Oxidation of CO by N₂O over Al- and Ti-doped graphene: a comparative study

Abstract

In this work, we employ density functional theory calculations to investigate the CO oxidation mechanisms by N₂O molecules over Al- or Ti-doped graphene (Al–/Ti–graphene). The reaction barriers and thermodynamic parameters are calculated using the M06-2X density functional with the 6-31G* basis set. The possible reaction pathway proposed for the oxidation of CO with N₂O molecules is as follows: N₂O → N₂ + O_ads and CO + O_ads → CO₂. Unlike Al–graphene, upon adsorption of the N₂O molecules over Ti–graphene, they are quickly dissociated into N₂ and O_ads species via a barrier-less reaction. Then, the activated O_ads reacts with CO molecules to form CO₂ molecules. The calculated activation energies of the reaction CO + O_ads → CO₂ on Al– and Ti–graphene are calculated to be 0.06 and 0.16 eV, which are lower than those on traditional noble metal catalysts. Our results indicate that both Al– and Ti–graphene can be used as a potential catalyst for low-temperature CO oxidation by N₂O molecules.

---

1. Introduction

Carbon monoxide (CO) is one of the most prominent atmospheric pollutants produced largely by incomplete burning of fuels in industrial and automotive combustion processes. The health significance of CO is due to the fact that it is able to combine irreversibly with the iron atom of haemoglobin to produce carboxyhaemoglobin, whose formation prevents blood from having its normal oxygen-carrying capacity.¹ Hence, the catalytic oxidation of CO to carbon dioxide (CO₂) has become one of the most efficient reactions to clean the exhausts from automobiles and industries. In recent years, numerous studies have been devoted to the CO oxidation reaction by O₂ molecules as the most common oxidizing agent.^2–9 Also, a few investigations have been reported with other oxidizing agents such as nitrous oxide (N₂O).^10–12 The N₂O molecule is known as a major greenhouse gas with a larger global warming potential than CO₂. It is emitted to the atmosphere mainly from agricultural and industrial activities. Thus, besides the oxidation of CO molecules, the reduction of N₂O becomes one of the main important topics for scientific investigations. It should be noted that the direct reaction of N₂O with CO molecules has an extremely high activation energy which hinders the reaction to take place at room temperature. Therefore, it can be catalyzed by different types of catalysts in the gas phase.^13–15 According to previous studies,^16,17 the catalytic oxidation of CO by N₂O proceeds as the following reactions:

N₂O → N₂ + O_ads (1)

O_ads + CO → CO₂ (2)

A variety of investigations have been performed on the oxidation of CO by N₂O molecules using transition metal cations.^18–20 For instance, Kartha et al.²¹ indicated that substituting Fe atoms in lanthanum titanates can significantly enhance the catalytic activity of CO oxidation by N₂O molecules. However, despite these extensive investigations performed on the CO oxidation by N₂O molecules over various metal surfaces, recently there have been considerable efforts to do this oxidation/reduction reaction over cheaper and favourable carbon-based materials such as graphene^22,23 or boron nitride nanotubes (BNNTs).²⁴

Graphene is a novel form of carbon that can act as a potential support for metal atoms and clusters to realize new carbon–metal nanocomposite catalysts. Recently, graphene as a matrix of a heterogeneous catalyst has been extensively studied due to its huge surface-to-volume ratio that provides a large catalytic reaction area. The exclusive π-conjugated structure of graphene results in outstanding thermal, mechanical, and electrical properties^25–28 and can be considered for the next generation of electronic materials.²⁹ According to previous studies,³⁰ the existence of the strong sp² bonding between the carbon atoms of pure graphene sheets makes it chemically inert. In order to solve this problem, researchers have proposed that substituting the carbon atoms of graphene sheets with foreign atoms can extensively enhance its catalytic activity³¹ or its sensitivity toward the adsorption of gas molecules.^32,33 Some of these metallic or metal-free atoms are Fe,^34,35 Mn,^36,37 Cu,³⁸ Pd,³⁹ Sn,⁹ N,^40,41 P⁴² or Si.^43–45 For example, Wang et al.⁴⁶ studied the adsorption of CO and H₂ molecules on Fe-, Co-, Ni- and Cu-embedded graphene theoretically. They found that the Fe- and Co-embedded graphene can preferably capture up to three CO molecules per metal atom, rather than H₂ molecules. They finally estimated that Fe– and Co–graphene can be used as a filter membrane for removing CO efficiently in the feed gas of hydrogen fuel cells. Besides these light 2p (C, N) or 3p (Si, P) dopants, Al atoms as a 3p element can induced strong structural changes in graphene sheets due to their considerably larger atomic radius than those of 2p atoms.^47–50 For instance, Jiang et al.⁵¹ investigated the oxidation of CO by O₂ molecules over Al-doped graphene theoretically. They found that the amount of charge transfer from the O₂ to the CO molecule via the embedded Al atom plays an important role in the oxidation of CO molecules. So, the low-cost Al-doped graphene can be used as an efficient catalyst toward CO oxidation. Another theoretical investigation performed by Dai et al.⁵² refers to the adsorption of some common gases over B-, N-, S- and Al-embedded graphene. They demonstrated that only NO and NO₂ molecules can bind to B-doped graphene, while NO₂ molecules bind to S-doped graphene. Therefore, they revealed that B- and S-doped graphene could be used as a favourable sensor for the adsorption of gas pollutants such as NO and NO₂. Moreover, the effects of doped titanium (Ti) or N atoms on the interaction of CO, NO, SO₂, and HCHO gases with graphene sheets were investigated by Zhang et al.⁵³ According to their findings, the adsorption energy and structural analysis indicated that the doped Ti atom could greatly improve the interaction of these gas molecules with the graphene surface. They also declared that Ti-doped graphene can act as a selective gas absorber while N-doped graphene did not exhibit this capability apparently. In another work, Rojas and Leiva et al.⁵⁴ investigated the interaction between some small gas molecules and Ti-doped graphene sheets. They found that the Ti-doped graphene surface can effectively adsorb H₂ molecules. These results imply that Ti-doped graphene is more effective than N-doped graphene in detecting and removing gas molecules because of its high selectivity.

