Sensitivity of Nanostructured Mn-Doped Cobalt Oxide Films for Gas Sensor Application

The effect of manganese doped cobalt oxide (Co3O4:Mn) was investigated by two different ratios (1% and 3%), which were precipitated by spray pyrolysis technique (SPT), and was adopted using a laboratory designed glass atomizer. Glass substrates were used to deposit films on them, heated at a temperature of 420 °C. The structural properties were studied through X-ray diffraction. The results showed that all deposit nanostructured films were polycrystalline and there was a decrease in the preferred reflection intensity along (311) plane resulting in a decrease in the crystallite size. Surface properties were analyzed through atomic force microscopy (AFM), which showed a decrease in the roughness and the particle size growth was a vertical columnar rod. The optical characterization displayed that the transmittance of pure Co3O4 nanostructured films was 48% and decreased to 35% for 1% of the Mn concentration, and continued to decrease to 33% with the increase of manganese concentration up to 3%. Optical energy bandgap of pure Co3O4 nanostructured films was 1.435 eV and decreased to 1.419 eV for 1% of Mn concentration, and continued to decrease to 1.367 eV with the increase of Mn concentration up to 3%. The highest percentage sensitivity was for the sample doped with 3% Mn, which was about 65%, for NO2 gas concentration of 600 ppm, at an operating temperature of 200 °C.

Recently the consideration regarding transition metal oxides has been taken into account. Metal oxide nanostructures play an essential role in numerous locales of science, physical science, ecological science, and material science [8][9][10]. These TMO, while falling in the nano estimated administration, are relied upon to have significantly witnessed an increasing interest in the fields; for example, information stockpiling, spintronics biomedicine, and broadcast communications [11][12][13][14]. These features are firmly subject to their dimension and surface properties. The Co 3 O 4 shows an ordinary spinal structure, in which Co 2+ is engaged in tetrahedral destinations and Co 3+ is engaged in octahedral locales [15]. This is very attractive because of the twists of Co 2+ particles, with a little commitment from turn circle coupling. Then again, Co 3+ particles have no perpetual attractive point as a result of the part of 3D levels of the octahedral precious stone field and finish filling of levels [16]. Given the attractive properties of permeable Co 3 O 4 small scale 3D squares, the antiferromagnetic Co 3 O 4 displays a specific level of ferromagnetism under outside attractive field because of the unpaired particles' closeness to the surface of the nanocrystals and the association of molecules. The unique properties of nanoparticle unequivocally reliant change of shapes and sizes and crystallization, charge heading [17]. Recently, a few procedures have been utilized to enhance the conduct of metal oxides by methods for presenting different dopants [18], shaping nanocomposites with p-n intersection [19], and by tuning the morphology [7], have focused on the electrical and surface properties of nanoparticles to accomplish enhanced reactant, electro-optical, attractive, substance, and physical properties [20]. Co 3 O 4 :Mn produced by solgel spin coat method has polycrystalline nature with degradation in the crystallization with increased manganese concentration, and the output phase is cubic with the appearance of orthorhombic MnO 2 phase. Shan, and Dai [21] explained that the doping of cobalt by magnesium causes a decrease in particle size and most particles had a nano-spherical shape and the size were between 82-244 nm. In this research Mn doped cobalt oxide is prepared to study their physical properties to use these films as a gas sensor [22][23][24]. The PLAL synthesized MoS 2 NS explained the vast absorbance in Vis region. Photocatalytic activity of TiO+-ZSM 5 -MoS 2 nanocomposite was examined with arsenite. TiO 2 -ZSM 5 -MoS 2 nanocomposite displayed nearly 100% arsenic photo-conversion to Usenet. This work aims to consider the effect of deformation on the sensitivity of the deposited films for NO 2 gas.

