Photocatalytic activity of carbon nanotube/Ag₃PO₄ hybrid from first-principles study

${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ is an excellent visible light photocatalyst with high photooxidative capability [1]. It is used to degrade organic contaminants and serves as a photofunctional material for wastewater cleaning [2]. This photocatalyst has high quantum efficiency under visible light irradiation, but, it is unstable under irradiation [3]. Numerous methods have been explored for improving and increasing the stability and photocatalytic activity of this photocatalyst. This includes the combination of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ with different materials, including ${\rm Sn}{{{\rm O}}_{2}}$ [4], AgX (X  =  Cl, Br and I) [5], ${\rm Ti}{{{\rm O}}_{2}}$ [6], ${\rm F}{{{\rm e}}_{3}}{{{\rm O}}_{4}}$ [7] and GO [8, 9].

The ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}/g-{{{\rm C}}_{3}}{{{\rm N}}_{4}}$ composite was synthesized for water oxidation [10], oxygen production and pollutant degradation [11]. Also, $g-{{{\rm C}}_{3}}{{{\rm N}}_{4}}$ nanorod/${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ composites have been considered as one of the most effective techniques for achieving the conversion of clean and sustainable sunlight to solar fuel [12]. The bifunctional ${\rm Ti}{{{\rm O}}_{{\rm 2}}}{\rm /A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$/graphene (GR) composites exhibited highly efficient visible light photocatalytic activity toward organic dye molecule degradation and showed excellent bactericidal performance [13]. Under LED illumination, ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$/Ag/graphene/graphitic carbon nitride ($g-{{{\rm C}}_{3}}{{{\rm N}}_{4}}$) hetero structured materials can drive photocatalytic water oxidation efficiently [14].

As a result of the exceptional structure and properties of CNT, it could be used as a dopant for improving the photocatalytic degradation efficiency. The electronic properties of a CNT prepare continuous electronic states in the conduction band (CB) for donating the transferring electrons [15].

The capabilities of charge transfer of CNT can promote the excited electron in the conduction band of the semiconductor to drift into the CNTs, thereby decreasing the ability of electron–hole pairs to recombine [16], and increasing the photocatalytic activity under visible light. Some composite materials have been proven to be effective, such as CNT/ZnO [17], ${\rm CNT/}{{{\rm C}}_{3}}{{{\rm N}}_{4}}$ [18] and ${\rm CNT/Ti}{{{\rm O}}_{2}}$ [19, 20]. CNT could be used as electron capture agents to increase the activity and stability of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$. The photocatalytic activity of ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ has been investigated with Rhodamine B (RhB) as a model contaminant by experimental method. The experimental results showed that ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ displayed much higher photocatalytic activity than the pure ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ [16]. The chemical bonds according to covalent interaction at the interface are supposed to be charge transfer channels [21] and the Van der Waals (vdW) forces between CNTs and the semiconductor were also revealed at the interfaces [22], thereby increasing the photocatalytic activity of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrids. Various mechanisms have been suggested for increasing the photocatalytic properties of ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrids. One is that in the time of photocatalysis, a high-energy photon stimulates an electron from the valence band (VB) to the CB of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$, and the generated electrons formed in the space-charge districts are carried into the CNTs, thereafter holes stay on ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ to take part in redox reactions [23, 24]. In this work, the interaction between ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ and SWCNT (6, 0) was investigated using large-scale density functional theory (DFT) computations to disclose the enhanced photocatalytic performance.

The most stable ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ surface, cubic ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surface and metallic CNT (6, 0) were selected. The theoretical calculations were performed using the plane wave pseudopotential DFT method, as implemented in the CASTEP code [25]. The generalized gradient approximation (GGA) was used to describe the exchange and correlation energy of the electrons [26]. Geometry optimizations and single-point energy calculation were performed. The calculated supercells consisted of a ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surface which contained 32 O, 8 P and 24 Ag atoms and the (6, 0) tube which contained 108 atoms, length of $12.8{{\rm }\mathring{\rm A} }$ in its axial direction.

The experimental lattice parameter $a=6.013{{\rm }\mathring{\rm A} }$ of cubic ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ with space group $P4-3n$ (NO. 218) was taken to get the crystal cell [27]. By minimizing the total crystal energy, the equilibrium lattice parameter was calculated using the GGA method and the results are presented in table 1 with the experimental values. The lattice parameter includes $a=6.004\,{{\rm }\mathring{\rm A} }$, which is slightly overestimated less than the experimental values. There is a strong correlation between structural parameters and the experimental values.

Table 1. Structure parameters of Ag₃PO₄: lattice parameters and atom positions.