To date, numerous studies have been devoted to understanding the mechanism of CO oxidation or N₂O reduction over different surfaces.^11,23,55 Among these, supported noble metal catalysts like Pt^56,57 or Ag⁵⁸ exhibit outstanding performance, due to their large surface area and unique electronic properties. However, the problem is that these noble metal-based catalysts are expensive and have disadvantages in terms of toxicity and environmental consideration. Moreover, they usually require a high reaction temperature for a significant operation due to their high energy barrier. So, it is desirable to replace these costly catalysts with low cost materials for different industrial reactions at room temperature. Al-doped graphene has been widely studied as a low-cost metal-free catalyst for the oxidation of CO^51,59 and the dissociation of H₂O.⁶⁰ Recently, Jiang et al.⁵¹ indicated that Al-doped graphene can exhibit a high catalytic activity for the oxidation of CO with reaction barriers comparable with those on noble metal-doped graphene sheets. On the other hand, the Ti atom has been known as one of the promising hydrogen adsorbing materials for many years.^54,61 Also, it is found that Ti atoms with strong binding energy can be utilized as a functional material to uniformly coat graphene and they can strongly be adsorbed on the graphene sheet without diffusion.^62,63 Thus, it is expected that Ti-doped graphene may provide an important model to evaluate the catalytic activity of single Ti atoms on graphene sheets. Besides, it can be regarded as a reference model for developing non-metal atomic catalysts for N₂O oxidation.

In the present work, we explore and compare the geometry, electronic structure and catalytic properties of Al- and Ti-doped graphene (Al–/Ti–graphene) by means of density functional theory (DFT) calculations. It is important to mention that no previous study has been reported for the CO oxidation mechanism by N₂O over Al–/Ti–graphene. In addition, our results can be useful for better understanding the chemical properties of Al–/Ti–graphene and are worthy for the development of an automobile catalytic converter in order to clean up more than one gas by a single process.

2. Computational details

All of the geometry optimizations and frequency calculations were performed at the M06-2X/6-31G* level of theory using the Gaussian 09 suite of programs.⁶⁴ M06-2X is a global-hybrid meta-GGA functional which has been exclusively designed to treat medium range correlation energy, non-covalent interactions, main group thermochemistry, and reaction barriers.^65,66 In order to determine the minimum energy pathways, intrinsic reaction coordinate (IRC) calculations were also performed in the forward and reverse directions. A hexagonal graphene supercell (4 × 4 graphene unit cell), containing 48 carbon atoms, was chosen as the basic model for the calculations. Al– and Ti–graphene were then modelled by substituting a single Al or Ti atom with one C atom on the surface. The adsorption energies (E_ads) of the adsorbed molecules over Al–/Ti–graphene were calculated using the following equation:

E_ads(A) = E_A–M − E_M − E_A (3)

where E_A–M, E_M and E_A are the total energies of the adsorbate–substrate (A–M) system, the substrate (M), and adsorbate (A). Natural bond orbital (NBO) analysis⁶⁷ was performed at the M06-2X/6-31G* level using the NBO-5 program.⁶⁸

3. Results and discussion

3.1. Geometric structures of Al– and Ti–graphene

As noted above, the aim of the present theoretical study is to investigate the catalytic oxidation of CO molecules by N₂O over Al–/Ti–graphene. To this aim, the interaction of CO and N₂O molecules with these surfaces is investigated at the M06-2X/6-31G* level. Fig. 1 depicts the optimized structures of Al– and Ti–graphene. It is seen that after Al is embedded into graphene, the geometric structure of the graphene changes dramatically, where the Al atom projects out of the surface due to its larger size than C atoms and its preference to sp³ hybridization. Covalent Al–C bonds (≈1.84 Å) are formed in Al–graphene, that are quite a bit larger than the C–C bonds in the pure graphene sheet (1.42 Å).²³ These obtained Al–C bond lengths are larger than those theoretically reported by Ao et al. (1.63 Å),⁶⁹ but in good agreement with those mentioned by Wen and co-workers (1.85 Å).⁵¹ Note that the discrepancy between the calculated Al–C bond lengths in this study and those of ref. 69 can be attributed to the distinction between the used cluster and periodic slab models or to different DFT calculations. Meanwhile, about 0.3 e is transferred from the Al atom to the graphene sheet according to the NBO analysis, which may further confirm the strong interaction between Al and its nearest C atoms.

