The Phosphine Oxide Route toward Lead Halide Perovskite Nanocrystals

We report an amine-free synthesis of lead halide perovskite (LHP) nanocrystals, using trioctylphosphine oxide (TOPO) instead of aliphatic amines, in combination with a protic acid (e.g., oleic acid). The overall synthesis scheme bears many similarities to the chemistry behind the preparation of LHP thin films and single crystals, in terms of ligand coordination to the chemical precursors. The acidity of the environment and hence the extent of protonation of the TOPO molecules tune the reactivity of the PbX2 precursor, regulating the size of the nanocrystals. On the other hand, TOPO molecules are virtually absent from the surface of our nanocrystals, which are simply passivated by one type of ligand (e.g., Cs-oleate). Furthermore, our studies reveal that Cs-oleate is dynamically bound to the surface of the nanocrystals and that an optimal surface coverage is critical for achieving high photoluminescence quantum yield. Our scheme delivers NCs with a controlled size and shape: only cubes are formed, with no contamination with platelets, regardless of the reaction conditions that were tested. We attribute such a shape homogeneity to the absence of primary aliphatic amines in our reaction environment, since these are known to promote the formation of nanocrystals with sheet/platelet morphologies or layered phases under certain reaction conditions. The TOPO route is particularly appealing with regard to synthesizing LHP nanocrystals for large-scale manufacturing, as the yield in terms of material produced is close to the theoretical limit: i.e., almost all precursors employed in the synthesis are converted into nanocrystals.


S2. 1 H NMR OA Concentration Dependence
In order to ascertain that our shift in the α-CH2 in OA is due to interactions with TOPO ( Figure 2a of the main text) and not due to concentration, we perform 1 H NMR on OA alone at the same concentrations as the solutions containing TOPO : OA. As can be seen in Figure S2, the gap between the α-CH2 of OA and that of TOPO : OA widens with increasing TOPO : OA ratios (i.e. decreasing OA concentrations). Therefore, we can conclude that the shift we report in Figure 2a is an effect which the interactions with TOPO contribute towards.

S6. Synthesis of CsPbBr 3 NCs via the Amine Route
We follow the procedure reported in a previous paper from our group. 1 In short, 72 mg of PbBr2 (0.2 mmol, 33 mM), 500 μL of oleylamine, 50 μL of oleic acid and 5.45 mL of 1-octadecene are mixed and degassed for 15 minutes at 100 0C in order to obtain a colorless solution. Thereafter, the temperature is ramped to 140 0C under a dry nitrogen flow and 0.5 mL of a 0.15 M Cs-oleate solution in 1-octadecene (previously heated for 10 minutes on a hot-plate set at 200 0C) is swiftly injected. The reaction mixture is immediately cooled after injection with an ice bath and diluted with 5 mL of toluene. The dispersion is centrifuged at 2500 rpm (for 3 minutes) and the NCs are re-dispersed in 2.0 mL of hexane

S7. Synthesis of CsPbBr 3 and Cs 4 PbBr 6 NCs
Synthesis of Cs4PbBr6 NCs. The same standard procedure employed for the CsPbBr3 NCs is used here, but with 4.0 g TOPO (10.36 mmol) and 2.36 mL ODE instead. To separate the NCs, we add toluene to the reaction mixture prior to centrifuging to avoid the excess TOPO from solidifying. (Table S1, Entry iv), (b) Absorption (top) and XRD (bottom) of Cs4PbBr6 NCs (modified version of Table S1, Entry iv, requiring 4 g of TOPO) (c)-(d) corresponding TEM images. XRD reference pattern for CsPbBr3 is given in green and is from COD 4510745, reference pattern for Cs4PbBr6 is given in red and is from ICSD 98-009-7851. S10

