1. Introduction
The release of these toxic gases (NO, SO2, CO, CO2, SO2, H2S, and NH3) from chemical industries, mines, and power plants has put a strain on the health of human beings, resulting in cardiovascular illnesses such as asthma, arrhythmia, and heart failure etc.), especially in pregnant women and elderly people. Moreover, these toxic gases and volatile gases have also contributed immensely to climate change resulting in extreme weather occurrences. The demand for highly efficient, reliable, and low-cost gas sensors has increased tremendously all over the world in recent years. With the advances of the fourth industrial revolution in bridging gaps between the field of science and technology, nanotechnology has been marked to be a major player among 4.0 emerging technologies. Nano gas sensors have gained extensive use in the medical field, to diagnose diseases and identify the nature of illness through monitoring the exhaled breath of patients. However, the most reliable and highly recommended technique for gas detection nanoparticle-based materials is to be considered, especially in the medical field for long-term health purposes.
The release of toxic gases (e.g., NO, SO₂, CO, CO₂, H₂S, and NH₃) from chemical industries, mines, and power plants poses a significant threat to human health, leading to cardiovascular illnesses such as asthma, arrhythmia, and heart failure, particularly among pregnant women and the elderly. Furthermore, these toxic and volatile gases contribute immensely to climate change and extreme weather events. Consequently, global demand for highly efficient, reliable, and low-cost gas sensors has increased tremendously in recent years.
Advancements linked to the Fourth Industrial Revolution, which bridges science and technology, have positioned nanotechnology as a major player among emerging Industry 4.0 technologies. Nano gas sensors are now extensively used in medicine to diagnose diseases and identify illnesses by monitoring patients' exhaled breath [
5,
6,
7,
8]. For reliable, long-term health monitoring, nanoparticle-based materials are considered a highly recommended technique for gas detection.
The use of metal oxide semiconductors (MOS) materials for the development of gas sensors can be dated back to 1954 and has continued to be one of the most effective techniques in monitoring the concentration of gases in the atmosphere. Nano MOS materials like SnO2, ZnO, WO3, V2O5, Fe2O3, and TiO2 are widely investigated for gas sensor applications because of their high surface to volume ratio, low manufacturing cost, and capability of detecting a large number of toxic gases at different temperature. The sensing mechanism of these MOS materials lies in the changes in electrical conductivity upon exposure to a concentration of gases due to catalytic reduction/oxidation reactions occurring at the oxide surface. In recent years, gas sensors based on 2-dimensional structures (2D)- such as vanadium pentoxide (V2O5), molybdenum disulphide (MoS2), tungsten disulphide (WSe2), hexagonal boron nitride (h-BN), Carbon materials, and graphene nanostructures have been attracting tremendous attention for gas sensor fabrication.
The use of metal oxide semiconductor (MOS) materials for gas sensor development dates back to 1954 and remains one of the most effective techniques for monitoring atmospheric gas concentrations. Nano-MOS materials such as SnO2, ZnO, WO3, V2O5, Fe2O3, and TiO2 are widely investigated for gas sensing due to their high surface-to-volume ratio, low manufacturing cost, and capability of detecting numerous toxic gases at different temperatures. The sensing mechanism of these materials relies on changes in electrical conductivity upon exposure to target gases, driven by catalytic reduction/oxidation reactions at the oxide surface. Recently, gas sensors based on two-dimensional (2D) structures—such as vanadium pentoxide (V₂O₅), molybdenum disulphide (MoS₂), tungsten disulphide (WSe₂), hexagonal boron nitride (h-BN), carbon materials, and graphene nanostructures—have attracted tremendous attention for sensor fabrication.
Among various 2D-based structures, V2O5 is an upcoming gas-sensing material because of its low optical bandgap of 2.5 eV, and excellent thermal and thermoelectric properties relative to other semiconducting metal oxides. This material is among the top 10 most researched oxides. Moreover, this material is considered the most stable among other vanadium oxide members such as vanadium monoxide (VO), vanadium dioxide (VO2), and vanadium sesquioxide (V2O3). V2O5 possesses different polymorphs namely α-V2O5 (orthorhombic), metastable β-V2O5 (tetragonal or monoclinic), γ-V2O5 (orthorhombic), and δ-V2O5 (monoclinic) with each being stable at different temperature and pressure. α-V2O5 is the most commonly used material for gas sensing applications compared to the other polymers. This material exhibits Mott Hubbard’s metal-to-insulator (MTI) at a given temperature (375°C). Its unique crystallographic unit cell structure, electronic properties, high surface area to volume ratios, and several reaction sites enable it to exhibit high selectivity and response towards different gases.
