First-principles calculations on the resistance and electronic properties of H₂ adsorption on a CoO–SnO₂ heterojunction surface

Abstract

Compared with pure metal oxides, heterojunctions greatly change the response to gas by the synergistic effect of the interface. In this work, density functional theory was used to reveal the adsorption performance of H₂ on the heterojunction under oxygen conditions. First, we determined the most reasonable heterojunction structure based on the adhesion work. According to the adsorption energy, the presence of SnO₂₍100)(I)/CoO(110)(II) made the adsorption of H₂ more stable. The DOS results showed that the resistance of the heterojunction increased with H₂ adsorption, following the same trend as that of CoO(110) with H₂ adsorption, although that of the heterojunction increased more. The electron density and electron density difference indicated that the heterojunction improved the reaction between H₂ and oxygen ions on CoO(110). However, the resistance of CoO(110)(II)/SnO₂(100)(II) increased after H₂ adsorption, contrary to the resistance change of SnO₂(100). Besides, the bonding energy between H₂ and the adsorption site became worse. The above results demonstrated that the presence of the heterojunction could indeed change the response trend and the adsorption behavior of H₂. Interestingly, the adsorption sites and effects of H₂ were different when two metal oxides were used as the substrate of the heterojunction, respectively.

---

1. Introduction

Semiconductor metal oxides as a material for chemical gas sensors play a significant role thanks to their abundance, low cost, and easy manufacture. Among these semiconductor metal oxide gas sensors, compounds made of adsorbed air oxygen ions and the target gas on the surface cause a change in the electric resistance of the materials. These semiconductor metal oxide gas devices have sensitivity to various toxic, flammable, and explosive gases, and are widely applied in various fields.^1–4

However, metal oxide gas devices suffer from the effects of baseline resistance drift and from poisoning interactions. These problems make it progressively more and more difficult for metal oxide surfaces to approach reactive gases.^5,6 SnO₂ semiconductors are among the best n-type metal oxide gas-sensing materials for gas-sensitive devices owing to their wide band gap, excellent chemical stability, and prominent sensitivity.⁷ However in practical applications, the gas sensitivity to H₂ still needs to be improved for the accurate identification of H₂. Surface modification^8–10 is the most commonly used approach to enhance the H₂ sensitivity.¹¹ However, the response effect is not obvious. Decorating H₂-response sieves^12–15 and polymers¹⁶ on the surface of gas-sensing materials has also been explored to enhance H₂ sensitivity. These are new innovation, but the accurate identification for H₂ needs to be improved. In addition, as one p-type metal oxide gas-sensing material, CoO has shown an affinity with oxygen and possesses multivalent characteristics.¹⁷ The pristine form of CoO is hardly used in the field of sensing due to its lower response than n-type semiconductors,¹⁸ though it has been reported that its gas-sensing can be significantly improved by introducing dopants,¹⁹ or by proper control of the morphology.^20,21 However, the accurate identification of gases still needs to be improved.

The idea of combining different metal oxide materials to form a heterojunction was recently proposed in order to further improve the important sensing characteristics of resistive-type gas sensors.^22,23 Generally, n-type materials have a higher Fermi energy level than that of p-type materials.²⁴ Hence, when n-type and p-type heterojunctions are fabricated, the electrons of n-type metal oxides with a higher energy will pass through the interface to the vacant lower-energy states until the Fermi energy level achieves a stable condition. The potential barrier, which is attributed to the difference in Fermi energy levels, will show an obvious change when a heterojunction is surrounded by different oxidizing or reducing gases. Furthermore, the changes will be observed as a high response in composite heterojunction gas sensors.²⁵ The synthesis of a heterojunction provides a significant method to incorporate different physical and chemical properties into one system.²⁶

A large number of studies have applied first-principle calculations on the various properties of heterojunctions,^27–30 and extensive research studies on the formation of heterojunctions have been published. Katoch et al.³¹ researched an n-p ZnO/CuO heterojunction as s nanofiber sensor to strengthen the primary ZnO nanofiber sensor sensing in H₂S. Similarly, Choi et al.³² reported the gas-sensing of H₂S was boosted in a CuO/SnO₂ heterojunction more than in the bulk of SnO₂ owing to the formation and disruption of the heterojunction. Ju et al.³³ proved that a NiO/SnO₂ heterojunction had a more rapid response to trimethylamine than pure SnO₂. Other research into metal oxide p–n heterojunction NiO/ZnO,³⁴ NiO/Fe₂O₃,³⁵ and NiO/SnO₂³⁶ have experimentally proved that the formation of a metal oxide heterojunction would obviously strengthen a sensor's gas sensitivity. The performances of metal oxide of Sn and Co can also be improved greatly by forming a p–n heterojunction.^37,38 Heterojunctions have been applied to strengthen the sensing properties of sensors based on SnO₂ through altering the influence at the synthesis interface.

However, explanations about the microstructure and theories about the mechanism of heterojunctions are obscure. Not only does this show a misfit, but also the structure and adhesion strength of the interface affect the effectiveness of promoting heterogeneous nucleation.^39,40 It is very difficult to observe structures and the adhesion strength of an interface or the adsorption of H₂ on a heterojunction surface through experimental methods. In this respect, an atomic level understanding about the H₂-sensing mechanism on SnO₂–CoO materials is crucial. Meanwhile, the gas-sensing characteristics, such as adsorption performance of metal oxide sensors, should be visualized for practical application.⁴¹ First-principles calculations have been employed to reveal the structures and adhesion strength of the interface at the atomic scale,^42–45 and the adsorption of gases on a single metal oxide calculated,^46–51 including the adsorption energy, adsorption distance, DOS, and electronic structure.

In this work, we studied the microstructure of SnO₂/CoO and CoO/SnO₂ heterojunctions by DFT, analyzing the transformation of the resistance, electronic structure, and other properties of H₂ adsorption on different sites of the heterojunction surface. Moreover, this work can provide theoretical guidance for the fabrication of heterojunction gas sensors in order to accurately identify H₂.

