† Corresponding author. E-mail:
Project supported by the National Natural Science Foundation of China (Grant No. 51172067), the Hunan Provincial Natural Science Fund for Distinguished Young Scholars, China (Grant No. 13JJ1013), the Specialized Research Fund for the Doctoral Program of Higher Education, China (Grant No. 20130161110036), and the New Century Excellent Talents in University, China (Grant No. NCET-12-0171.D).
We calculate the electronic properties and carrier mobility of perovskite CH3NH3SnI3 as a solar cell absorber by using the hybrid functional method. The calculated result shows that the electron and hole mobilities have anisotropies with a large magnitude of 1.4 × 104 cm2·V−1·s−1 along the y direction. In view of the huge difference between hole and electron mobilities, the perovskite CH3NH3SnI3 can be considered as a p-type semiconductor. We also discover a relationship between the effective mass anisotropy and electronic occupation anisotropy. The above results can provide reliable guidance for its experimental applications in electronics and optoelectronics.
The recent emergence of halide perovskites as light harvesters and transport materials has revolutionized the scenario of emerging photovoltaic technologies. Organic–inorganic hybrid perovskite based on metal halides adopts the ABX3 perovskite structure (B cation = Sn2+ or Pb2+; X is typically Br1 −, or I1 −; and the A cation is selected to balance the total charge and it is a small organic cation) solar cells are revolutionizing the field because of the potential for high efficiency and long dispersion length.[1–5] Therefore, in order to realize the sustainable development of energy and commercialization of technology, it is necessary to reach analogous optical and photovoltaic performance developing environmentally friendly solar cells. As is well known, lead poisoning is a type of metal poisoning and does harm to both the human body and the natural environment.[6] Motivated by non-toxic atoms of Sn in the same group compared with Pb, one substituted Sn2+ for Pb2+. However, a recent study[7] has reported that the strong acidification induced by the SnI2 is found to be more harmful than the combined effect of milder acidification and the expected Pb intoxication induced by PbI2, indicating that Sn-based perovskite may not be the ideal Pb surrogate. Fortunately, our experimental group has substituted Ge for Pb and Sn, considering this problem, tuning the superior solar cell performance of effective mass and absorption in perovskite CH3NH3GeCl3 under hydrostatic pressure.[8] Recent implementation of CH3NH3SnI3 (X = Cl, Br, I) perovskite absorbers enabled power conversion efficiency of 12%–15%, which has been recognized as “The Next Big Thing in Photovoltaics”. This makes them a possible light absorber in energy conversion or detector devices.[2] The perovskite CH3NH3SnI3 has an optical band gap of 1.3 eV,[9] indicating a significant red shift compared with the CH3NH3PbI3 (about 1.55 eV).[10]
Recently, lead-free solid-state organic–inorganic hybrid halide perovskite solar cells based on methylammonium tin iodide perovskite semiconductor as the light harvester, such as CH3NH3SnI3, has been identified as a promising photovoltaic (PV) material.[2] Bandgap engineering was implemented by chemical substitution in the form of CH3NH3SnI3-xBrx solid solutions, which can be controllably tuned to cover much of the visible spectrum, thus enabling the realization of lead-free solar cells with an initial power conversion efficiency of 5.73% under simulated full sunlight.[9] Thus, many studies have focused on CH3NH3SnX3. For example, Lang et al.[11] studied the chemical trends of the electronic properties in halide perovskites and proposed that CH3NH3SnBr3 could be a promising light-harvesting material because of appropriate band gap and optical absorption. Feng et al.[12] reported that the band gaps of CH3NH3SnX3 can be engineered to match the visible light spectrum under the low hydrostatic pressure (including compression and expansion). Umari et al. predicted that CH3NH3SnI3 has better electron transport properties than CH3NH3PbI3.[13] The efficiency is expected to be further increased by optimizing and better fundamental understanding of the internal electron dynamics and corresponding interfacial engineering. Papaassiliou and Koutselas[10] reported that perovskite solar cells represent a step towards the realization of low-cost, environmentally friendly solid-state solar cells. Despite prosperous applications of the Sn-based hybrid perovskites in solar cells, a systematic fundamental understanding of their electronic properties and band-edge positions is important for further optimizing their material properties. Jia et al.[14] reported improving light trapping and conversion efficiency of a solar cell by modifying and randomly distributing ZnO nanorods. Carrier mobility is the central issue for optoelectronic semiconducting materials. However, there is rare information about quantitative analysis of the carrier mobility of CH3NH3SnX3 theoretically.
In this work, we investigate the electronic properties, including the effective mass values of hole and electron at the Γ point. According to the calculated effective mass values, we further predict the charge mobility for nontoxic methylammonium tin halide (CH3NH3SnI3) at the equilibrium geometry by using the Heyd–Scuseria–Ernzerhof (HSE06) hybrid functional method.[15] We also clarify the relationship between the carrier mobility anisotropy and electron occupation anisotropy in p orbit of Sn and I atoms to offer constructive guidance to tune the carrier mobility along a certain direction in future.