Fig. 1 Optimized structures of Al– and Ti–graphene. All bond distances are in Å.

Analogous with Al–graphene, when the Ti atom is replaced with one C atom of the graphene sheet, the Ti atom moves out of the plane with the average Ti–C bond length of 2.00 Å (Fig. 1), displacing also the positions of the first, second, and third out-of-plane neighbours. This value is smaller than that reported by Hu et al. (2.28 Å),⁶² but larger than that by Zhang (1.78 Å).⁷⁰ According to the NBO charge density analysis, a sizable charge of about 1.0 e is transferred from the Ti atom to its nearest C atoms, which may reveal the ionic nature of the Ti–C bonds.

It should be noted that the aggregation of metal atoms, such as Al or Ti, in high concentrations over the catalyst is a considerable problem.⁵¹ Therefore, to evaluate this possibility, the diffusion of Al and Ti atoms to their nearest position on the graphene sheets is studied. As Fig. S1 of the ESI† indicates, the calculated diffusion barrier is obtained to be about 2.95 and 3.42 eV for the Al and Ti atoms, respectively. On the other hand, the relatively large adsorption energies of these dopants over the vacancy site of the graphene guarantees that the Al and Ti dopants can disperse on graphene quite stably without a clustering problem. Thus, it is concluded that both Al– and Ti–graphene are stable enough to be utilized in the catalytic oxidation of CO by N₂O.

3.2. Adsorption of N₂O and CO molecules over Al– and Ti–graphene

First, we explore the adsorption of the isolated N₂O molecule on Al– and Ti–graphene. In order to find stable configurations of N₂O adsorbed over these surfaces, different adsorption sites and patterns (including side-on and end-on) were considered. Among these different adsorbed complexes, the most favourable configurations were chosen for the study of N₂O adsorption. The most stable adsorption complexes of N₂O over these surfaces along with their geometric values are shown in Fig. 2. Besides, Table 1 lists the corresponding binding distances, net charge transfers (q_CT), adsorption energies (E_ads) and thermodynamic parameters (ΔG₂₉₈ and ΔH₂₉₈) of these complexes.

Fig. 2 Optimized structures of adsorbed N₂O (from the O end (complexes A and C) and N end (complexes B and D)) and CO (complexes E and F) molecules over Al–/Ti–graphene along with their electron density difference (EDD) maps (±0.001 au). In the EDD maps, the charge depletion and accumulation sites are displayed in red and blue, respectively. All bond distances are in Å.

Table 1 Calculated binding distances (R), net charge transfer (q_CT), adsorption energy (E_ads), change of Gibbs free energy (ΔG₂₉₈) and change of enthalpy (ΔH₂₉₈) of the N₂O adsorption from its O end (complexes A and C), and N end (complexes B and D) over Al– and Ti–graphene

Complex	R (Å)	qCT (e)	Eads (eV)	ΔG298 (eV)	ΔH298 (eV)
Al–graphene
A	1.81	0.6	−0.81	−0.25	−0.78
B	1.91	0.6	−0.06	0.37	−0.04
E	2.20	0.2	−0.62	−0.19	−0.58
Ti–graphene
C	1.89	0.6	−3.05	−2.72	−3.06
D	2.00	0.5	−1.24	−0.81	−1.21
F	2.18	0.1	−1.03	−0.58	−0.98

---

The attachment of a single N₂O molecule to the Al–graphene sheet is initially considered by both the O and N end attacking in a cycloaddition configuration. In the case of the O end attacking, the N₂O molecule uses its O and N atoms to bind the Al and C atoms of Al–graphene, respectively, forming a five-membered Al–O–N–N–C ring (Fig. 2 (complex A)). The E_ads of this configuration is −0.81 eV which is larger than that reported by Lv et al. (−0.57 eV)⁷¹ as obtained at the PW91/6-31G** level. The difference may be due to the proper description of London dispersion effects with the M06-2X density functional, which is the main constituent part of the weak intermolecular interactions. The binding distance of Al–O is 1.81 Å, which is in the range of chemisorption and is larger than the Si–O bond in Si-embedded graphene (1.61 Å).²² Moreover, according to this adsorption, the O–N–N bond angle of N₂O is decreased significantly from 180° in the free N₂O to 117.6° in the adsorbed form. From the NBO analysis, a net charge of 0.6 e is transferred from the surface to the N₂O molecule which causes the elongation of the N–O bond length from 1.19 Å in the gas form to 1.34 Å in the adsorbed configuration. Due to the N–O bond elongation, the N₂O molecule is activated, therefore, its dissociation can occur easily. To investigate the electron density redistribution due to the adsorption of the N₂O molecule over Al–graphene, an electron density difference (EDD) plot is also provided (Fig. 2). As can be seen, the blue colour in the EDD plot clearly indicates that there is a sizable electron density accumulation between the O atom of the N₂O molecule and the Al atom of the surface, which confirms the chemisorption nature of the adsorption. It can be seen from thermochemistry results that the formation of complex A is exothermic (ΔH₂₉₈ = −0.78 eV) with a negative ΔG₂₉₈ value of about −0.25 eV (Table 1). Another stable configuration is achieved when the N₂O molecule is attached to the Al atom by its end N atom in which it is placed in a tilted position over the Al–graphene sheet (complex B). In this case, compared to complex A, the N₂O molecule is adsorbed weakly (E_ads = −0.06 eV) on Al–graphene while the N atom is about 1.91 Å away from the Al atom (Table 1 and Fig. 2). Herein, the N–N–O angle and the N–O bond length are 139° and 1.22 Å, respectively, which are 41° and 0.01 Å smaller than those in the isolated state. The EDD plot confirms the great charge accumulation around the Al–N bond while the charge is depleted between the O atom of the N₂O molecule and the C atom of the Al–graphene surface. Additionally, a total charge of about 0.6 e is transferred from the Al atom to the N atom of the N₂O molecule. As is also evident from Table 1, the adsorption process to form complex B is an exothermic reaction (ΔH₂₉₈ = −0.04 eV), with a positive ΔG₂₉₈ value of 0.37 eV.