2. Methods

2.1 Calculation details

The Cambridge Serial Total Energy Package (CASTEP) Code was used for all the calculations based on density functional theory (DFT).^52–54 The Generalized Gradient Approximation (GGA) of the Perdew–Burke–Ernzerhof (PBE) scheme was employed to describe the exchange–correlation functional. Considering the stability of the system and for optimization of the calculation speed, the plane wave cut-of energy was selected as 400 eV. The Brillouin zone was sampled by 3 × 3 × 1 k-points using the Monkhorst–Pack scheme. The convergence tolerance for the energy was selected as 2.0 × 10⁻⁶ eV per atom. The force, stress, and displacement tolerance of the convergence tolerance were set as 0.05 eV Å⁻¹, 0.1 GPa, and 0.002 Å, respectively. To avoid the interaction between surface atoms, a vacuum layer of 15 Å was selected for each heterojunction surface system.

2.2 CoO–SnO₂ heterojunction models

Chemiresistive gas sensors employing a p–n heterojunction offer a compelling high sensitivity and specific response.⁵⁵ The interfacial free energy between the precipitations and the substrates control the process of particle coarsening. Being based on the lattice mismatch theory,⁵⁶ the interface with the lowest lattice mismatch between the second particle and the pre-existing particle will be the preferred growth direction of the second particle on the pre-exiting particle.⁵⁷ The primitive cells made from the CoO(110) surface and SnO₂(100) surface possessed lattice constants of a_CoO(110) = 4.2667 Å, b_CoO(110) = 3.017013 Å, a_SnO2(100) = 4.73727 Å, b_SnO2(100) = 3.186383 Å. The mismatch between CoO(110) and SnO₂(100) was less than 6%. Therefore, SnO₂(100) and CoO(110) were selected to structure the heterojunction. As shown in Fig. 1(a and b), the surface termination of CoO(110) has two forms:^58–60 CoO(110)(I) (1O and 1/4Co as termination) and CoO(110)(II) (1O and 1/2Co as termination). In addition, the surface termination of SnO₂(100) has four different surface terminations: SnO₂(100)(I) (1/4Sn-termination), SnO₂(100)(II) (1/2O-termination), SnO₂(100)(III) (1Sn-termination), and SnO₂(100)(IV) (1O-termination), as shown in Fig. 1(c–f). There are sixteen kinds of heterojunction due to the four terminations of SnO₂(100) and two terminations of CoO(110), which are shown in Fig. 2(a) and (b). In order to reduce the effect of lattice interactions on the adsorption, (1 × 2) supercellular SnO₂(100)–CoO(110) heterojunctions were fabricated.

Fig. 1 Surface structure of CoO(110) and SnO₂(100). (a) CoO(110)(I), (b) CoO(110)(II), (c) SnO₂(100)(I), (d) SnO₂(100)(II), (e) SnO₂(100)(III), (f) SnO₂(100)(IV).

Fig. 2 Different models of CoO/SnO₂ heterojunction and SnO₂/CoO heterojunction. (a) SnO₂(100) was put on CoO(110); (b) CoO(110) was put on SnO₂(100).

3. Results and discussion

3.1 Adsorption properties of H₂ on the heterojunction surface

3.1.1 Adhesion work of the heterojunction. The adhesion work (W_ad) needed to be calculated, in order to gain an insight into the binding strength of the interface, where W_ad is defined as the reversible work of separating the interface into two free surfaces, and is given by the difference in the total energy between the interfaces. Generally, a larger W_ad is associated with a more stable interface. The adhesion work is defined as follows:^61–63where E_SnO2 and E_CoO are the total energy of SnO₂ and CoO slabs, respectively, E_SnO2/CoO represents the total energy of the heterojunction interface, and A is the interface area.

The interfaces of the SnO₂(100)/CoO(110) system and CoO(110)/SnO₂(100) system showed the same distances (2 Å), which were optimized by energy calculation for evaluating the interface properties of the heterojunction. The interface structures were optimized with in a fixed cell volume, and all the atoms at the interface models were allowed to relax in three directions.⁶⁴ The W_ad values of the heterojunction are shown in Table 1.

Table 1 Bonding energy (J m⁻²) of SnO₂–CoO heterojunctions

Heterojunction	W ad (J m^−2)
SnO2(100)(I)/CoO(110)(I)	−0.5179
SnO2(100)(II)/CoO(110)(I)	−0.5187
SnO2(100)(III)/CoO(110)(I)	−0.5203
SnO2(100)(IV)/CoO(110)(I)	−0.5194
SnO2(100)(I)/CoO(110)(II)	−0.5175
SnO2(100)(II)/CoO(110)(II)	−0.5191
SnO2(100)(III)/CoO(110)(II)	−0.5177
SnO2(100)(IV)/CoO(110)(II)	−0.5191
CoO(110)(I)/SnO2(100)(I)	−0.5189
CoO(110)(I)/SnO2(100)(II)	−0.5185
CoO(110)(I)/SnO2(100)(III)	−0.5195
CoO(110)(I)/SnO2(100)(IV)	−0.5176
CoO(110)(II)/SnO2(100)(I)	−0.5176
CoO(110)(II)/SnO2(100)(II)	−0.5175
CoO(110)(II)/SnO2(100)(III)	−0.5195
CoO(110)(II)/SnO2(100)(IV)	−0.5182

---

A larger W_ad is associated with a stronger binding force of the interfacial atoms, which suggests the stability of the SnO₂(100)/CoO(110) and CoO(110)/SnO₂(100) heterojunctions interfaces.⁶⁵

Table 1 shows the adhesion work of the SnO₂(100)/CoO(110) heterojunctions, where it can be seen that the adhesion work of the SnO₂(100)(I)/CoO(110)(I) and SnO₂(100)(I)/CoO(110)(II) heterojunctions were larger than that of the other heterojunctions. Therefore, they were more stable than the other SnO₂(100)/CoO(110) heterojunctions. However, SnO₂(100)(I)/CoO(110)(II) heterojunction had a larger W_ad than that of the SnO₂(100)(I)/CoO(110)(I) heterojunction. Therefore, the SnO₂(100)(I)/CoO(110)(II) heterojunction was selected as the adsorption carrier of H₂, which was attributed to its larger adhesion work.