Among various 2D-based structures, V₂O₅ is an emerging gas-sensing material owing to its low optical bandgap (~2.5 eV) and excellent thermal and thermoelectric properties compared to other semiconducting metal oxides [
1,
2,
3]. It ranks among the top ten most researched oxides [
4] and is considered the most stable member of the vanadium oxide family, which includes vanadium monoxide (VO), vanadium dioxide (VO₂), and vanadium sesquioxide (V₂O₃) [
5]. V₂O₅ exhibits several polymorphs, including α-V₂O₅ (orthorhombic), metastable β-V₂O₅ (tetragonal or monoclinic), γ-V₂O₅ (orthorhombic), and δ-V₂O₅ (monoclinic), each stable under different temperature and pressure conditions [
6,
7]. The α-V₂O₅ phase is the most commonly used for gas sensing and exhibits a Mott–Hubbard metal-to-insulator transition (MTI) at approximately 375 °C [
8,
9]. Its unique crystallographic structure, electronic properties, high surface-area-to-volume ratio, and abundant reaction sites enable high selectivity and sensitivity toward various gases [
10,
11,
12,
13,
14].
Research has shown that different atomic arrangements or structures of semiconducting materials called hierarchical structures, often result in different chemical, optical-electronic, magnetic, etc. properties [
15,
16]. This has also been demonstrated in the case of gas and chemical sensing where hierarchical structures of materials exhibited different adsorption properties (surface areas, pores volumes, and pore diameters) which consequently yielded different sensitivity, response time, recovery time, and gas selective ability [
17,
18,
19]. For instance, the large surface area of WO
3 hierarchical structures has been said to provide more channels which not only make gas diffusion great but also allow more active sites for adsorbing gases, thus leading to high gas sensing performance [
19]. Higher gas response of SnO
2 nanoneedles with quicker response and recovery of the same material when in nanosheets assemble was reported [
20]. The high response property of the nanoneedles structures is attributed to the large surface area from the Brunauer-Emmett-Teller (BET) test. More adsorption and desorption sites were also reported for nanoneedle structures which allowed gas molecules to easily diffuse across the surface. Spacing among the nanosheets is large (compared to an open room) relative to nanoneedles which give high pores for quicker response and recovery. Other reports have also shown that nanowires are the most researched morphology with 40%, followed by nanobelts and nanorods with 20%, nanotubes with 20%, and nanobelts with 20% [
4].
Research has shown that hierarchical structures—different atomic arrangements or assemblies of semiconducting materials—can lead to distinct chemical, optoelectronic, and magnetic properties [
15,
16]. This principle also applies to gas and chemical sensing, where hierarchical morphologies influence adsorption properties (e.g., surface area, pore volume, pore diameter), thereby affecting sensitivity, response time, recovery time, and selectivity [
17,
18,
19]. For instance, the large surface area of WO₃ hierarchical structures provides more channels for gas diffusion and additional active sites for adsorption, leading to enhanced sensing performance [
19]. Similarly, SnO₂ nanoneedles have demonstrated higher gas response, while SnO₂ nanosheet assemblies enable quicker response and recovery [
20]. The improved performance of nanoneedles is attributed to their large BET surface area, which offers abundant adsorption/desorption sites and facilitates gas molecule diffusion on the materials ‘surface. In contrast, the wider spacing between nanosheets creates more open pores, promoting faster response and recovery. Reports indicate that nanowires are the most researched morphology (≈40%), followed by nanobelts, nanorods, and nanotubes [
4].