In this work, first-principles computations on the basis of density functional theory (DFT) are performed in the orthorhombic phase of MASnI3 lattice by using the Vienna ab initio simulation package (VASP).[16] Projector-augmented plane wave (PAW) approach is used to describe the ion–electron interactions.[17,18] We used the generalized gradient approximation (GGA) involving Perdew–Burke–Ernzerhof (PBE) functional and the Grimme vdW correction, which can accurately describe the weak interactions.[19,20] A cutoff energy of 400 eV for the plane-wave basis set is adopted in all computations. The k-point mesh of the Brillouin zone is set to be 4 × 4 × 4 for geometry optimization. The convergence threshold for self-consistent-field iteration is set to be 10−6 eV, and the atomic positions are fully optimized until all components of the residual forces are smaller than 0.001 eV/Å. The density of states (DOS) of perovskite CH3NH3SnI3 is calculated by using the tetrahedron method with Blöchl corrections.[17] As is well known, the standard DFT usually underestimates the band gap of semiconductor due to the inherent lack of derivative discontinuity and the self-correlation error of electrons.[21] Yuan et al.[22] reported that the band gap for cubic phase of CH3NH3SnI3 is just 0.244 eV, which underestimates about 1 eV compared with the experimental data (1.3 eV). To account for the underestimation of the band gap in standard DFT calculations, the accurate electronic structures are then calculated by using the Heyd–Scuseria–Ernzerhof (HSE06) hybrid functional.[15] The scalar-relativistic effects are included in the PAW Pseudopotentials. In addition, we realize that strong spin–orbit coupling (SOC) effects play a significant role in reducing the band gaps of compounds consisting of heavy elements. However, according to Bernal and Yang’s work,[23] we can ignore SOC effects in Sn-based perovskites. So, the band calculated by the HSE06 methods does not consider the spin–orbit coupling (SOC). As is well known, the more serious distortion may induce stronger ferroelectricity, which might play a vital role in the high carrier mobility. So, we use an orthorhombic cell to approximate the realistic structure. We start our calculations by focusing on the orthorhombic phase of CH3NH3SnI3,[24] as shown in Fig.
Crystal structures of orthorhombic in CH3NH3SnI3 perovskite. The left panel is a side view, and the right panel shows the structure in polyhedron graphs (dark: tin; purple: iodine; brown: carbon; blue: nitrogen; pink: hydrogen).
In the matrix (Eq. (
Like MAPbX3, MASnX3 also exhibits a very rich phase diagram as a function of temperature, i.e., the crystal structure goes from cubic, tetragonal, orthorhombic, and monoclinic to triclinic class by cooling.[17,23,27–29] Hao et al. reported a tetragonal CH3NH3SnI3 phase at room temperature.[2] It was found that CH3NH3SnI3 compound crystallizes into monoclinic phase at low temperature experimentally. However, the unit cell of monoclinic phase deviates from the orthorhombic crystal class very slightly.[28,29] The complicated phase transition behavior of CH3NH3SnI3 is attributed to the disordering of (CH3NH3)+ cations and the distortions of (SnX3)− octahedra at finite temperature. The geometrical structure of the hybrid compound is illustrated in Fig.
For orthorhombic phase in perovskite CH3NH3SnI3, the energy band structure is shown in Fig.
As is well known, the standard DFT usually underestimates the band gap of semiconductors due to the inherent lack of derivative discontinuity and the self-correlation error of electrons.[32] Therefore, results (1.27 eV) obtained by the hybrid functional within the framework of density functional theory (DFT-D2) method are closer to the experimental data (1.3 eV).[9] We ascribe the difference between the calculated data and experimental data to the following reasons: (i) the experimental materials are flawed, while the calculated structure is considered as an ideal model; (ii) though HSE06 calculations improve the exchange action, the improved exchange effect is still from a semi-empirical theory. According to the calculated band structure, the carrier effective mass values are shown in Table
The density of states (DOS), and the projected density of states (PDOS) of CH3NH3SnI3 are shown in Fig.
In order to further illustrate our results, the calculated projected DOS in valence band maximum and conduction band minimum are shown in Fig.
In order to evaluate the transfer efficiency of the carrier in perovskite CH3NH3SnI3, we systematically investigate the carrier mobility of CH3NH3SnI3. From experimental measurement, the system of CH3NH3SnI3 was reported to have high electron mobility (μe ∼ 2320 cm2· V−1·s−1),[8] which is higher than CH3NH3PbI3 (μe ∼ 66 cm2· V−1·s−1).[8] The charge mobility is calculated according to Eq. (
In this work, we employ DFT + D2 and HSE06 to study the equilibrium lattice parameters, electronic structures and the intrinsic charge carrier mobility of orthorhombic phase in CH3NH3SnI3. We conclude that CH3NH3SnI3 is semiconductor with direct band gaps at the Γ point. The anisotropy of carrier effective mass can be attributed to the anisotropy of electron occupation probabilities along the three principal directions, so can the anisotropy of carrier mobility. The results reveal that the high carrier mobility is ascribed to the intrinsically small effective mass, in addition to a relatively weak electron–phonon coupling. Owing to the hole mobility higher than the electron mobility, the CH3NH3SnI3 perovskite can be considered as p-type semiconductor in electrical conduction. The suitable energy bad gaps and impressively large carrier mobilities reported here indicate that CH3NH3SnI3 is a very promising material in the field of application and designing for solar cell devices.
1 | |
2 | |
3 | |
4 | |
5 | |
6 | |
7 | |
8 | |
9 | |
10 | |
11 | |
12 | |
13 | |
14 | |
15 | |
16 | |
17 | |
18 | |
19 | |
20 | |
21 | |
22 | |
23 | |
24 | |
25 | |
26 | |
27 | |
28 | |
29 | |
30 | |
31 | |
32 | |
33 | |
34 | |
35 | |
36 |