Table 1 shows the adhesion work of the CoO(110)/SnO₂(100) heterojunctions. It shows that the W_ad values of the CoO(110)(II)/SnO₂(100)(I), CoO(110)(II)/SnO₂(100)(II), and CoO(110)(I)/SnO₂(100)(IV) heterojunctions were larger than those of the other CoO(110)/SnO₂(100) heterojunctions. Therefore, they were more stable than the other CoO(110)/SnO₂(100) heterojunctions. Although three heterojunctions had approximately the same W_ad, the CoO(110)(II)/SnO₂(100)(II) heterojunction had the larger W_ad. Consequently, the CoO(110)(II)/SnO₂(100)(II) heterojunction system was chosen as the adsorption carrier of H₂.

3.1.2 Adsorption energy of H₂ on the heterojunction. In the air atmosphere, the oxygen in the air will capture the electrons on the surface of a heterojunction to form adsorbed oxygen anions.⁶⁶ Consequently, it was necessary to first find out the most stable adsorption site and configuration of O₂ molecule on the heterojunction surface. The adsorption behaviors of O₂ molecules on the heterojunction surface are shown in Fig. S1 and S2 (ESI†); in addition, the related parameters are shown in Tables S1 and S2 (ESI†). The results showed that different heterojunction terminals have different oxygen adsorption configurations. The most stable adsorption structure of O₂ on the heterojunction surface is shown in Fig. 3. Therefore, we used these structures as the initial configurations to explore the adsorption behavior of H₂ on the heterojunctions.

Fig. 3 The most stable adsorption structure of O₂ molecule on the heterojunction surface. (a) O₂ on the SnO₂(100)(I)/CoO(110)(II) heterojunction, (b) O₂ on the CoO(110)(II)/SnO₂(100)(II) heterojunction.

There were nine different types of adsorption conditions of H₂ on SnO₂(100)(I)/CoO(110)(II) and CoO(110)(II)/SnO₂(100)(II) heterojunctions in an oxygen atmosphere (including the adsorption on the O₂ molecule and different sites of the heterojunction surface), which are shown in Fig. 4(a–e) and (f–i).⁶⁷

Fig. 4 H₂ adsorbed on different sites of heterojunction surfaces. (a–e) H₂ on the SnO₂(100)(I)/CoO(110)(II) heterojunction, (f–i) H₂ on the CoO(110)(II)/SnO₂(100)(II) heterojunction.

The adsorption energy (E_ads) is a key standard to estimate the adsorption property of H₂ molecules on the heterojunction surface. It can be adopted to judge the adsorption strength of H₂ molecules on a carrier, which is defined as follows:^68,69

E_ads = E_total − (E_{heterojunction} + E_gas) (2)

where E_ads is the adsorption energy (eV), E_total represents the total energy of the whole adsorption system, E_{heterojunction} is the total free energy of the heterojunction system (eV), and E_gas is the total free energy of an isolated gas molecule (eV). In general, E_ads with a negative value indicates that the adsorption is exothermic, and the system of adsorption is energetically stable.^70,71

The E_ads and the adsorbed distance of H₂ adsorption on the SnO₂(100)(I)/CoO(110)(II) and CoO(110)(II)/SnO₂(100)(II) heterojunctions are shown in Table 2. In addition, the parameters about E_ads and the adsorbed distance of H₂ on single material oxide CoO and SnO₂ are shown in Tables S3 and S4 (ESI†).

Table 2 Adsorption of H₂ on the heterojunctions

Heterojunction	Site	E ads (eV)	Distance (Å)
SnO2(100)(I)/CoO(110)(II)	O2	−0.2286	2.4615
	O1	−0.6962	3.3761
	O2	−0.3900	4.1691
	Co1	−0.0898	3.1388
	Co2	1.6229	4.3275
CoO(110)(II)/SnO2(100)(II)	O2	−0.2957	3.0844
	Sn	−0.3855	3.4432
	O1	−0.2551	3.2475
	O2	−0.2496	3.8159

---

Table 2 shows the E_ads and adsorption distance when H₂ was adsorbed on different sites of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface. There were five forms where H₂ was adsorbed on the heterojunction surface, including the O site of an O₂ molecule, and O1 site, O2 site, Co1 site, Co2 site of the heterojunction surface. The adsorption distance and E_ads varied greatly. When H₂ was adsorbed on the O site of an O₂ molecule, the distance of H–O bonding was less than 3 Å, indicating the formation of a chemical bond. Except at the Co2 site, the E_ads values were all negative when H₂ was adsorbed on other sites of the heterojunction surface. This reveals the preferable stabilization of adsorption, especially the adsorption on an O site of an O₂ molecule, and O1 site and O2 site of the heterojunction surface. Therefore, according to the W_ads and adsorption distance, it was easy for H₂ to be adsorbed on an O site of an O₂ molecule. Compared with pure metal oxides CoO, the SnO₂(100)(I)/CoO(110)(II) heterojunction improves the adsorption and bonding of H₂ on an O₂ molecule. Also, the adsorption of H₂ on the heterojunction was more stable.

Table 2 shows the E_ads and adsorbed distance of H₂ adsorption on O site of an O₂ molecule, Sn (Sn is equivalent because of supercell) site, and O1 site and O2 site of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface. The adsorption distance and E_ads varied greatly. The adsorbed distances were all more than 3 Å when H₂ was adsorbed on different sites of the heterojunction surface. This implies the existence of weak chemical bonding between H atoms and the adsorbed sites. In addition, the E_ads values were obviously different when H₂ was adsorbed on different surface sites. The E_ads values were all negative when H₂ was adsorbed on the heterojunction surface, revealing that the adsorption of H₂ on the heterojunction was thermodynamically stable. Consequently, it was easy for H₂ to be adsorbed on the CoO(110)(II)/SnO₂(100)(II) heterojunction surface by analyzing the adsorbed distance and E_ads. However, there was no formation of stronger chemical bonding between H₂ molecules and the adsorbed sites. Compared with pure metal oxides SnO₂, the existence of the CoO(110)(II)/SnO₂(100)(II) heterojunction made the bonding effect between H₂ and adsorption sites worse, but made the adsorption more stable, especially the adsorption of H₂ on the O1 site.