Continuously, the N₂O adsorption over Ti–graphene is also studied and the corresponding favourable adsorption structures are depicted in Fig. 2. Analogous with Al–graphene, here, the N₂O molecule can be attached to the Ti atom of the surface via its O or N end and forms two different adsorption configurations. When the N₂O molecule approaches from its O site to the Ti atom, the dissociative complex C is achieved. In this configuration, upon the adsorption of the N₂O molecule, it is easily dissociated into N₂ and activated atomic oxygen (O_ads) species. In addition, the N₂O molecule has a larger E_ads value (−3.05 eV) on Ti–graphene than on Al–graphene. Thus, it is evident that compared to the Al atom, the Ti atom in Ti–graphene plays a more important role in the adsorption and activation of the N₂O molecule, which is most likely due to its large polarization and charge transfer effects. The binding distance between the remaining O_ads and Ti atom is 1.89 Å and the N₂ molecule is placed about 2.64 Å away from it. It is of particular interest that the physisorbed N₂ molecule (Ti–N = 2.88 Å) can desorb easily from the surface and the activated O_ads remains over the Ti atom. As is evident from the EDD map, there is a significant build up of electron density around the Ti–O_ads bond, which indicates the strong binding of O_ads to the Ti atom. In addition, the total net charge transfer from the Ti atom to O_ads (0.6 e) is another proof for the strong interaction between the Ti and O_ads atoms. It is interesting to know that according to NBO atomic charges, a major negative charge (−0.5 e) is on O_ads and only 0.05 e is accumulated over the N₂ molecule. The formation of this complex is extremely exothermic (ΔH = −3.06 eV) and has a large (more negative) ΔG₂₉₈ value which reveals its easy formation process in ambient conditions (Table 1). Complex D is the other possible configuration considered for the adsorption of N₂O molecules over Ti–graphene. In this configuration, the gas molecule adsorbs via its N atom over the Ti atom of the surface while it is completely in a vertical position right above the Ti atom of the surface. The Ti–N bond length and E_ads are about 2.00 Å and −1.24 eV, respectively. It can be seen from Fig. 2 that the N₂O molecule is weakly adsorbed over the surface and the EDD map also verifies this fact. Table 1 indicates that a net charge of 0.5 e is transferred from the surface to the N₂O molecule. It is noticeable that the main negative charge (−0.5 e) is accumulated over the N atom that is directly bonded to the Ti atom of Ti–graphene. Like complex C, the formation of this complex is exothermic and is spontaneous at normal temperatures (Table 1).

On the other hand, seeking the most stable configurations of a single CO molecule adsorbed on Al–/Ti–graphene leads to the characterization of different possible adsorption complexes. The most stable configurations of CO adsorption over Al–/Ti–graphene are displayed in Fig. 2 (complexes E and F). Also, the corresponding binding distance, q_CT, E_ads, ΔH₂₉₈ and ΔG₂₉₈ values are summarized in Table 1. It is found that the CO molecule in the end-on position, in which the C atom of the CO molecule directly bonds to the Al or Ti atom of the surface, is the most energetically probable adsorption configuration. In these structures, the CO molecule is weakly adsorbed over the surface with the C–Al (C–Ti) bond length of 2.20 (2.18) Å. The corresponding E_ads value of the adsorbed CO over Al– and Ti–graphene is −0.62 and −1.03 eV, respectively. One can see from the Table 1 results that analogous with the N₂O adsorption, the absolute value of E_ads for the CO molecule over Ti–graphene is larger than that of Al–graphene. This can be related to the higher catalytic activity of Ti–graphene than that of Al–graphene. Table 1 shows that in complexes E and F, a net charge of about 0.2 (0.04) e is transferred from the Al (Ti) atom to the 2π* orbitals of the CO molecule. The EDD map can verify the physical adsorption of the CO molecule over Al– and Ti–graphene (Fig. 2), where there exists a small electron density accumulation region around the Al–C or Ti–C bond. Generally, although the absolute values of the thermodynamic parameters (ΔH₂₉₈, ΔG₂₉₈) of the adsorbed CO over Ti–graphene are larger than those of Al–graphene, totally, the formation of both these complexes is exothermic and they are possible at room temperature (Table 1).

In summary, comparing the E_ads values of the adsorbed N₂O and CO molecules over Al–/Ti–graphene demonstrates that the N₂O molecule binds more strongly than CO, so the doped graphene will be probably covered with the N₂O species. Also, the N₂O molecule can be easily decomposed into N₂ and O_ads species over the Ti–graphene sheet. Hence, it can be proposed that Ti–graphene might be a better catalyst for N₂O decomposition than Al–graphene, which can be attributed to its larger polarization and charge transfer effects.