Experimental
An aqueous solution of 0.1 M of CoCl 2 .6H 2 O supplied from Merck Chemicals (Germany) was used to prepare thin films of Co 3 O 4 [25], and 0.1 M of Manganese chloride MnCl 2 ·4H 2 O was used as doping agent with a concentration of 1% and 3% to obtain Mn doped cobalt chloride [26]. A few drops of HCl were added to obtain a homogeneous solution. The spray pyrolysis technique was adopted with an experimental intended glass spray [27]. Glass substrate was used for deposition heated at a temperature of 420 °C. The following optimal status was taken into account, with spray time of 8 s. The spray was sprayed with two limits of 1.5 min to avoid excessive cooling and to prevent cracks. Nitrogen carrier gas was upheld at a pressure of 10 5 Nm -2 . The space between the nozzle and the substrate was 28±1 cm, at solution flow rate of 4 mL/min. Film thickness was dignified using gravimetric process and was found to be about 320 nm. Optical transmittance and absorbance employing UV-visible spectrophotometer were verified in the wavelengths range of 300-1100 nm (Shimadzu Company, Japan). High accuracy XRD was utilized to investigate the structural qualifications of thin films employing D8 Advance Bruker system, CuKα (λ = 0.154056 Å).

Results and Discussion
XRD spectra of pure and Mn:Co 3 O 4 nanoparticles are illustrated in Fig. 1(a), (b) and (c), from which the phase and crystalline structure were diagnosed. The samples were polycrystalline with a clear difference in intensity of reflections from crystalline levels for all diffraction peaks. The preferred reflection was (311) plane at 2θ = 36.92°, with several secondary reflections of (220), (222), (400), (511) and (440) at 2θ = 31.27°, 38.32°, 44°, 60.12° and 56.19° respectively. These results assured predominant of the cubic phase with no reflections for any phases concerning manganese. These peaks were in a perfect match with JCPDS data (JCPDS Card No. 43-1003) [28]. Note that the effect doping concentrations increasing up to 3% did not change the phase of cobalt oxide, and these were the most important conditions of doping, as well as through the alteration in intensity of scattering for the peaks produced. The doping of ions with a smaller ionic radius Mn 2+ (0.80 Å) than the host material Co 2+ (0.72 Å) led to a change in the values of lattice constant which could be followed by a deviation in the angle of reflection of the preferred plane towards the smaller angle. A noticeable decrease in the intensity of the preferred reflection (311) plane as shown in Fig. 1(d), (e) and (f) indicated an increase in the lattice constant.  (1) [29] showed a decrease in crystallite size (D) via an increase in doping ratio due to interstitial doping processes. The results are shown in Table 1.
where λ is x-ray wavelength, θ is Bragg's angle and K is the shape factor.
The Microstrain was calculated from Equation (2) [30,31], resulting from the growth of the film, which in turn produced pressure and expansion in the (  lattice. Varying the displacement of the atom based on their original location increased the microstrain ε. Fig. 2 shows the increase in doping ratio leading to an increase in microstrain, which was caused by the decrease in crystallinity of the film. Equation (3) was used to calculate the dislocation density δ; Fig. 2 displays the dislocation density increasing as Mn concentration increased, which agreed with the results obtained in previous studies [32,33].
The results of ε and δ values are listed in Table 1.
Topography of pure and Mn-doped Co 3 O 4 films was calculated by AFM (Fig. 3). The roughness, RMS and average grain size were also collected from 2D images.  The AFM images of the pure Co 3 O 4 showed that the particle size was regular in shape with a maximum RMS roughness of 0.431 nm. RMS roughness acts as the standard deviation, which is important to describe surface roughness by statistical methods. As the Mn dopant increased from 1% to 3% (Fig. 3(b) and (c)), the columnar growth enclosed grains with a rod profile could be affected by the merger of Mn on the developing surface. The spreading of the crystallites on surface morphology was uniform with decreases in the RMS from 1.36 nm to 1.23 nm as the Mn concentration increased from 1% to 3%. Apart from that, the average diameter decreased from 86 nm to 76 nm due to decomposition in the lattice.
The optical transmittance of undoped and Mn doped Co 3 O 4 films was measured by UV-Vis spectrophotometer. The first beam passed through the undoubted substrate while the second beam passed through the thin film. The transmittance depended strongly on Mn concentration as shown in Fig. 4. The transmittance of the undoped Co 3 O 4 film approached 48% in the near-IR spectrum. Fig. 4 demonstrated lower transmittance of 38% to 33% with the increase of doping concentration from 1% to 3%, respectively. The decrease in surface roughness with increasing doping concentration is shown in Fig. 4(a)-(c). Fig. 4(d) represents the absorption values that were observed to increase with increasing doping concentration that could be related to the doped thin films that could absorb light in a broader range of wavelengths and operate extra light energy than Co 3 O 4 thin film. The absorption spectrum took exponential decay with increasing wavelength due to the low energy of the falling photons and their inability to lift electrons from the valence band to the conduction band ( Fig. 4(d)).