In summary, based on the E_ads and adsorption distance, H₂ readily adsorbs on an O₂ molecule of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface. In addition, H₂ tends to be adsorbed on different sites of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface; however, stronger chemical bonds did not exist.

3.2 Influence of gas adsorption on the electronic conductivity and electronic structure

3.2.1 Density of states (DOS). To understand the adsorption effect of H₂ on the conductivity of a heterojunction in the oxygen atmosphere, DOS analysis was performed of single metal oxide CoO and SnO₂ adsorbed O₂ on surface. The adsorption of one O₂ molecule on the CoO and SnO₂ surface is shown in Fig. S3 and S4 (ESI†). Then, the effect of H₂ adsorption on the surface conductivity of single metal oxide was studied under oxygen conditions. Finally, the gas-sensing properties of the heterojunction were studied under oxygen conditions.

The adsorption of H₂ in the oxygen atmosphere will affect the heterojunction resistance, which could be reflected in DOS. Therefore, the change in the resistance can be studied by analyzing the DOS. The generation of more peaks and increase in the DOS values were found over the tested energy range, which were attributed to the gas adsorption. The variations in DOS would further account for the increase in the electron energy and the better conductivity.^72–76

The DOS values for H₂ on the different sites of the SnO₂(100)(I)/CoO(110)(II) and CoO(110)(II)/SnO₂(100)(II) heterojunction surfaces are shown in Fig. 5(a) and (b). In addition, the DOS for H₂ adsorbed on different sites of single metal oxides CoO and SnO₂ are shown in Fig. S5(a) and (b) (ESI†).

Fig. 5 DOS of H₂ adsorption on different sites of the heterojunction surface. (a) H₂ on O₂ molecule, and O1 site, O2 site, Co1 site, Co2 site of the SnO₂(100)(I)/CoO(110)(II) heterojunction. (b) H₂ on O₂ molecule, and O1 site, O2 site, Sn site of the CoO(110)(II)/SnO₂(100)(II) heterojunction.

Fig. 5(a) shows the DOS for H₂ adsorption on different sites of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface. The DOS value was reduced when H₂ was adsorbed on an O site of the O₂ molecule, labeled as 1 and 2, revealing the heterojunction resistance had increased. The DOS value showed obvious decreases (labeled as 3, 4, and 5) when H₂ was adsorbed on the O1 site, indicating the heterojunction resistance increased along with the adsorption of H₂. When H₂ was adsorbed on the O2 site, a decrease and disappearance of DOS peaks (labeled as 6, 7, and 8) were observed. All these results prove that the resistance increased when H₂ was adsorbed on the O2 site of the heterojunction surface. The DOS value was lower and some peaks disappeared (labeled as 9 and 10) with H₂ adsorbed on the Co1 site, reflecting that the heterojunction resistance had increased. When H₂ was adsorbed on the Co2 site, the changes in the DOS occurred at the Fermi level (labeled as 11, 12, and 13). The DOS value was then reduced and some peaks disappeared, revealing that the heterojunction resistance had increased. On the whole, when H₂ was adsorbed on different sites of the heterojunction surface, the heterojunction resistance increased. Compared with the DOS of the metal oxide CoO (Fig. S5(a), ESI†), the DOS value of the SnO₂(100)(I)/CoO(110)(II) heterojunction decreases more obviously. Therefore, the heterojunction changed the gas adsorption performance for H₂.

Fig. 5(b) shows the DOS of H₂ adsorption on different sites of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface. With the adsorption of H₂ on an O₂ molecule, the DOS value was reduced distinctly (labeled as 1 and 2), implying the heterojunction resistance had improved. The DOS showed a slight alteration when H₂ was adsorbed on the Sn site, indicated that the resistance of the heterojunction had slightly changed, together with the disappearance of peaks in the conduction band (labeled as 3 and 4), indicating that the heterojunction resistance was enlarged. In the DOS, the number of peaks disappeared or reduced (labeled as 5, 6, and 7) when H₂ was adsorbed on the O1 site, indicating that the heterojunction resistance had increased. DOS showed a slight reduction and some peaks disappeared (labeled as 8, 9, 10, and 11) when H₂ was adsorbed on the O2 site, indicating that the heterojunction resistance had increased. These results proved the heterojunction resistance increased with the adsorption of H₂. The DOS changes of H₂ on the heterojunction surface was opposite to that for H₂ on the SnO₂ surface (Fig. S5(b), ESI†). This shows that the presence of a heterojunction will change the response trend of H₂ on the SnO₂ surface.

In summary, compared with single metal oxides CoO and SnO₂, the existence of the SnO₂(100)(I)/CoO(110)(II) heterojunction and CoO(110)(II)/SnO₂(100)(II) heterojunction changed the adsorption performance to H₂. The response trend of the SnO₂(100)(I)/CoO(110)(II) heterojunction to H₂ was the same as that with CoO, but the response trend of the CoO(110)(II)/SnO₂(100)(II) heterojunction to H₂ was the opposite as that with SnO₂. This will provide theoretical guidance for the accurate identification of H₂ and the manufacture of sensors.

3.2.2 Electronic structure of H₂ on the SnO₂(100)/CoO(110) heterojunction surface. In order to understand the essence of adsorption, the electronic structure of H₂ adsorption on the SnO₂(100)/CoO(110) heterojunction in an oxygen atmosphere was assessed. The binding of atoms has a strong relationship with the electronic structure, which can be characterized by the electronic density and electronic density difference. The total electronic density and electronic density difference were calculated, and are shown in Fig. 6 and 7, respectively.

Fig. 6 Electronic structure of H₂ adsorption on different sites of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface. Optimized structure of the heterojunction; electronic density; electronic density difference. (a) H₂ on O₂ molecule, (b) H₂ on O1 site, (c) H₂ on O2 site, (d) H₂ on Co1 site, (e) H₂ on Co2 site.

