Growth mechanism of oleylammonium-based tin and lead bromide perovskite nanostructures

Metal halide perovskites, particularly using tin and lead as bivalent cations, are well known for their synthetic versatility and ion mobility. These materials possess intriguing ionic properties that allow the formation of 2D Ruddlesden–Popper (RP) and 3D metal halide perovskite nanocrystals (NCs) under similar synthetic conditions. We studied the synthesis mechanism of oleylammonium-based Sn and Pb bromide perovskites 2D Ruddlesden–Popper (RP) in comparison with the 3D CsPbBr3 and CsSnBr3 NCs. Using experimental techniques in combination with theoretical calculations, we studied the interactions of the long-chain organic cations with the inorganic layers and between each other to assess their stability. Our findings suggest that tin bromide is more inclined toward forming higher-order RP phases or 3D NCs than lead bromide. Furthermore, we demonstrate the synthesis of precisely tuned CsSnBr3 3D NCs (7 and 10 nm) using standard surface ligands. When the 3D and 2D tin halide perovskite nanostructures coexist in suspension, the obtained drop-cast thin films showed the preferential positioning of residual RP nanostructures at the interface with the substrate. This study encourages further exploration of low-dimensional hybrid materials and emphasizes the need for understanding mechanisms to develop efficient synthetic routes for high-quality tin-halide perovskite NCs.


Table S1
Elemental Analysis of 7 nm and 10 nm CsSnBr 3 NCs

I. Theoretical Calculations
The minimum-energy structures of the 2D Ruddlesden-Popper phases (L 2 Cs n-1 M n X 3n+1 , with n in the range 1-4) and those of the crystalline bulk CsMX 3 phases, where M stands for the metal cation (Pb 2+ , Sn 2+ ), X for the anion (I -, Br -) and L for the OLA + ligand, were calculated under the Density Functional Theory (DFT) framework.Periodic boundary conditions were imposed and tier-2 numerical atom-centered orbitals (NAO) basis functions were used in conjunction with the GGA PBEsol functional, 2 as implemented in the FHI-aims code. 3 Relativistic effects were considered through the use of the scalar ZORA scheme.The initial structures for the bulk CsMX 3 phases were obtained from the Materials Project database, The experimental evidences that the [R-NH 3 ] 2 MX 4 phases could act as precursors for the bulk CsMX 3 structures, that structures of type [R-NH 3 ] 2 Cs n-1 M n X 3n+1 with n ≥ 1 were detected via XRD measurements, and that the conversion to 3D NCs was more complete for the case of CsSnBr 3 , prompted us to theoretically investigate the formation energies of the different phases and to devise the driving forces and interactions pointing to this behavior.The formation energies phases were evaluated as follows 5 The lattice parameters for the (OLA) 2 Cs n-1 M n X 3n+1 phases (with n > 1) resulting from the PBEsol/tier-2 NAO optimization are listed in Table S3.The values of the [R-NH 3 ] 2 Cs n-1 M n X 3n+1 formation energies are listed in Table S4.From them, it can be concluded that the formation of the [R-NH 3 ] 2 MX 4 phase is in all cases favorable, as negative values were predicted for the three systems.Furthermore, the formation energies of  To disentangle the trends observed both experimentally and by means of DFT calculations, the interactions between the oleyl amine chains within the [R-NH 3 ] 2 MX 4 structures were investigated through the calculation of the non-covalent index (NCI). 6This magnitude is based on the reduced density gradient s and the electron density p according to the following equation In the presence of intermolecular interactions there is a shift in the reduced density, pointing to the creation of critical interacting points between atoms or molecular fragments that can be visualized (see Figure 2a).Furthermore, the value of the density can be used to assign the strength of the interaction, while the sign of the Laplacian of the density ( 2 ) can be used to ∇ 2  distinguish between repulsive interactions ( 2 ) and those with van der Waals character ( 2 > 0 ).The NCIPLOT code was used on top of a 3x3x1 supercell of the different [R-NH 3 ] 2 MX 4 ≲ 0 systems, 7 resulting from the DFT optimization, but removing the MX 6 octahedra.Promolecular densities were used as they provide a reasonable ratio between the computational cost and the accuracy of the calculations.The magnitude of interaction between the OLA + cations and the perovskite MX 6 octahedra is crucial for the desorption of the cations, required to form more complex structures of increased dimensionality.To estimate the OLA + -perovskite interaction, 3x3x1 supercells were constructed from the PBEsol/tier-2 NAO minimum-energy structures of [R-NH 3 ] 2 MX 4 and removing the central cation.As the terminal OLA + mainly interacts with the perovskite through the ammonium group, the oleyl amine chains were substituted by shorter cations with five C atoms to reduce the computational cost.The interaction energy E i was then calculated at the PBEsol/tier-2 NAO basis set level of theory as follows where E bulk corresponds to the energy of the 3x3x1 supercell with short cations, E cation is the energy of the desorbed short cation, and E desorb is the energy of the 3x3x1 supercell once the central short cation was desorbed.All the calculations were performed with periodic boundary conditions.The results, which are listed in Table S5, were calculated accounting for intermolecular interactions with the use of the Tkatchenko-Scheffler dispersion correction as available in the FHI-aims code. 8respective of the use of the dispersion correction, the interaction energy is lower for the [R-NH 3 ] 2 SnBr 4 structure, which ensures a weaker oleyl amine-perovskite octahedra interaction.This result, combined with the more destabilizing interaction between oleyl amine chains, favors the cation desorption required to form bulkier phases in the case of [R-NH 3 ] 2 SnBr 4 .
Table S5.Oleyl amine-perovskite interaction energy (in eV) calculated with and without dispersion correction at the PBEsol/tier-2 NAO basis set level of theory.The system was parametrized according to the CHARMM force field using CGenFF through the CHARMM-GUI interface. 9The DFT lattice parameters from the optimization were used to impose periodic boundary conditions and the N atoms were frozen during the trajectories.First, 1000 steps of classical minimization were performed keeping the unit cell fixed.Then, a NVT thermalization trajectory of 50 ns with a time step of 2 fs was used to ensure that the system was correctly equilibrated and thermalized.A significant contraction of the oleylamine chains along the c axis occurred and therefore several NVT trajectories were performed, adjusting the size of the unit cell, until no further contraction was observed.The d-spacing value was estimated as the distance between the planes generated by the cation N atoms plus an average distance of 2.25 Å between the N and the plane that crosses the center of the perovskite octahedra.Even though our simulations neglect the effect of the MX 6 octahedra and that of the displacement of the chains on the final arrangement, our results show that large intercalation of the chains is not required to attain the short d-spacings obtained experimentally.

