with exposed {104} facets. This work was based on previously reported research, which stated that hematite nanostructures with specific exposed facets, such as {012}, {014}, and {210} exhibited enhanced gas-sensing performance toward CH₃COCH₃ and C₂H₅OH [25]. In this work, the hematite network structures were synthesized by heating Fe(NO₃)₃ solution containing PVP at 650 °C in air. Fig. 5(a) shows SEM images of the products prepared using the synthesizing process. The network structures assembled from hematite nanosheets have a thickness of approximately 23–34 nm. The XRD diffraction peaks are indexed as hexagonal structures of hematite. The TEM images confirm that the hematite nanosheet is single crystalline and enclosed by {104} facets. To optimize the operating temperature, gas sensors based on the hematite nanosheet network structures were exposed to 100 ppm C₂H₅OH and N(C₂H₅)₃ at different temperatures. The maximum measured response was at 300 ^oC. The gas-sensing performance of the sensors based on the hematite nanosheet network structure was compared with that of sensors based on hematite commercial powders. After being exposed to various concentrations of C₂H₅OH and N(C₂H₅)₃, both sensors exhibited a short response time and good reversibility, as shown in Fig. 5(b). However, the response value curve of the two types of sensors to C₂H₅OH and N(C₂H₅)₃ revealed that the hematite nanosheet network structures display a higher response than hematite nanopowders to all measured concentrations. To explain the difference in the sensing properties, the sample’s response per unit surface area to 100 ppm of C₂H₅OH and N(C₂H₅)₃ was calculated and plotted, as shown in Fig. 5(c). The normalized response value of the hematite nanosheet network structure is evidently higher than that of hematite commercial powder, implying that hematite nanosheets with exposed {104} facets exhibit much better gas-sensing performance than hematite commercial powders. Furthermore, the XPS spectra of hematite nanosheet network structures and commercial powders revealed oxygen adsorption ability. The hematite nanosheet network structures had more chemisorbed oxygen species and O₂-ions in oxygen-deficient regions than hematite commercial powders, indicating a superior oxygen chemisorbing ability and a higher number of oxygen vacancies. These results suggest that controlling the exposed facets of hematite nanosheets results in extraordinary gas-sensing performance.

Fig. 5. 
(a) SEM images of hematite nanosheet network structure, (b) Response and recovery curve of the hematite nanosheet network structure and hematite commercial powder sensor, (c) Responses (per unit surface area) of hematite nanosheet network structure and hematite commercial powder sensor to 100 ppm C₂H₅OH and N(C₂H₅)₃. Reprinted with permission from [10]. Copyright ⓒ 2017, American Chemical Society.