Fig. 7 Electronic structure of H₂ adsorption on different sites of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface. Optimized structure of the heterojunction; electronic density; electronic density difference. (a) H₂ on O₂ molecule, (b) H₂ on O1 site, (c) H₂ on O2 site, (d) H₂ on Sn site.

Fig. 6 displays the optimized structure, electronic density, and electronic density difference of H₂ adsorption on different sites (including O₂ molecule, O1 site, O2 site, Co1 site and Co2 site) of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface. The optimized structure of H₂ adsorption on an O₂ molecule of the SnO₂(100)(I)/CoO(110)(II) heterojunction surface is shown in Fig. 6(a). The adsorbed distance was 2.4615 Å and E_ads was −0.2286 eV, as shown in Table 2. The adsorbed distance was smaller than that for H₂ on other sites of the heterojunction surface.

The blue ranges represent the depletion areas of the electrons, and the red areas show the accumulation region of the electrons. As displayed in Fig. 6(a), H₂ was adsorbed on the O₂ molecule of the heterojunction surface. The adsorbed distance was less than 3 Å and E_ads was negative, and the electronic sharing occurred between H atoms and the O₂ molecule. Therefore, the H atoms share an extensive amount of electrons with the O₂ molecule, which indicates that H–O bonding has stronger covalence. In addition, the O₂ molecule shares extensive electrons with atoms of the heterojunction surface, illustrating the stable adsorption of the O₂ molecule on the heterojunction surface and the formation of a covalent bond with atoms of the heterojunction surface. As shown in Fig. 6(a), the H₂ molecule and O₂ molecule, which are adsorbed on the heterojunction surface, get an amount of electrons simultaneously. This implies the existence of an electronic transformation and chemical reaction between H atoms and O atoms of O₂.

The electron sharing hardly occurred and the adsorption distance was greater than 3 Å when H₂ was adsorbed on other sites of the heterojunction surface, and although there was slight electron transfer between the sites, there was no bonding. Therefore, O₂ molecule was the best adsorption site of H₂ on the heterojunction surface and the heterojunction was the best carrier of H₂. In the air environment, the adsorbed O₂ molecules extract electrons bands of the heterojunction and convert to oxygen species, such as O²⁻, O⁻, and O₂⁻, covering the surface of the heterojunction.^77,78 The subsequent reaction (H₂ + O^x− → H₂O + xe⁻) releases electrons back into the heterojunction surface.⁵⁴ An atomic level understanding of the H₂-sensing mechanism on the heterojunction was thus proved.

3.2.3 Electronic structure of H₂ on the CoO(110)(II)/SnO₂(100)(II) heterojunction. The electronic structures of H₂ adsorption on different sites of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface are shown in Fig. 7, together with the optimized structure of H₂ adsorption on different sites of the heterojunction. The E_ads values were all negative when H₂ was adsorbed on different sites of the heterojunction surface. This shows that H₂ can be stably adsorbed on the heterojunction surface.

Fig. 7(a) reveals the electronic density when H₂ was adsorbed on the O₂ molecule. As shown, there was little electronic accumulation between the H atoms and O₂ molecule when H₂ was adsorbed on the O₂ molecule. The H atoms shared a bit of electronic density with the O₂ molecule on the heterojunction surface and the adsorbed distance was more than 3 Å, indicating the existence of a slight H–O covalent bond. Fig. 7(a) shows the electronic density difference when H₂ was adsorbed on the O₂ molecule of the heterojunction surface, whereby the H atoms acquired extensive electrons, and at the same time, a gain and loss of electrons occurred on the O₂ molecule. Consequently, the transformation of the electronic density and the chemical reaction between H atoms and the O₂ molecule were proved. Fig. 7(b) reveals the electronic density when H₂ was adsorbed on the O1 atom of the CoO(110)(II)/SnO₂(100)(II) heterojunction surface. There was clear electronic accumulation between H atoms and the O1 atom when H₂ was adsorbed on the O1 atom. The H atoms shared some electrons with the O1 atom on the SnO₂ side and the adsorbed distance was more than 3 Å, which suggests the existence of H–O bonding. Fig. 7(b) shows the electronic density difference when H₂ was adsorbed on the O1 site of the heterojunction surface; whereby the H atoms acquired some electrons and the O1 atom of the heterojunction surface obtained some electrons too. Consequently, the slight transformation of the electronic density and a slight reaction between the H atoms and O1 atom were proved.

Fig. 7(c) and (d) show the electronic structure when H₂ was adsorbed on the O2 site and Sn site of the heterojunction surface. Barely any electrons existed between H atoms and the O2 atom and Sn atom of the heterojunction surface. This indicated the formation of weaker H–O bonding and H-Sn bonding for H atoms sharing a few electrons with the O2 atom of the heterojunction surface. This means a small transformation in the electronic density and a slight reaction between H₂ and the O1 atom and Sn atom. In summary, O₂ molecule on the heterojunction was the best site for reacting with H₂, and the O1 site was the best adsorption site for H₂.

4. Conclusion

In this study, DFT was used to calculate the adsorption properties of H₂ on CoO/SnO₂ and SnO₂/CoO heterojunction surfaces under air conditions. First, the optimal heterojunction structure was determined based on the W_ad. Second, the most stable adsorption configuration of H₂ in the air was identified based on the results of the E_ads calculation. Finally, the mechanism of the gas adsorption process was revealed by DOS and by the electronic structure, and the following conclusions were obtained:

(1) A smaller lattice mismatch and larger W_ad are associated with a more stable interface. The CoO(110) surface and SnO₂(100) surface were chosen for forming the structure of the heterojunction. SnO₂(100)(I)/CoO(110)(II) and CoO(110)(II)/SnO₂(100)(II) were more stable than other SnO₂(100)/CoO(110) and CoO(110)/SnO₂(100) heterojunctions. Therefore, the stability, resistance, and electronic structure of SnO₂(100)(I)/CoO(110)(II) and CoO(110)(II)/SnO₂(100)(II) heterojunctions were mainly studied as carriers of H₂ in oxygen atmosphere.