It has been reported that the hierarchical hollow structure of Zinc stannate (ZnSnO
3 ) composed of ultra-thin nanorods as building blocks has shown fast response and recovery capacities, and good repeatability in the detection of ethanol. This is related to the hollow structure having efficient surface area and surface accessibility [
21]. Due to these exceptional properties, ZnSnO
3 shows permeable surfaces that allow easy absorption of ethanol gas molecules [
22] resulting in a fast response. Zinc oxide (ZnO) nanorod is another unique material that exhibits a quicker response and recovery speed towards ethanol detection, as compared to nanosheets which display a larger ethanol response. The quicker response of ZnO nanorod is due to its better conductivity and lower potential barrier, while the larger response of nanosheets results from its higher specific surface area [
23]. Other studies have also reported that hierarchical and hollow Indium oxide (In
2O
3) material has shown higher response and recovery times as compared to its agglomerated counterparts (powder) in the detection of carbon monoxide (CO) [
24]. This is a result of its structural assembly which allows effective and rapid gas diffusion toward the entire sensing surface. Metal oxide nanostructures/Porous Silicon (PS) composites, such as PS/WO
3 [
25] and PS/ZnO [
26,
27] have been reported to show good sensing abilities towards different gases. PS/V
2O
5 nanorod composite material exhibits high response and good selectivity towards NO
2 at room temperature [
28]. The porous silicon provides a high surface-to-volume ratio in composites (PS/V
2O
5) while the nanorods have specific surface areas and dimensions comparable to Debye length. The heterojunction between the two helps the material in achieving good sensing performance [
28].
Hierarchical hollow structures also show promise. For example, zinc stannate (ZnSnO₃) composed of ultrathin nanorod building blocks exhibits fast response/recovery, good repeatability, and efficient ethanol detection due to its high surface area and accessibility [
21,
22]. Zinc oxide (ZnO) nanorods demonstrate quicker response and recovery toward ethanol compared to nanosheets, owing to better conductivity and a lower potential barrier, whereas nanosheets show a larger response due to their higher specific surface area [
23]. Hierarchical and hollow indium oxide (In₂O₃) structures exhibit higher response and faster recovery times than their agglomerated powder counterparts in carbon monoxide (CO) detection, a result of their assembly promoting effective and rapid gas diffusion [
24].
Metal oxide nanostructure/porous silicon (PS) composites, such as PS/WO₃ [
25] and PS/ZnO [
26,
27], have also demonstrated good sensing abilities. Notably, a PS/V₂O₅ nanorod composite exhibits high response and good selectivity toward NO₂ at room temperature [
28]. In such composites, porous silicon provides a high surface-to-volume ratio, while the nanorods offer specific surface areas and dimensions comparable to the Debye length. The heterojunction between the two materials further enhances sensing performance [
28].
Having established that the morphologies or structures of metal oxide nanomaterials have a relationship with the chemical, physical, and gas-sensing properties of a material, it is obvious that crucial information about the choice of gas-sensing materials can be obtained. It is then essential to critically review the structures of V2O5 for the benefit of exploring them for gas sensor development. Thus, this paper presents a systematic review of the sensing properties of V2O5 nanomaterials in hierarchical structure assembly.
Given the established relationship between morphology (hierarchical structure) and the chemical, physical, and gas-sensing properties of metal oxide nanomaterials, crucial insights can be gained for selecting optimal sensing materials. Therefore, a critical review of V₂O₅ nanostructures is essential to explore their full potential for gas sensor development. This paper presents a systematic review of the sensing properties of V₂O₅ nanomaterials with a focus on hierarchical structure assemblies.
Figure 1.
Summary of V2O5 morphologies with their possible detectable gases.
Figure 1.
Summary of V2O5 morphologies with their possible detectable gases.