The values of the energy gap depended on the precise structure of the deposition membrane on the arrangement and distribution of atoms in the crystalline bond [34]. Cobalt oxide compounds have a direct and indirect energy gap [35] and may vary depending on the type of deposition or conditions. Fig. 5(a)-(c) shows the graph of (αhν) 2 vs E (eV) for direct energy bandgap of pure and Mn-doped Co 3 O 4 thin film. The value obtained for the energy gap of the cobalt oxide of 1.435 eV differed from the standard value of 1.419 eV [36]. This difference in the value of energy gap could be achieved by deposition mechanisms or conditions accompanying this deposition. Fig. 5(a) and (b) shows a decrease in the energy gap values from 1.419 eV to 1.367 eV with the increase of Mn concentration from 1% to 3%, respectively. This might be due to a decrease in crystallite size values calculated from the X-ray diffraction pattern and the atomic force microscope, which led to crystallization weakness [37].
Absorption coefficient (α) can be defined through Equation (4). α decreased with increases of λ where t is the film thickness.
The refractive index (n) is shown in Fig. 6, calculated from reflectance (R) via Equation (5) [40]. The study of refractive index obtained a clear picture of the properties and visual behavior of the films recorded for the pure film's refractive index values of 2.6 which were taken down to 2.4 then to 2.3 with an increase in dopant concentration from 1% to 3%, respectively, due to the increased density of the medium, besides the scattering of the incident light at the grain boundaries [41]. Fig. 7 represents the dynamic resistance change as a function of time for several NO 2 gas concentrations (200, 400, 600, and 800 ppm) at operating temperature of 200 °C for Co 3 O 4 and Mn doped Co 3 O 4 . It was noted that the resistance values were limited to 5-45 kΩ.m; the greatest resistance was recorded in the presence of air R a for the undoped samples; the resistance values decreased with the doping ratio due to increasing number of charge carriers. The resistance of the sample decreased with increasing gas concentration from 200 to 800 ppm, which can be due to surface roughness. All recorded samples had a negative connection feature, a decrease in resistance with an increase in concentration of NO 2 gas. When O 2 molecules are adsorbed on the surface of metal oxides (Fig. 7(b)), which leads to the capture of electrons from the conduction band (E c ), the electrons are held on the surface of the films where R a is resistance of the film sensor in the presence of air, and R g resistance of the film in NO 2 gas. Fig. 9(a) represents the response time as a function of gas concentration of NO 2 . It was observed the minimum response time was in the 3% doping concentration, and at the concentration of the gas 800 ppm, it was about 4 s. (Reducing time is a good feature of sensors.) In general, the response time decreased with the increase of gas concentration, which could be due to the increased charge carrier concentration by adsorption and absorption of the gas to the surface. Fig. 9(b) represents the recovery time or return time as a function of NO 2 gas concentrations, where the minimum recovery time of the 1% model was equal to 10 s (Reducing time is a good feature of sensors). This occurred due to the deforming of films at 150 °C. Continuous exposure to operating temperature caused the film properties to change with the change of gas sensing properties [43].

Conclusions
Undoped and Mn-doped Co 3 O 4 were produced by the SPT. The structural description of the samples studied by XRD displayed polycrystalline with cubic structure; the crystallite size reduced from 15.62 nm for the pure Co 3 O 4 to 11.2 nm upon 3% Mn-doped Co 3 O 4 samples. The morphology investigation by AFM showed a columnar growth containing grains with a rod shape, which can be caused by a combination of Mn on the developing surface. The incorporation of Mn into Co 3 O 4 weakened the energy band due to the formation of surface-associated defects in the nanoparticles. We concluded that the electrical resistance of the films were reduced from 45 kΩ to 18 kΩ with the increasing of doping rate, and that the best sensitivity to the gas NO 2 was 65% by 3% Mn content. With the concentration of gas 800 ppm at a temperature of 150 °C, the best recovery time was 10 s, and the best response time was 4 s to the doped film by 3%, with the concentration of gas 800 ppm at a temperature of 150 °C.