(2) The heterojunction resistance increased when H₂ was adsorbed on different sites of the SnO₂(100)(I)/CoO(110)(II) surface, which, followed the same trend as for H₂ adsorption on the CoO surface. Moreover, its trend of change was greater according to DOS. This shows that the heterojunction can change the gas adsorption performance of H₂. The O₂ molecule on the heterojunction surface is the best adsorption site for H₂, because of the larger electron density distribution between them. In addition, the heterojunction is the best carrier of H₂ for the existence of the reaction between H₂ and surface oxygen ions. An atomic level understanding of the H₂ sensing mechanism on the heterojunction was proved.

(3) The heterojunction resistance increased when H₂ was adsorbed on different sites of the CoO(110)(II)/SnO₂(100)(II) surface. Also, the heterojunction resistance showed the opposite trend to H₂ adsorption on the SnO₂ surface. When the SnO₂/CoO heterojunction was exposed to H₂ gas, H₂ reacted with the oxygen ions on the SnO₂ surface and the electrons returned to the conduction band, which led to energy band bending. Nonetheless, H₂ hardly reacted with the oxygen ions on the CoO surface. As a result, the potential barrier of SnO₂/CoO increased, leading to an increase in resistance and a p-type response to H₂. Therefore, the presence of the heterojunction will change the trend of the response to H₂. Meanwhile, according to the electronic density, the O1 site of the heterojunction surface was the best adsorption site for H₂, because of the larger electron density distribution among them. Finally, as judged by E_ads, the heterojunction improved the stability of H₂ adsorption on the surface.

Conflicts of interest

There are no conflicts declare.

Acknowledgements

We are very grateful for the support of the National Natural Science Foundation of China (Fund number: 51974157, 51874169, 51774180 and 51634004).