S8. Transformation of NCs
CsPbBr3 ↔ Cs4PbBr6 transformation reactions. 3D to 0D. In this transformation, 100 μL of a NC dispersion ([Pb] = 0.025 mM, as measured by ICP) is mixed with 100 μL of toluene and stirred at room temperature. A 0.96 M TOPO in toluene solution is added, 10 μL at a time with stirring intervals, until the solution turns from a bright green suspension to white, indicating the formation of Cs4PbBr6. The total added was 150 μL over 30 minutes. We can achieve the same transformation by adding 50 μL of NCs to a 0.58 M TOPO solution in toluene and stirring for 15 minutes. The transformation does not occur in the presence of oleic acid. 0D to 3D. To reverse the previous transformation, we add 100 μL of oleic acid to the white Cs4PbBr6 solution with stirring, and a yellow-green solution forms. We attribute this to CsPbBr3 but note the loss in confinement of the NCs, as shown by the colour change. A model of a CsPbBr3 NC with edge size of about 2.4 nm was built by cutting a cubic bulk structure along the (100) facets, leaving Cs and Br on the surface. This NC presents a stoichiometry of Cs125Pb64Br240, with an excess positive charge when each ion is considered in its more stable thermodynamic electronic configuration (i.e. Cs + , Pb 2+ and Br -). This excess was compensated by randomly removing 13 Cs ions from the surface, resulting in the following stoichiometry Cs112Pb64Br240. We model four reactions to calculate the stability of the ligands proposed, where each ligand, formed of a cation and an anion, is considered to replace a CsBr pair of adjacent atoms on a facet of the NC: Here the underscore before the ion indicates it is bound to the NC surface, and NC(o) and NC(o) refer to the NC described in the second paragraph of this section but with one and five CsBr pairs of atoms extracted from its surface respectively. The energies -21.34, -9.72, -26.24 and -2.54 kcal/mol are found, respectively, for the four reactions, a negative number indicating a more stable configuration of the left side of the equilibrium.
To study the electronic properties of the NC at different Cs concentrations, we consider different degrees of surface passivation of our NC with Cs-oleate ligands. We start from an ideal fully passivated NC, with Oleate ions substituting all Bron the surface, yielding 16 Cs-oleate ligands per face. From this point we add and subtract Cs-oleate ligands, in order to see its effect on the density of states. Adding two Cs-oleate ligands on the surface is enough to induce two clear trap levels localized at the added extra ligands.
To study the removal of Cs-oleate ligands from the surface of the initial NC, we first calculate the most stable ligand removal sequence. In the structure we can find vertex ligands, where a Pb 2+ cation is bound to three oleate anions; edge ligands, where a Pb 2+ cation is bound to two oleate anions, and finally facet ligands, where a Pb 2+ cation is bound to an oleate anion. We find that the higher the oleate coordination around Pb, the easier it is to remove a Cs-oleate ligand. Hence, we respect this order for ligand removal. Here, we find that, when about 55 % of the ligands are removed, trap states localized at ligands start to reduce the band gap. This last structure presents the stoichiometry of Cs65Pb64Br144. Figure S10 Relaxed structures with the Cs-oleate and TOPOH..Br attached to the NC surface. In the left, TOPOH + substitutes Cs + and Broccupies a Brsite. On the right side, Cs + enters a Cs + site and the oleate anion substitutes Br -S11. NMR Study of Ligands   When no additional OA is present (Washed NCs) the right part of the resonance slightly shows a shoulder, which is attributable to the percentage of OAfree in exchange with OAbound to NC. When OA is added, the peak of OAbound fractions (indicated by the red asterisk, *) partially overlaps the intense OAfree signal. On adding more OA, this shoulder becomes less visible, but is still present and accounts for the line broadening which is affected by the relative populations of the two exchanging sites. Full details of the washing procedure of NCs can be found in the Experimental Section of the main text. Briefly, NCs are washed by adding an equal volume of acetone and centrifuging. The NCs are re-dispersed in toluene and OA is added, if being used. This process is repeated twice to obtain washed NCs. S15 S12. Colloidal Stability of Nanocrystals Figure S12 Dynamic light scattering measurements of as-synthesized and washed (with added OA) CsPbBr3 NC dispersions over an eight day period, related to the NCs in Figure 6 of the main text where their PLQY stability is presented. S16 S13. Alternative Acids Study We are able to expand our NC synthesis to use acids other than OA. We synthesize NCs using TDPA and DOPA with modifications to the TOPO-acid ratios to reflect the affinity of the acid to TOPO, which we establish by using 31 P NMR.
Figure S13 a) Differences of 31 P chemical shift of TOPO with acids TDPA, OA and DOPA, all in a 1:1 ratio with TOPO, b) TEM image of NCs made with TDPA instead of OA (10 nm, σ = 10 %). c) TEM image of NCs made with DOPA instead of OA (7 nm, σ = 28 %).