3 22906)
where all structures were optimized at the PBEsol/tier-2 NAO level of theory, considering periodic solid phases, and a k-grid size of 6×6×6.The initial structures for the optimization were retrieved from the Materials Project database, considering Pnma orthorhombic phases for PbBr 2 (mp-28077) and SnBr 2 (mp-29862), C 2/m phase for SnI 2 (mp-27194) and cubic Pm m phases for CsBr (mp-̅ and CsI (mp-1056920), as experimentally determined.The structures of OLABr and OLAI were constructed by example from that of the orthorhombic Pbcm phase of MAI (mp-997570).

[R-NH 3 ] 2
Cs n-1 M n X 3n+1 become more negative as the value of n increases.These two results agree with the isolation of the [R-NH 3 ] 2 MX 4 phases and the co-existence of higher-order n structures depending on the experimental conditions.The formation energy in the case of [R-NH 3 ] 2 SnBr 4 is significantly smaller than that for [R-NH 3 ] 2 PbBr 4 and [R-NH 3 ] 2 SnI 4 .Moreover, the net stabilization of each of the higher-order n structures (n = 4) compared to the [R-NH 3 ] 2 MX 4 phase is larger for the Sn-Br combination (0.79 eV), in comparison with that obtained for Sn-I (0.50 eV) and Pb-Br (0.68 eV).These results point towards the more complete formation of the 3D bulk CsSnBr 3 phase, as well as for higher-order [R-NH 3 ] 2 Cs n-1 M n X 3n+1 structures.For comparison, the formation energies of the [R-NH 3 ] 2 Cs n-1 M n X 3n+1 phases with n = 4 were considered as those for the 3D bulk structures.
The interlayer distance (d-spacing) for the [R-NH 3 ] 2 MX 4 materials was also inspected by computational simulations to devise the degree of chain intercalation that could correspond to the experimental values shown by XRD measurements in the range of 3.8-4.2nm.To inspect the effect of temperature and dynamic disorder in the d-spacing, classical Molecular Dynamics simulations were performed on top of a 3x3x1 supercell of [R-NH 3 ] 2 SnBr 4 , as a representative example, with complete oleyl amine chains but without the perovskite octahedra (see Figure2b).

Figure 1 Figure
Figure S3..The stability of CsSnX 3 NCs (in days) is plotted with SnI 2 concentration (M) in the reaction mixture with marked in blue the concentration that works well for Pb-halide perovskite NCs as reported by Protesescu et al. 1

Figure S6 . 13 Figure
Figure S6.(a) XRD of SnX 2 halide salts in DOPA quenched to room temperature without Cs injection.The SnBr 2 Pnma reference is reproduced from Eckold et al. 10 (b) XRD of SnX 2 halide salts in DOPA quenched to room temperature with Cs injection producing bulk Sn perovskites.The bulk references are reproduced from cited literature.[11][12][13]

Figure S9 .
Figure S9.UV-Visible and PL spectroscopy of CsSnBr 3 NCs synthesized with different cation rations of Cs: Sn at 200⁰C.

Figure S10 .
Figure S10.(a) XRD of CsSnBr 3 NCs synthesized in ODE at 200⁰C (grey) and 100⁰C (red) plotted with CsSnBr 3 cubic Pm3m bulk reference 12 showing the co-existence of 2D and 3D perovskite structures at low temperature injection.(b) UV-Visible and PL spectroscopy of CsSnBr 3 NCs in ODE at different temperatures.

Figure S11 .
Figure S11.SEM Images showing the co-existence of 3D CsSnBr 3 NCs and 2D nanosheets when the reaction is performed at 100⁰C.

Figure S13 .
Figure S13.Comparison of optical properties of 2D and 3D tin-halide perovskite materials.
4considering γ-orthorhombic phases for CsPbBr 3 and CsSnI 3 (code references mp-567629 and mp-568570, respectively) and the α-cubic phase for CsSnBr 3 (code reference mp-27214).The structural parameters of the minimum-energy crystal structures resulting from the full lattice and ionic optimization are listed in TableS1.The 2D Ruddlesden-Popper structures were constructed by hand from the optimized structures of the bulk phases and therefore α-cubic phase was assumed for [R-NH 3 ] 2 SnBr 4 and γorthorhombic phases were constructed for [R-NH 3 ] 2 PbBr 4 and [R-NH 3 ] 2 SnI 4 .The k-grid size was fixed for all systems at 6×6×6 due to the convergence of the unit cell energy and crystal lattice parameters, which are listed in TableS2.Large vacuum distances of at least 50 Å between the oleyl amine chains were set along the c axis to avoid clashes and interactions that could modify their arrangement.TableS1.Minimum-energy bulk CsMX 3 phase crystal lattice parameters obtained from the PBEsol/tier-2 NAO optimization.