3-D nanostructures exhibit the most complex and diverse morphology. Because of the possibility of heterojunctions, 3-D hierarchical heterostructures are being actively studied [26]. The different Fermi levels of two metal oxide semiconductors comprising heterostructures generates a potential barrier at their interfaces, which is an essential factor in enhancing the gas-sensing properties of sensors based on heterojunctions [27]. Suh et al. reported a selective toluene sensor based on hematite-decorated nickel oxide nanocorals. Fig. 6(a) shows a schematic of the preparation method for hematite-decorated nickel oxide nanocorals. Vertically ordered nickel oxide nanorods with hematite decoration were deposited on an SiO₂/Si substrate with Pt-interdigitated electrodes using the glancing angle deposition method based on electron beam evaporation. A 100 nm thick nickel oxide nanorod layer was deposited with a tilt angle of 83°, followed by a 3 nm-thick Fe layer deposition at a tilt angle of 0°. The deposition procedure was repeated until the total thickness reached approximately 600 nm. The nickel oxide and iron films were annealed at 550 ^oC for 2 h to oxidize iron into hematite. The SEM images in Fig. 6(b) show the typically smooth and extremely rough surfaces of the bare nickel oxide nanorods and hematite-decorated nickel oxide nanocorals, respectively. The repetitive Fe layer deposition between NiO nanorods caused an adhesion force change between the Fe layers and NiO, yielding a rough surface [28]. Energy dispersive X-ray spectroscopy (EDS) mapping (Fig. 6(c)) of hematite-decorated nickel oxide nanocorals revealed a uniform distribution of Ni and O elements and sparse distribution of the Fe element, which indicates that the Fe film was converted to nanoparticles. The authors manipulated the deposition angle of nickel oxide nanorods (75°, 80°, 83°, and 85°) to control their density (Fig. 6(d)). After being exposed to 50 ppm C₂H₅OH, the nickel oxide nanorods deposited on the substrate at a tilt angle of 83° exhibited the maximum response. For optimizing the gas-sensing performance further, the operating temperature and thickness of Fe film deposition were modified. The optimal conditions for ensuring a high sensing performance from hematite-decorated nickel oxide nanocorals were determined to be an operational temperature of 350 ^oC and Fe-film deposition thickness of 3 nm. Furthermore, the VOC selectivity at optimum conditions was measured from the responses of bare nickel oxide nanorods and hematite-decorated nickel oxide nanocorals to various gases, including C₂H₅OH, C₇H₈, CH₃CHO, CH₃COCH₃, C₆H₆, CO, NO₂, and H₂. As shown in Fig. 6(e), both specimens displayed the highest response to C₂H₅OH. However, there was a difference in the second highest response, with bare nickel oxide nanorods responding to CH₃CHO and hematite-decorated nickel oxide nanocorals to C₇H₈. Especially, the response ratio of hematite-decorated nickel oxide nanocorals to bare nickel oxide nanorods for C₇H₈ showed an extremely high value, which is sufficient to allow selective detection of C₇H₈. To study the selectivity more deeply, the authors investigated the detection results for three consecutive pulses of various gases based on Principal Component Analysis (PCA). Bare nickel oxide nanorods did not show any obvious selectivity because of the overlapping area of most gases. However, hematite-decorated nickel oxide nanocorals showed excellent selectivity and a clear distinction to various gases, especially C₇H₈. To investigate the possibility of practical application, the long-term stability and humidity dependence were also investigated. The response curve to eight consecutive pulses of 50 ppm C₇H₈ for 30000 s confirmed their stability. Humidity induced a sharp increase in the base resistance; however, the gas response of hematite-decorated nickel oxide nanocorals to C₇H₈ was maintained. The response values of hematite-decorated nickel oxide nanocorals were observed as 0.38, 0.48, 0.53, 0.55, and 0.59 to 1–5 ppm C₇H₈, respectively, at 350 °C, as shown in Fig. 6(f). From these response values, the slope of the concentration–response curve was calculated to be 101 ppb^-1, and the theoretical detection limit was calculated to be 22 ppb. Therefore, a suitably designed metal oxide heterostructure and fully decorated nanostructure can be effective for developing gas sensors. The comparative gas-sensing properties of gas sensors based on various hematite nanostructures are listed in Table 1.

Fig. 6. 
(a) Schematic of glancing angle deposition, self-shadowing effects when increasing the incident angle, and vertically ordered NiO nanorods without and with hematite decoration, (b) Cross-sectional SEM images of NiO nanorods without and with hematite decoration, (c) EDS mapping of Fe, O, and Ni for hematite-decorated NiO nanocorals. (d) Response transient curves of bare NiO nanorods according to incident angles, operating temperature, and Fe interlayer thickness, (e) Response and response-time values for bare NiO nanorods and hematite-decorated NiO nanocorals to various gases (5 ppm for NO₂ and 50 ppm for other gases), (f) Responses and calibration of response for hematite-decorated NiO nanocorals to 1–5 ppm C₇H₈ at 350 ^oC. Reprinted with permission from [11]. Copyright ⓒ 2017 WILEY-VCH Verlag GmbH & Co. KGaA.