3.3. Proposed reaction pathways for CO oxidation over Al–/Ti–graphene

CO oxidation by N₂O molecules is proposed in step-wise reactions, which are the decomposition of N₂O followed by CO oxidation; i.e. N₂O → N₂ + O_ads and O_ads + CO → CO₂. In order to investigate the reaction mechanisms of CO oxidation by N₂O molecules over Al–graphene, we start from the favourable configuration A. The configurations of the initial state (IS), transition state (TS) and final state (FS) with their related geometric parameters and the reaction pathway are shown in Fig. 3. Additionally, the corresponding activation energy (E_act), ΔH₂₉₈ and ΔG₂₉₈ of these elementary reactions are reported in Table 2. As mentioned before, the stable complex A is chosen as the initial state (IS-1) in which the N₂O molecule forms a five-membered ring over the Al–graphene sheet. In this configuration the N₂ and O_ads species are produced passing through the transition state TS-1 with an activation energy of 0.24 eV, which is smaller than those of noble metal catalysts⁷² or other metal-doped graphene sheets.^11,71–73 This small E_act value shows that the N₂O decomposition over Al–graphene is achieved quite easily. In TS-1, the Al–O and O–N bond lengths are increased to 1.85 and 2.21 Å, respectively. The N₂ molecule is finally produced in FS-1 with the N≡N bond length of 1.11 Å while the Al–O bond is significantly decreased from 1.85 Å in TS-1 to 1.77 Å in FS-1. In this configuration, the N₂ molecule and atomic oxygen atom (O_ads) are completely separated from each other (O_ads–N = 3.08 Å) and O_ads remains over the Al atom of the surface to react with the CO molecule. It is noteworthy that the reaction step IS-1 → FS-1 is exothermic with a negative ΔH₂₉₈ value and can be performed at ambient conditions (Table 2). Besides, the adsorption energy of O_ads over Al–graphene is about −6.2 eV, which may indicate its strong interaction with the Al atom. In the next step (O_ads + CO → CO₂), the CO molecule can be oxidized by the activated O_ads atom. Although the CO molecule can approach the Al atom of Al–graphene in different co-adsorption configurations, the reaction starts with the stable configuration as IS-2 (Fig. 3). In this configuration, the CO molecule is located about 2.48 Å away from O_ads while it is a little tilted to the surface. In TS-2, the distance between the C atom of the CO molecule and O_ads is reduced to 2.03 Å. Also, the Al–O_ads bond length is elongated from 1.39 Å in IS-2 to 1.78 Å in TS-2, in order to facilitate the formation of the CO₂ molecule in FS-2. In FS-2, the CO₂ molecule is formed about 2.10 Å away from the Al atom and easily desorbs from the surface due to its small E_ads of about −0.52 eV. The barrier energy of the IS-2 → FS-2 step is very small (0.06 eV) with the reaction energy of −2.21 eV. Note that the activation energy obtained for the oxidation of CO + O_ads in the present study is smaller than those of noble metal catalysts such as Au-doped graphene⁷⁴ or those using normal transition metals (e.g. Cu, Fe).^34,75 Besides, as the results of Table 2 indicate, the reaction IS-2 → FS-2 has a negative value of ΔH₂₉₈ and ΔG₂₉₈ which confirms the feasibility of this reaction at room temperature (Table 2).

Fig. 3 Schematic energy profile corresponding to local configurations along (a) N₂O → N₂ + O_ads and (b) O_ads + CO → CO₂ reaction pathways over the Al–/Ti–graphene sheet. All energies and bond distances are in eV and Å, respectively.

Table 2 Calculated activation energy (E_act), reaction energy (ΔE), change of Gibbs free energy (ΔG₂₉₈) and change of enthalpy (ΔH₂₉₈) for different pathways of CO oxidation by N₂O molecules over Al– and Ti–graphene sheets

Reaction	Eact (eV)	ΔE (eV)	ΔG298 (eV)	ΔH298 (eV)
Al–graphene
IS-1 → P-1	0.24	0.02	−0.02	0.03
IS-2 → P-2	0.06	−2.21	−2.19	−2.24
Ti–graphene
IS-3 → P-3	0.16	−1.18	−1.90	−1.23