References

1. C. B. Lim and S. Oh, Microstructure evolution and gas sensitivities of Pd-doped SnO₂-based sensor prepared by three different catalyst-addition processes, Sens. Actuators, B, 1996, 30, 223–231.
2. F. Hartmann, Semiconductor sensors, Nucl. Instrum. Methods Phys. Res, 2011, 628, 40–49.
3. A. T. Güntner and N. J. Pineau, et al., Sniffing Entrapped Humans with Sensor Arrays, Anal. Chem., 2018, 90, 4940–4945.
4. M. Righettoni, A. Amann and E. S. Pratsinis, Breath analysis by nanostructured metal oxides as chemo-resistive gas sensors, Mater. Today, 2015, 18, 163–171.
5. P. Reimann, A. Dausend and A. Schutze, A self-monitoring and self-diagnosis strategy for semiconductor gas sensor systems, SENSORS, 2008 IEEE, 2008, 192–195.
6. M. Schüler, T. Sauerwald and A. Schütze, A novel approach for detecting HMDSO poisoning of metal oxide gas sensors and improving their stability by temperature cycled operation, J. Sens. Sens. Syst., 2015, 4, 305–311.
7. L. P. Huo, Y. Xi and Z. Liu, et al., Modulation of Potential Barrier Heights in Co₃O₄/SnO₂ Heterojunction for Highly H₂-Selective Sensors, Sens. Actuators, B, 2017, 244, 694–700.
8. N. V. Toan and N. V. Duy, et al., Fabrication of highly sensitive and selective H₂ gas sensor based on SnO₂ thin film sensitized with microsized Pd islands, J. Hazard. Mater., 2015, 301, 433–442.
9. C. Liewhiran, N. Tamaekong and A. Wisitsoraat, et al., Ultra-sensitive H₂ sensors based on flame-spray-made Pd-loaded SnO₂ sensing films, Sens. Actuators, B, 2013, 176, 893–905.
10. C. B. Lim and S. Oh, Microstructure evolution and gas sensitivities of Pd-doped SnO₂-based sensor prepared by three different catalyst-addition processes, Sens. Actuators, B, 1996, 30, 223–231.
11. F. D. Manchester, A. San-Martin and J. M. Pitre, The H-Pd (hydrogen-palladium) System, J. Phase Equilib., 1994, 15, 62–83.
12. G. Lu and J. T. Hupp, Metal–organic frameworks as sensors: a ZIF-8 based Fabry-Pérot device as a selective sensor for chemical vapors and gases, J. Am. Chem. Soc., 2010, 132, 7832–7833.
13. M. Drobek, J. H. Kim and M. Bechelany, et al., MOF-Based Membrane Encapsulated ZnO Nanowires for Enhanced Gas Sensor Selectivity, ACS Appl. Mater. Interfaces, 2016, 8, 8323–8328.
14. H. Tian, H. Fan and M. Li, et al., Zeolitic Imidazolate Framework Coated ZnO Nanorods as Molecular Sieving to Improve Selectivity of Formaldehyde Gas Sensor, ACS Sens., 2016, 1, 243–250.
15. M. S. Yao and W. Xiang, et al., MOF Thin Film-Coated Metal Oxide Nanowire Array: Significantly Improved Chemiresistor Sensor Performance, Adv. Mater., 2016, 28, 5229–5234.
16. J. Hong, S. Lee, J. Seo, S. Pyo, J. Kim and T. Lee, A highly sensitive hydrogen sensor with gas selectivity using a PMMA membrane-coated Pd nanoparticle/single-layer graphene hybrid, ACS Appl. Mater. Interfaces, 2015, 7, 3554–3561.
17. O. Lupan, V. Postica and J. Ttrup, et al., Hybridization of Zinc Oxide Tetrapods for Selective Gas Sensing Applications, ACS Appl. Mater. Interfaces, 2017, 9, 4084–4099.
18. V. Musat, E. Fortunato and A. M. Botelho do Rego, et al., Sol-gel cobalt oxide-silica nanocomposite thin films for gas sensing applications, Thin Solid Films, 2008, 516, 1499–1502.
19. M. Matsumiya, F. Qiu and W. Shin, et al., Thermoelectric CO Gas Sensor Using Thin-Film Catalyst of Au and Co₃O₄, J. Electrochem. Soc., 2003, 151, H7.
20. Q. Jiao, M. Fu and C. You, et al., Preparation of hollow Co₃O₄ microspheres and their ethanol sensing properties, Inorg. Chem., 2012, 51, 11513–11520.
21. X. Yu, X. Liu and M. Wu, et al., Hierarchical radial Co₃O₄ microcrystal and application in gas sensor, Chin. J. Chem. Phys., 2014, 27, 99–102.
22. D. Barreca, E. Comini and A. P. Ferrucci, et al., First Example of ZnO-TiO₂ Nanocomposites by Chemical Vapor Deposition: Structure, Morphology, Composition, and Gas Sensing Performances, Chem. Mater., 2007, 19, 5642–5649.
23. W. J. Moon, H. J. Yu and G. M. Choi, The CO and H₂ gas selectivity of CuO-doped SnO₂–ZnO composite gas sensors, Sens. Actuators, B, 2002, 87, 464–470.
24. J. H. Kim, J. H. Lee and A. Mirzaei, et al., SnO₂ (n)-NiO (p) composite nanowebs: gas sensing properties and sensing mechanisms, Sens. Actuators, B, 2017, 258, 204–214.
25. D. R. Miller, S. A. Akbar and P. A. Morris, Nanoscale metal oxide-based heterojunction for gas sensing: a review, Sens. Actuators, B, 2014, 204, 250–272.
26. Y. Liu, G. Li and R. Mi, et al., An environment-benign method for the synthesis of p-NiO/n-ZnO heterostructure with excellent performance for gas sensing and photocatalysis, Sens. Actuators, B, 2014, 191, 537–544.
27. J. Martinez, S. B. Sinnott and S. R. Phillpot, Adhesion and diffusion at TiN/TiO₂ interfaces: a first principles study, Comput. Mater. Sci., 2017, 130, 249–256.
28. J. H. Kim, J. H. Lee and A. Mirzaei, et al., SnO₂(n)-NiO(p) composite nanowebs: gas sensing properties and sensing mechanisms, Sens. Actuators, B, 2017, 258, 204–214.
29. B. Xu, C. Zhu and X. He, et al., First-Principles Calculations on Atomic and Electronic Properties of Ge/4H-SiC Heterojunction, Adv. Conden. Matter Phys., 2018, 3, 1–9.
30. X. Fan, B. Chen and M. Zhang, et al., First-principles calculations on bonding characteristic and electronic property of TiC(111)/TiN(111) interface, Mater. Des., 2016, 112, 282–289.
31. A. Katoch, S. W. Choi and J. H. Kim, et al., Importance of the nanograin size on the H₂S-sensing properties of ZnO–CuO composite nanofibers, Sens. Actuators, B, 2015, 214, 111–116.
32. S. W. Choi, Z. Jin and K. Akash, et al., H₂S sensing performance of electrospun CuO-loaded SnO₂ nanofibers, Sens. Actuators, B, 2012, 169, 54–60.
33. D. Ju, H. Xu and Q. Xu, et al., High triethylamine-sensing properties of NiO/SnO₂ hollow sphere P–N heterojunction sensors, Sens. Actuators, B, 2015, 215, 39–44.
34. Q. Zhi, Y. Fu and B. Yu, et al., High and fast H₂S response of NiO/ZnO nanowire nanogenerator as a self-powered gas sensor, Sens. Actuators, B, 2016, 222, 78–86.
35. Y. T. Kwon and W. H. Yeo, Flexible, Wearable, and Stretchable Electronics, Flexible, Wearable, and Stretchable Electronics, 2020, 1–30.
36. J. J. Kim, S. J. Chung and J. W. Choi, et al., Redox Reaction between Nitric Oxide and Carbon Monoxide over Perovskite-type Catalysts, J. Korean Inst. Chem. Eng., 1982, 20, 245.
37. L. Wang, J. Deng and Z. Lou, et al., Cross-linked p-type Co₃O₄ octahedral nanoparticles in 1D n-type TiO₂ nanofibers for high-performance sensing devices, J. Mater. Chem. A, 2014, 2, 10022–10028.
38. X. T. Yin, J. Li and D. Dastan, et al., Ultra-high selectivity of H₂ over CO with a p–n nanojunction based gas sensors and its mechanism, Sens. Actuators, B, 2020, 319, 128330.