3. Density Functional Theory (DFT) Perspective
This review article has so far intensively discussed the impact of hierarchical structures towards gas sensing using the experimental approach. However, to explore the full extent of the analyses, we need to consider the quantum mechanical aspect based on density functional theory (DFT) calculations. Previous studies have used DFT to investigate different properties of hierarchical structures, including band structure, density of states, elastic constants, absorption energy, and optical properties. For example, first principles calculations have been used to explore the energetics, electronic and geometric structures of zinc sulphide (ZnS) nanotubes, nanorods, nanosheets, and nanowires based on their thickness or diameter, and its effect on the band gap. Results revealed that ZnS nanowires and double-wired nanotube with higher thickness diameter were the most energetically favoured hierarchical structures, compared to single-wired ZnS nanotubes [
118]. The band gap size of these hierarchical structures was found to decrease with an increase in diameter, resulting in high conductivity. Other works also used DFT to estimate the band gap of copper oxide (CuO) thin films, which was found to be 1.66 eV, in agreement with experimental results [
119]. In addition, this tool has also been used to gain an atomic-level understanding of how hierarchical structures interact with adsorbate gas molecules. It was observed that ZnS nanotubes showed a high response to ammonia and phosphine, due to the structural orientation of the gas molecules. The sensitivity of a tube was found to be more favourable towards the chemisorption mode [
120].
This review has thus far intensively discussed the impact of hierarchical structures on gas sensing from an experimental perspective. To explore the full extent of the analysis, we must also consider the quantum mechanical aspect based on density functional theory (DFT) calculations. Previous studies have used DFT to investigate various properties of hierarchical structures, including band structure, density of states, elastic constants, adsorption energy, and optical properties. For example, first-principles calculations have been employed to explore the energetics and the electronic and geometric structures of zinc sulphide (ZnS) nanotubes, nanorods, nanosheets, and nanowires as a function of their thickness or diameter, and the consequent effect on band gap. Results revealed that ZnS nanowires and double-walled nanotubes with larger diameters were the most energetically favorable hierarchical structures compared to single-walled ZnS nanotubes [
118]. The band gap of these structures was found to decrease with increasing diameter, resulting in higher conductivity. DFT has also been used to estimate the band gap of copper oxide (CuO) thin films, yielding a value of 1.66 eV which agrees with experimental results [
119]. Furthermore, this computational tool provides an atomic-level understanding of how hierarchical structures interact with adsorbate gas molecules. For instance, ZnS nanotubes showed a high response to ammonia and phosphine due to the structural orientation of the gas molecules, with sensitivity found to be more favorable toward the chemisorption mode [
120].
Other studies have also employed the DFT to simulate stable surfaces of various hierarchical structures. Reports have demonstrated that (010) V
2O
5 nanobelts have a good response to the adsorption of ethanol gas. Atomic Mulliken population analysis revealed a transfer of 0.18e electrons from the ethanol gas into the nanobelt's conduction band, resulting in increased conductivity. Furthermore, the reaction's enthalpy change was reported to be −2.84 eV, signifying an exothermic reaction, which is in agreement with experimental findings [
121]. Similar effects were observed when (001) nanorods were loaded with NH
3 molecules, wherein the band gap of the nanorods reduced with increasing loading of NH
3 molecules, indicating electron transfer from NH
3 to the nanorod and corresponding shift of the Fermi level towards the conduction band [
122].
Other studies have employed DFT to simulate stable surfaces of various hierarchical structures. Reports demonstrate that the (010) surface of V₂O₅ nanobelts exhibits a strong response to ethanol adsorption. Atomic Mulliken population analysis revealed a charge transfer of 0.18 e from ethanol to the nanobelt's conduction band, increasing conductivity. The reaction enthalpy change was reported to be −2.84 eV, signifying an exothermic process consistent with experimental findings [
121]. A similar effect was observed when (001) nanorods were exposed to NH₃ molecules; the band gap decreased with increasing NH₃ loading, indicating electron transfer from NH₃ to the nanorod and a corresponding shift of the Fermi level toward the conduction band [
122].