---

Furthermore, the CO oxidation reaction over Ti–graphene is also studied. Like Al–graphene, a two-step reaction pathway is also taken into account for the CO oxidation reaction over this surface. However, as mentioned before, the N₂O molecule in complex C is dissociated into N₂ and O_ads species. In other words, we can say that the C → O_ads + N₂ step occurs as a barrier-less reaction. In the second step, the ability of O_ads toward the oxidation of CO molecules over Ti–graphene is investigated. It should be noted that the dissociative complex C is chosen as the final state in which the physisorbed N₂ molecule can easily desorb, leaving O_ads with an adsorption energy of −7.3 eV over the surface. The NBO analysis also reveals that the charge of the O_ads atom on the surface of Ti–graphene (−0.73 e) is more negative than that of Al–graphene (−0.62 e), due to the large polarization of the Ti atom. This may provide a basis for the higher catalytic performance of Ti–graphene compared to that of Al–graphene. Upon desorption of N₂ molecules, different co-adsorption configurations are examined from which the most favourable configuration is selected as the initial state (IS-3). In this complex, the CO molecule is placed far from O_ads with the C–O_ads and Ti–O_ads bond distances of 2.71 and 1.87 Å, respectively (Fig. 3). The CO molecule reaches and reacts with O_ads in TS-3 with the activation energy of 0.16 eV. This value is smaller than that of Pt–graphene (1.02 eV),⁷⁶ P–graphene (1.1 eV) and N–graphene (0.53 eV),⁴² which suggests that Ti–graphene can be regarded as a promising candidate for the low-temperature oxidation of CO. The Ti–O_ads bond is increased from 1.87 Å in IS-3 to 1.90 Å in TS-3 to facilitate the dissociation of O_ads from the Ti atom. Also, the CO molecule gets closer (C–O_ads = 1.96 Å) to the atomic O_ads in order to react and form the CO₂ molecule. Passing from TS-3, FS-3 is finally obtained and the physisorbed CO₂ molecule is easily released from the surface with a small E_ads of about −0.32 eV. It can be seen from Table 2 that the reaction step IS-3 → FS-3 has a negative ΔH₂₉₈ and ΔG₂₉₈ of −1.23 and −1.90 eV, respectively, which confirms the exothermic nature of this reaction and also the thermodynamically probable situation at ambient conditions.

Another reaction path which involves the reaction of the O_ads moiety with a N₂O molecule to produce N₂ and O₂ molecules is also studied. Fig. S2† shows the decomposition process of the second N₂O molecule adsorbed on the O_ads–Al–/Ti–graphene sheets. As is evident from Fig. S2,† the decomposition of N₂O molecules over O_ads–Al–graphene and O_ads–Ti–graphene starts from the adsorption configurations IS-3 and IS-4, respectively. The calculated adsorption energy for the N₂O molecule over O_ads–Al–graphene and O_ads–Ti–graphene is −0.26 and −0.33 eV, respectively. The NBO analysis results also indicate a small amount of charge transfer from N₂O to the graphene surface. Meanwhile, the adsorption of the second N₂O molecule tends to elongate the Al–O_ads or Ti–O_ads bond (Fig. S2†), which reveals the activation of the O_ads moiety in the presence of the N₂O molecule. In the transition state, the oxygen atom of N₂O approaches the O_ads moiety and the O–N bond of N₂O is slightly elongated. The reaction is then continuously proceeded from the transition state to the final state where the O₂ molecule is formed, leaving the N₂ molecule physisorbed over the Al–/Ti–graphene system with the E_ads value of −0.06 and −0.08 eV, respectively. The calculated activation energies for this reaction step are calculated to be 2.43 and 2.51 eV for Al– and Ti–graphene, respectively. These are consistent with the previous theoretical study reported by Song et al.,³⁶ indicating that the N₂O + O_ads → N₂ + O₂ reaction is unlikely to take place over both Al– and Ti–graphene sheets under normal conditions.

Based on these results, we can predict the feasibility of using Al– and Ti–graphene in practical applications. According to our findings, it is found that both Al– and Ti–graphene have an outstanding catalytic activity toward the oxidation of CO by N₂O molecules. Compared with Al–graphene, the decomposition of N₂O over Ti–graphene can occur rapidly at room temperature due to its barrier-less reaction (C → O_ads + N₂). Although E_act of the second process of IS-2 → FS-2 in Al–graphene is smaller than that of IS-3 → FS-3 in Ti–graphene, the latter is preferred because the dissociative process of the N₂O molecule takes place automatically via the negligible barrier energy. Besides, the relatively large activation energies obtained for the decomposition of N₂O over O_ads–Al–/Ti–graphene indicate the impossibility of this side reaction over these systems. Generally, these results can be helpful for expanding efficient noble metal-free catalysts toward the oxidation of CO based on graphene.

4. Conclusions

In this study, the catalytic properties of Al– and Ti–graphene for the oxidation of CO by N₂O molecules were investigated by means of DFT calculations. The equilibrium geometries of Al–/Ti–graphene sheets along with the N₂O adsorption and dissociation configurations over them were studied in detail. Comparing the adsorption of N₂O and CO molecules over Al–/Ti–graphene indicated that the N₂O molecule is a stronger electron acceptor than the CO molecule. So, in the presence of both N₂O and CO molecules as reaction gases, the Al or Ti site of the doped graphene is covered mainly by N₂O molecules. Unlike Al–graphene, upon the adsorption of N₂O molecules over Ti–graphene, it is quickly dissociated to N₂ and O_ads species via the barrier-less reaction pathway C → N₂ + O_ads. Then, the activated O_ads goes through the second reaction (O_ads + CO → CO₂) in order to form CO₂ molecules with the E_act of about 0.16 eV. It can be concluded that Ti–graphene has a greater catalytic activity than Al–graphene for CO oxidation by N₂O molecules. Moreover, the decomposition of the second N₂O molecule is unlikely to take place over O_ads–Al–graphene and O_ads–Ti–graphene sheets due to the calculated large activation energies. Overall, both Al– and Ti–graphene can be considered as promising non-noble metal catalysts for the low-temperature CO oxidation reaction by N₂O molecules.