39. B. L. Bramfitt, The effect of carbide and nitride additions on the heterogeneous nucleation behavior of liquid iron, Metall. Trans., 1970, 1, 1987–1995.
40. H. Zhang and H. Xiong, First-principles study of NbC heterogeneous nucleation on TiC vs. TiN in microalloy steel, Ironmaking Steelmaking, 2018, 47, 77–83.
41. A. Mirzaei, S. G. Leonardi and G. Neri, Detection of hazardous volatile organic compounds (VOCs) by metal oxide nanostructures-based gas sensors: a review, Ceram. Int., 2016, 42, 15119–15141.
42. Y. Jian, J. Huang and D. Fan, et al., First-principles investigation on the electronic property and bonding configuration of NbC(111)/NbN(111) interface, J. Alloys Compd., 2016, 689, 874–884.
43. X. Fan, B. Chen and M. Zhang, et al., First-principles calculations on bonding characteristic and electronic property of TiC(111)/TiN(111) interface, Mater. Des., 2016, 112, 282–289.
44. L. Shi, H. Yan and Z. H. Lu, et al., Surface Property of the Cu Doped γ-Al₂O₃: A Density Functional Theory Study, Appl. Surf. Sci., 2020, 535, 147651.
45. G. Feng, Q. Xiao and D. S. Wang, et al., Acid properties of Ni-modified ZSM-12: a first-principles Study, J. Fuel Chem. Technol., 2020, 48, 704–712.
46. G. Feng, J. Yang and C. Wang, et al., A first-principles study on Pd modified ZSM-12 zeolites, Microporous Mesoporous Mater., 2018, 260, 227–234.
47. G. Feng, M. V. Ganduglia-Pirovano and C. F. Huo, et al., Hydrogen spillover to copper clusters on hydroxylated γ-Al₂O₃, J. Phys. Chem. C, 2018, 122, 18445–18455.
48. X. Yu, X. Zhang and H. Wang, et al., High-coverage H₂ adsorption on the reconstructed Cu₂O(111) surface, J. Phys. Chem. C, 2017, 121, 22081–22091.
49. L. Shi, S. Meng and S. Jungsuttiwong, et al., High coverage H₂O adsorption on CuAl₂O₄ surface: a DFT study, Appl. Surf. Sci., 2020, 507, 145162.
50. L. Ou, Y. Fan and Y. Liu, et al., First-Principle Study of the Adsorption and Dissociation of O₂ on Pt(111) in Acidic Media, J. Phys. Chem. C, 2009, 113, 20657–20665.
51. J. Chen and C. Ye, A first-principle study of the effect of vacancy defects and impurities on the adsorption of O₂ on sphalerite surfaces – Science Direct, Colloids Surf., A, 2010, 363, 56–63.
52. H. Xiong and Z. Liu, et al., First principles calculation of interfacial stability, energy and electronic properties of SiC/ZrB₂ interface-Science Direct, J. Phys. Chem. Solids, 2017, 107, 162–169.
53. Y. Jian, J. Huang and D. Fan, et al., First-principles investigation on the electronic property and bonding configuration of NbC(111)/NbN(111) interface, J. Alloys Compd., 2016, 689, 874–884.
54. W. H. Lee and X. H. Yao, First principle investigation of phase transition and thermodynamic properties of SiC, Comput. Mater. Sci., 2015, 106, 76–82.
55. L. Huo, Y. Xi and Z. Liu, et al., Modulation of Potential Barrier Heights in Co₃O₄/SnO₂ Heterojunction for Highly H₂-Selective Sensors, Sens. Actuators, B, 2017, 244, 694–700.
56. B. L. Bramfitt, The effect of carbide and nitride additions on the heterogeneous nucleation behavior of liquid iron, Metall. Trans., 1970, 1, 1987–1995.
57. H. Zhang and H. Xiong, First-principles study of NbC heterogeneous nucleation on TiC vs., TiN in microalloy steel, Ironmaking Steelmaking, 2018, 47, 77–83.
58. R. Liu, X. Yin and K. Feng, et al., First-principles calculations on Mg/TiB₂ interfaces, Comput. Mater. Sci., 2018, 149, 373–378.
59. L. Xiao, Q. Hui and D. Shao, et al., First-principles study on the stability and electronic structure of Mg/ZrB₂ interfaces, Sci. China: Mater., 2016, 59, 28–37.
60. H. H. Xiong, H. N. Zhang and J. H. Dong, Adhesion strength and stability of TiB₂/TiC interface in composite coatings by first principles calculation, Comput. Mater. Sci., 2017, 127, 244–250.
61. L. Jian, Y. Yang and G. Feng, et al., First-principles study of stability and properties on β-SiC/TiC(111) interface, J. Appl. Phys., 2013, 114, 54–59.
62. D. Yin, X. Peng and Y. Qin, et al., Electronic property and bonding configuration at the TiN(111)/VN(111) interface, J. Appl. Phys., 2010, 108, 033714.
63. D. Y. Dang, L. Y. Shi and J. L. Fan, et al., First-principles study of W-TiC interface cohesion, Surf. Coat. Technol., 2015, 276, 602–605.
64. N. Chandran, M. Sall and J. Arvanitidis, et al., On the formation of graphene by Ge intercalation of a 4H-SiC surface, Mater. Sci. Forum, 2015, 821, 961–964.
65. H. Zhang and H. Xiong, First-principles study of NbC heterogeneous nucleation on TiC vs. TiN in microalloy steel, Ironmaking Steelmaking, 2018, 47, 77–83.
66. H. J. Kim and J. H. Lee, Highly sensitive and selective gas sensors using p-type oxide semiconductors: overview, Sens. Actuators, B, 2014, 192, 607–627.
67. S. Chu and A. Majumdar, Opportunities and challenges for a sustainable energy future, Nature, 2012, 488, 294–303.
68. L. Tao, J. Huang and D. Dastan, et al., New insight into absorption characteristics of CO₂ on the surface of calcite, dolomite, and magnesite, Appl. Surf. Sci., 2021, 540, 148320.
69. L. Tao, J. Huang and D. Dastan, et al., CO₂ capture and separation on charge-modulated calcite, Appl. Surf. Sci., 2020, 530, 147265.
70. M. Viitala, et al., Conformation and energetics of benzene adsorbate on SnO₂(110) surfaces: a first principles study, Surf. Sci., 2011, 605, 1563–1567.
71. Z. Wen, T. M. Liu and D. J. Liu, Formaldehyde gas sensing property and mechanism of TiO₂–Ag nanocomposite, Phys. B, 2010, 405, 4235–4239.
72. M. H. Li, H. C. Zhu and G. F. Wei, et al., DFT calculation and analysis of the gas sensing mechanism of methoxy propanol on Ag decorated SnO₂(110) surface, RSC Adv., 2019, 9, 35862–35871.
73. W. Zeng, T. M. Liu and D. J. Liu, et al., Hydrogen sensing and mechanism of M-doped SnO₂ (M = Cr³⁺, Cu²⁺ and Pd²⁺) nanocomposite, Sens. Actuators, B, 2011, 160, 455–462.
74. W. Zeng, T. M. Liu and X. F. Lei, Hydrogen sensing properties of low-index surfaces of SnO₂ from first-principles, Phys. B, 2010, 405, 3458–3462.
75. S. Yang, Z. Lan and H. Xu, et al., A First-Principles Study on Hydrogen Sensing Properties of Pristine and Mo-Doped Graphene, J. Nanotechnol., 2018, 2018, 1–5.
76. Y. H. Zhang, L. F. Han and Y. H. Xiao, et al., Understanding dopant and defect effect on H₂S sensing performances of graphene: a first-principles study, Comput. Mater. Sci., 2013, 69, 222–228.
77. A. Gurlo and R. Riedel, In situ and operando spectroscopy for assessing mechanisms of gas sensing, Angew. Chem., Int. Ed., 2007, 46, 3826–3848.
78. A. Gurlo, Interplay between O₂ and SnO₂: oxygen ionosorption and spectroscopic evidence for adsorbed oxygen, Phys. Chem. Chem. Phys., 2006, 7, 2041–2052.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1cp04539c

---