Further, DFT has been found to play a significant role in determining the type of transition metal dopants likely to improve the adsorption energies and electronic properties of materials. Reports have shown that doping carbon nanotubes (CNT) with Sc, Ti, V and Cr can drastically alter the electronic properties, resulting in enhanced NH
3, PH
3 and AsH
3 adsorption energies. It was reported that doping CNT with Cr shows a band gap of 0.707 eV, resulting in higher adsorption energy to NH
3 in comparison to the other dopants [
123]. The natural bond orbital that indicates the charge transfer of NH
3 to the CNT was found to be 0.195e, indicating a strong covalent bond [
123]. Additionally, it has also been reported that intercalation of metal atoms such as Rhodium (Rh) into (001) phase of V
2O
5 reduces the band gap of the material from 2.2 eV to 0.3 eV [159]. Rh becomes the centre of attraction when the material is exposed to some gases such as CO, PH
3, and H
2S. In the case of CO exposure, the most stable adsorption is when the C atom is adsorbed on Rh, yielding an energy of -296 meV as compared to the pristine V
2O
5, which gives the adsorption energy of -137 meV. However, polar molecules such as SO
2, and CO
2 prefer to adsorb on the V atom, which is supported by charge contour plots analysis that shows polar molecules bond via the oxygen atom to the zone near the V and bridging O
2 atom [
123]. Thus, the pre-adsorption of Rh on the surface of (001) V
2O
5 results in a weaker interaction between (001) V
2O
5 and CO
2 [
123]
.
DFT also plays a significant role in screening transition metal dopants to improve adsorption energies and electronic properties. For example, doping carbon nanotubes (CNTs) with Sc, Ti, V, and Cr can drastically alter electronic properties, enhancing the adsorption energies for NH₃, PH₃, and AsH₃. Specifically, Cr-doped CNTs exhibit a band gap of 0.707 eV and higher adsorption energy for NH₃ compared to other dopants [
146]. The natural bond orbital analysis indicated a charge transfer of 0.195 e from NH₃ to the CNT, suggesting a strong covalent interaction [
123]. Additionally, intercalating metal atoms such as rhodium (Rh) into the (001) phase of V₂O₅ reduces the band gap from 2.2 eV to 0.3 eV [159]. Rh acts as an active centre when the material is exposed to gases such as CO, PH₃, and H₂S. For CO exposure, the most stable configuration involves adsorption of the C atom on Rh, with an adsorption energy of −296 meV, compared to −137 meV for pristine V₂O₅. In contrast, polar molecules such as SO₂ and CO₂ prefer to adsorb on V atoms, supported by charge contour plots showing that these molecules bond via an oxygen atom near the V and bridging O atoms [
123]. Thus, pre-adsorption of Rh on the (001) V₂O₅ surface weakens the interaction with CO₂ [
123].
This current study employed DFT to investigate the effect of doping vanadium pentoxide (V2O5) with transition metals including tungsten (W), copper (Cu), manganese (Mn), tin (Sn), and silver (Ag) to enhance its adsorption energy towards nitrogen dioxide (NO2). Incorporating foreign metals in the surface of V2O5 has been previously reported to improve the electromechical properties of the material. For instance, doping V2O5 with Ag and Cu have shown to improve the electric conductivity properties by two order magnitude. This dopant has been believed reduce the oxidation of vanadium ion state from V5+ to V4+ resulting in oxygen vacancies in the material which are more beneficial for gas adsorption, especially to reducing gases. As such in this present work we intensively reviewed the sensing performances of nano layered hierarchical V2O5 structures and comprehensively study the gas molecule adsorption properties of pristine and doped V2O5 from ab initio approach. The different transition metals were incorporated using the substitution methods.
In the current study, DFT was employed to investigate the effect of doping vanadium pentoxide (V₂O₅) with transition metals—tungsten (W), copper (Cu), manganese (Mn), tin (Sn), and silver (Ag)—to enhance its adsorption energy toward nitrogen dioxide (NO₂). Incorporating foreign metals into V₂O₅ has been previously reported to improve its electrochemical properties. For instance, doping with Ag and Cu can improve electrical conductivity by two orders of magnitude. These dopants are believed to reduce the oxidation state of vanadium from V⁵⁺ to V⁴⁺, creating oxygen vacancies that are beneficial for gas adsorption, particularly for reducing gases. Therefore, this work comprehensively reviews the sensing performance of nano-layered hierarchical V₂O₅ structures and provides a detailed ab initio study of gas molecule adsorption on pristine and doped V₂O₅. The different transition metals were incorporated using substitutional doping methods.