			Bond length/${{\rm }\mathring{\rm A} }$		Bond angle (deg)			Fractional coordinates
Method	$a/{{\rm }\mathring{\rm A} }$	${{E}_{g}}/{\rm eV}$	P-O	Ag-O	O-P-O	O-Ag-O		$x$	$y$	$z$
Exp [22].	6.013	2.45	1.539	2.345	109.470	93.660	Ag	0.231	0.000	0.500
			1.539	2.403			P	0.000	0.000	0.000
							O	0.148	0.148	0.148
GGA-PBE	6.004	0.285	1.549	2.364	109.307	92.995	Ag	0.250	0.000	0.500
			1.557	2.400			P	0.000	0.000	0.000
							O	0.150	0.150	0.150

For the noncovalent hybrid, the equilibrium distances between the CNTs and the top layer of the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surface after optimization was calculated to be $2.450\,{{\rm }\mathring{\rm A} }$ for ${\rm CNT}\left(6, 0 \right)/{\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$. A side view of the hybrid of CNT and the cubic ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surface is shown in figure 1. The strength of the ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid can be evaluated by their formation energy, which is defined as:

$$\begin{align} \newcommand{\e}{{\rm e}} \displaystyle {{E}_{{\rm formation}}}={{E}_{{\rm hybrid}}}-{{E}_{{\rm CNT}}}-{{E}_{{\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}}}\nonumber \end{align}$$

where ${{E}_{{\rm hybrid}}},\,{{E}_{{\rm CNT}}}$ and ${{E}_{{\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}}}$ display the total energy of ${\rm CNT}/{\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100), pure CNT, and clean ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surfaces, respectively. By this definition, negative ${{E}_{{\rm formation}}}$ shows that the interface is steady and stable. The interface formation energy was calculated to be $-1.020\,{\rm eV}$, indicating a rather strong interaction between CNTs and the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100) surface, and the thermodynamic stability of this hybrid.

Zoom In Zoom Out Reset image size

The density of states (DOSs) of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ and CNT, before and after the formation of the hybrid, was calculated so as to determine the effect of the type of interfacial interaction on the electronic properties of hybrid.

Pure ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ is an indirect semiconductor with a band gap $\left({{E}_{g}} \right)$ of 2.45 eV and its CB bottom is very diffusive, causing smaller useful masses of the photogenerated electrons in pure ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ [1]. While the surface of ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ is exposed, ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ becomes a direct band gap semiconductor and ${{E}_{g}}$ declines to $2.15{\rm eV}$ [28].

The DOS of pure ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ is presented in figure 2. Also, the DOS of single metallic (6, 0) CNT showed $0{\rm eV}$ band gap, which corroborates previous studies [22]. Figure 3 shows that the DOS of (6, 0) ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid changes a little from individuals. The interaction between the metallic (6, 0) CNTs and ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ may interact via a noncovalent vdWs force. The band gap of ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid is small $\left(0.16<0.3{\rm eV} \right)$ mentioning that the hybrid can absorb the most sunlight, thus raising their photocatalytic activity. For the noncovalent ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid, the CB bottom is formed mainly from Ag $5s$ orbitals, which can be more clearly observed from the electron density distributions of the lowest-unoccupied level (LUL), while the HOL is only composed of the C $2p$ orbitals.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The interaction in CNT (6, 0) and ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid depends on the nature of CNTs and their relative orientations. Metallic CNT (6, 0) may physically interact with the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ (100). In the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrid, the band gap is small which causes the CNT/Ag₃PO₄ hybrid to absorb sunlight from the ultraviolet to the infrared region. Moreover, CNTs are not only effective sensitizers, but are also highly active co-catalysts in hybrids. This study is useful for developing highly efficient carbon-based nanophotocatalysts. During photocatalysis, the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ can be excited to yield photon-generated carriers and photoinduced electrons from the valence band (VB) are transferred to the conduction band (CB), leaving the holes in the VB. In hybrid of the ${\rm A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ and CNT, the photo-generated electrons are effectual trapped by CNT. The CNT can be increased photocatalytic activity and stability due to the fact that the introduction of CNT significantly improves the separation of photogenerated charge carriers. Thus, electrons and holes could be usefully divided so that the photocatalytic performance of the catalyst would be modified. The optical absorption of ${\rm CNT/A}{{{\rm g}}_{3}}{\rm P}{{{\rm O}}_{4}}$ hybrids in the visible-light region can be greatly increased owing to their small band gap. Fascinatingly, it was realized that the CNTs are not only effective sensitizers, but also highly active co-catalysts in hybrids.

I gratefully thank Payame Noor University for financial support.