References

1. I. Blumenthal, J. R. Soc. Med., 2001, 94, 270–272.
2. N. Lopez and J. K. Nørskov, J. Am. Chem. Soc., 2002, 124, 11262–11263.
3. L. D. Socaciu, J. Hagen, T. M. Bernhardt, L. Wöste, U. Heiz, H. Häkkinen and U. Landman, J. Am. Chem. Soc., 2003, 125, 10437–10445.
4. M. Chen, Y. Cai, Z. Yan, K. Gath, S. Axnanda and D. W. Goodman, Surf. Sci., 2007, 601, 5326–5331.
5. D.-J. Liu, J. Phys. Chem. C, 2007, 111, 14698–14706.
6. L. Liu, F. Zhou, L. Wang, X. Qi, F. Shi and Y. Deng, J. Catal., 2010, 274, 1–10.
7. M. D. Esrafili and N. Saeidi, Superlattices Microstruct., 2015, 81, 7–15.
8. M. D. Esrafili and N. Saeidi, Phys. E, 2015, 74, 382–387.
9. M. D. Esrafili, P. Nematollahi and R. Nurazar, Superlattices Microstruct., 2016, 92, 60–67.
10. G. Walther, D. J. Mowbray, T. Jiang, G. Jones, S. Jensen, U. J. Quaade and S. Horch, J. Catal., 2008, 260, 86–92.
11. S. Wannakao, T. Nongnual, P. Khongpracha, T. Maihom and J. Limtrakul, J. Phys. Chem. C, 2012, 116, 16992–16998.
12. M. D. Esrafili, N. Saeidi and P. Nematollahi, RSC Adv., 2015, 5, 100290–100298.
13. H. Loirat, F. Caralp, M. Destriau and R. Lesclaux, J. Phys. Chem., 1987, 91, 6538–6542.
14. A. Gluhoi, M. Dekkers and B. Nieuwenhuys, J. Catal., 2003, 219, 197–205.
15. O. P. Balaj, I. Balteanu, T. T. Roßteuscher, M. K. Beyer and V. E. Bondybey, Angew. Chem., Int. Ed., 2004, 43, 6519–6522.
16. F. Rondinelli, N. Russo and M. Toscano, J. Chem. Theory Comput., 2008, 4, 1886–1890.
17. L. Zhao, Z. Liu, W. Guo, L. Zhang, F. Zhang, H. Zhu and H. Shan, Phys. Chem. Chem. Phys., 2009, 11, 4219–4229.
18. T. Baidya, A. Marimuthu, M. Hegde, N. Ravishankar and G. Madras, J. Phys. Chem. C, 2007, 111, 830–839.
19. C.-K. Siu, S. Reitmeier, I. Balteanu, V. Bondybey and M. Beyer, Eur. Phys. J. D, 2007, 43, 189–192.
20. X. L. Xu, E. Yang, J. Q. Li, Y. Li and W. K. Chen, ChemCatChem, 2009, 1, 384–392.
21. K. Kartha, M. Pai, A. Banerjee, R. Pai, S. Meena and S. Bharadwaj, J. Mol. Catal. A: Chem., 2011, 335, 158–168.
22. Y. Chen, B. Gao, J.-X. Zhao, Q.-H. Cai and H.-G. Fu, J. Mol. Model., 2012, 18, 2043–2054.
23. J.-X. Zhao, Y. Chen and H.-G. Fu, Theor. Chem. Acc., 2012, 131, 1242.
24. S. Lin, X. Ye and J. Huang, Phys. Chem. Chem. Phys., 2015, 17, 888–895.
25. E. Kan, Z. Li and J. Yang, Nano, 2008, 3, 433–442.
26. A. K. Geim, Science, 2009, 324, 1530–1534.
27. W. Choi, I. Lahiri, R. Seelaboyina and Y. S. Kang, Crit. Rev. Solid State Mater. Sci., 2010, 35, 52–71.
28. L. L. Zhang, R. Zhou and X. Zhao, J. Mater. Chem., 2010, 20, 5983–5992.
29. K. R. Ratinac, W. Yang, S. P. Ringer and F. Braet, Environ. Sci. Technol., 2010, 44, 1167–1176.
30. M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2009, 110, 132–145.
31. B. Guo, L. Fang, B. Zhang and J. R. Gong, Insci. J., 2011, 1, 80–89.
32. T. Wehling, K. Novoselov, S. Morozov, E. Vdovin, M. Katsnelson, A. Geim and A. Lichtenstein, Nano Lett., 2008, 8, 173–177.
33. M. Zhou, Y. H. Lu, Y. Q. Cai, C. Zhang and Y. P. Feng, Nanotechnology, 2011, 22, 385502.
34. Y. Li, Z. Zhou, G. Yu, W. Chen and Z. Chen, J. Phys. Chem. C, 2010, 114, 6250–6254.
35. H.-P. Zhang, X.-G. Luo, H.-T. Song, X.-Y. Lin, X. Lu and Y. Tang, Appl. Surf. Sci., 2014, 317, 511–516.
36. E. H. Song, J. M. Yan, J. S. Lian and Q. Jiang, J. Phys. Chem. C, 2012, 116, 20342–20348.
37. M. Wu, C. Cao and J. Jiang, New J. Phys., 2010, 12, 063020.
38. E. H. Song, Z. Wen and Q. Jiang, J. Phys. Chem. C, 2011, 115, 3678–3683.