3.1. Density Functional Theory Study of Transition-Metal-Doped V₂O₅ for Enhanced Adsorption of NO₂
Bulk alpha vanadium pentoxide (α-V2O5) is classified as an orthorhombic crystal structure with a space group 59 (Pmmn). The basis (Figure 9a) consists of identical vanadium atoms and three nonequivalent oxygen atoms, namely: vanadyl oxygen (Ov), bridging oxygen (Ob), and Chain oxygen (Oc). This structure is composed of a distorted octahedra with one central vanadium atom bonded to six oxygen atoms (VO6). (See Figure 9b) Each VO6 is linked together at the edges by the chain (Oc) and at the corners by the bridging (Ob). The vanadyl (V=Ov) is double-bonded to a single vanadium atom, with a bond distance of 1.610 Å, while the bridging (V-Ob) is bonded to two adjacent vanadium atoms with a bond distance of 1.806 Å, and chain (V-Oc) atoms are bonded to three vanadium atoms with a bond distance of 1.88 Å, respectively.
Bulk alpha vanadium pentoxide (α-V₂O₅) has an orthorhombic crystal structure with space group Pmmn (No. 59). Its unit cell (Figure 9a) contains identical vanadium atoms and three nonequivalent oxygen atoms: vanadyl oxygen (Oᵥ), bridging oxygen (O_b), and chain oxygen (O_c). The structure consists of distorted VO₆ octahedra (Figure 9b), each with a central vanadium atom bonded to six oxygen atoms. Adjacent VO₆ units are linked at edges by chain oxygen (O_c) and at corners by bridging oxygen (O_b). The vanadyl oxygen (V=Oᵥ) is double-bonded to a single vanadium atom with a bond distance of 1.610 Å, while the bridging oxygen (V–O_b) bonds to two adjacent vanadium atoms at 1.806 Å, and the chain oxygen (V–O_c) bonds to three vanadium atoms at 1.88 Å.
Figure 2.
(a) Crystal structure of V2O5 and (b) Coordination of a vanadium and oxygen atom in V2O5. The red balls represent oxygen, and the grey represents the vanadium atom.
Figure 2.
(a) Crystal structure of V2O5 and (b) Coordination of a vanadium and oxygen atom in V2O5. The red balls represent oxygen, and the grey represents the vanadium atom.
In order to determine the ground state properties of V
2O
5, Full geometry optimisation of the V
2O
5 unit cell was performed using plane pseudopotential wave density functional theory calculations, as implemented in the CASTEP. Vanderbilt ultrasoft pseudopotentials were used to describe the valence-core interactions, while the Perdew-Burke-Ernzerhof generalized gradient approximation functional was used to approximate the electronic exchange correlation effects. A plane wave cut-off of 1000 eV and a k-point sampling of 4×14×11 was found to be sufficient to converge the bulk V
2O
5. The optimised lattice constants and the bond distances were compared with the experiment and literature as summarised in
Table 8. Our calculated lattice parameters and bond lengths are in good agreement with experimental and literature values.
To determine the ground-state properties of V₂O₅, full geometry optimisation of the unit cell was performed using plane-wave pseudopotential density functional theory as implemented in CASTEP. Vanderbilt ultrasoft pseudopotentials described the valence–core interactions, and the Perdew–Burke–Ernzerhof generalized gradient approximation functional was used for the exchange–correlation effects. A plane-wave cut-off energy of 1000 eV and a *k*-point mesh of 4 × 14 × 11 were sufficient to converge the bulk V₂O₅ properties. The optimised lattice constants and bond distances were compared with experimental and literature values, as summarised in
Table 8. Our calculated parameters show good agreement with previous data.
3.2. Adsorption of NO2 Molecule V2O5 (011) Surface
The adsorption of NO
2 molecule in this study was modeled using a V
2O
5 (011) surface supercell. The (011) plane is proven to be the most thermodynamically stable compared to other V
2O
5 such as (100), (200), and (400). V
2O
5 surface was constructed by cleaving the optimised bulk V
2O
5 in the (011) direction and building a 2×2 supercell slab of four repeat units. A vacuum space of 20 Å was allowed to ensure negligible interaction with image slabs. Adsorption of the NO
2 molecule was performed for both the clean and transition metals (Cu, Mn, Ag, Sn) doped V
2O
5 surfaces, with the metal atom substituting the atom located on the top layer of the surface slab, as shown in
Figure 3. The adsorption energies in each case were calculated using the equation
The adsorption of NO₂ was modeled on a V₂O₅ (011) surface supercell, chosen for its thermodynamic stability relative to other surfaces such as (100), (200), and (400) [reference]. The surface was constructed by cleaving the optimised bulk V₂O₅ along the (011) direction to create a 2 × 2 supercell slab of four repeat units. A vacuum space of 20 Å was added to avoid interactions between periodic images. Adsorption was studied on both pristine and transition-metal-doped (Cu, Mn, Ag, Sn) V₂O₅ surfaces, with the dopant atom substituting a top-layer vanadium atom (
Figure 3). The adsorption energy (Eₐ) for each case was calculated using the equation:
where
is the total energy of the optimised (undoped or doped)V
2O
5 adsorbed surfaces whereas
and
denote the total energy of the doped and clean surfaces and energy of the NO
2 molecule, respectively. A negative value of
indicates that the process is exothermic, whereas a positive value indicates an endothermic process.
Where is the total energy of the optimised (doped or undoped) V₂O₅ surface with adsorbed NO₂, is the total energy of the clean (doped or undoped) surface, and is the energy of an isolated NO₂ molecule. A negative indicates an exothermic process, while a positive value indicates an endothermic one.
Figure 4 presents the adsorption energy of NO
2 on clean and transition metal-doped V
2O
5 (011) surface calculated for different numbers of NO
2 molecules. It is seen that all the calculated adsorption energies are negative, indicating that the reaction between the surface and the adsorbate occurs spontaneously. Increasing the load of NO
2 molecules leads to a reduction in the adsorption energy in both the clean and doped V
2O
5 surface, but while remaining in the negative regime, indicating a possibility of sensing application over a wide range of concentrations. Clearly, Ag-doped V
2O
5 presents the most negative adsorption energy as compared to the clean and the other dopants (see
Figure 5). However, at concentrations beyond four NO
2 molecules, the adsorption energy of the clean surface becomes more negative, indicating that Ag doping may play a critical role in increasing the sensing effectiveness of V
2O
5 at low concentrations of NO
2 molecules.
Figure 4 presents the adsorption energies of NO₂ on pristine and transition-metal-doped V₂O₅ (011) surfaces for varying numbers of NO₂ molecules. All calculated adsorption energies are negative, indicating spontaneous adsorption. Increasing the NO₂ load reduces the adsorption energy (though remaining negative) for both pristine and doped surfaces, suggesting potential for sensing over a wide concentration range. Notably, Ag-doped V₂O₅ exhibits the most negative adsorption energy compared to the pristine surface and other dopants (see
Figure 5). However, beyond four NO₂ molecules, the adsorption energy of the pristine surface becomes more negative, indicating that Ag doping may be particularly effective in enhancing V₂O₅ sensitivity at low NO₂ concentrations.
Similar results have been reported experimentally when comparing Ag
0.35 V
2O
5 with undoped V
2O
5. The Ag
0.35 V
2O
5 material tends to show high sensitivity and selectivity toward different gases (including ammonia, acetone, and amines) at 100 ppm concentration. This is due to the decrease in electronegativity of vanadium (V) in Ag
0.35 V
2O
5, which results in the high absorbing site [
126]. The other dopants, such as Cu, Ag, and Mn, have also enhanced the structural stability of V
2O
5 at higher concentrations of NO
2 molecules. The graph tends to increase beyond the saturation phase (four NO
2 molecules) on the undoped surface.
Similar results have been reported experimentally for Ag₀.₃₅V₂O₅, which shows higher sensitivity and selectivity toward gases such as ammonia, acetone, and amines at 100 ppm concentration. This is attributed to decreased electronegativity of vanadium in Ag₀.₃₅V₂O₅, leading to more favorable adsorption sites [
126]. Other dopants (Cu, Mn, Sn) also enhance the structural stability of V₂O₅ at higher NO₂ concentrations, as the adsorption energy trend rises beyond the saturation point observed for the undoped surface.