39. L. Ma, J.-M. Zhang, K.-W. Xu and V. Ji, Appl. Surf. Sci., 2015, 343, 121–127.
40. L. Zhang and Z. Xia, J. Phys. Chem. C, 2011, 115, 11170–11176.
41. L. Zhang, J. Niu, L. Dai and Z. Xia, Langmuir, 2012, 28, 7542–7550.
42. M. D. Esrafili, R. Mohammad-Valipour, S. M. Mousavi-Khoshdel and P. Nematollahi, ChemPhysChem, 2015, 16, 3719–3727.
43. Y. Tang, Z. Liu, X. Dai, Z. Yang, W. Chen, D. Ma and Z. Lu, Appl. Surf. Sci., 2014, 308, 402–407.
44. M. D. Esrafili, R. Nurazar and E. Vessally, Int. J. Quantum Chem., 2015, 115, 1153–1160.
45. M. D. Esrafili, N. Saeidi and P. Nematollahi, Chem. Phys. Lett., 2016, 649, 37–43.
46. L. Wang, Q. Luo, W. Zhang and J. Yang, Int. J. Hydrogen Energy, 2014, 39, 20190–20196.
47. Z. Ao, S. Li and Q. Jiang, Solid State Commun., 2010, 150, 680–683.
48. M. Chi and Y.-P. Zhao, Comput. Mater. Sci., 2009, 46, 1085–1090.
49. Y. Sun, L. Chen, F. Zhang, D. Li, H. Pan and J. Ye, Solid State Commun., 2010, 150, 1906–1910.
50. W. Zhao and Q. Y. Meng, Adv. Mater. Res., 2013, 602, 870–873.
51. Q. Jiang, Z. Ao, S. Li and Z. Wen, RSC Adv., 2014, 4, 20290–20296.
52. J. Dai, J. Yuan and P. Giannozzi, Appl. Phys. Lett., 2009, 95, 232105.
53. H.-P. Zhang, X.-G. Luo, X.-Y. Lin, X. Lu, Y. Leng and H.-T. Song, Appl. Surf. Sci., 2013, 283, 559–565.
54. M. Rojas and E. Leiva, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 155415.
55. R. Gholizadeh and Y.-X. Yu, Appl. Surf. Sci., 2015, 357, 1187–1195.
56. R. R. Sadhankar and D. T. Lynch, J. Catal., 1994, 149, 278–291.
57. J. Arenas-Alatorre, A. Gómez-Cortés, M. Avalos-Borja and G. Díaz, J. Phys. Chem. B, 2005, 109, 2371–2376.
58. P. Giese, H. Kirsch, M. Wolf and C. Frischkorn, J. Phys. Chem. C, 2011, 115, 10012–10018.
59. M. D. Esrafili, P. Nematollahi and H. Abdollahpour, Appl. Surf. Sci., 2016, 378, 418–425.
60. Q. G. Jiang, Z. M. Ao and Q. Jiang, Phys. Chem. Chem. Phys., 2013, 15, 10859–10865.
61. S. Chu, L. Hu, X. Hu, M. Yang and J. Deng, Int. J. Hydrogen Energy, 2011, 36, 12324–12328.
62. L. Hu, X. Hu, X. Wu, C. Du, Y. Dai and J. Deng, Phys. B, 2010, 405, 3337–3341.
63. Y. Okamoto, Chem. Phys. Lett., 2006, 420, 382–386.
64. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian, Inc, Wallingford CT, 2009.
65. Y. Zhao and D. G. Truhlar, J. Am. Chem. Soc., 2007, 129, 8440–8442.
66. M. D. Esrafili and R. Nurazar, Comput. Mater. Sci., 2014, 92, 172–177.
67. A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926.
68. E. Glendening, J. Badenhoop, A. Reed, J. Carpenter, J. Bohmann, C. Morales and F. Weinhold, NBO 5.0, Theoretical Chemistry Institute, University of Wisconsin, Madison, 2004.
69. Z. Ao, J. Yang, S. Li and Q. Jiang, Chem. Phys. Lett., 2008, 461, 276–279.
70. H.-P. Zhang, X.-G. Luo, X.-Y. Lin, X. Lu and Y. Leng, Int. J. Hydrogen Energy, 2013, 38, 14269–14275.
71. Y.-A. Lv, G.-L. Zhuang, J.-G. Wang, Y.-B. Jia and Q. Xie, Phys. Chem. Chem. Phys., 2011, 13, 12472–12477.
72. V. Tzitzios and V. Georgakilas, Chemosphere, 2005, 59, 887–891.
73. E. Song, J. Yan, J. Lian and Q. Jiang, J. Phys. Chem. C, 2012, 116, 20342–20348.
74. Y.-H. Lu, M. Zhou, C. Zhang and Y.-P. Feng, J. Phys. Chem. C, 2009, 113, 20156–20160.
75. E. Song, Z. Wen and Q. Jiang, J. Phys. Chem. C, 2011, 115, 3678–3683.
76. Y. Tang, Z. Yang and X. Dai, Phys. Chem. Chem. Phys., 2012, 14, 16566–16572.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04326g

---