Intercalates of Bi2Se3 studied in situ by time-resolved powder X-ray diffraction and neutron diffraction

Intercalation of lithium and ammonia into the layered semiconductor Bi2Se3 proceeds via a hyperextended (by >60%) ammonia-rich intercalate, to eventually produce a layered compound with lithium amide intercalated between the bismuth selenide layers which offers scope for further chemical manipulation.


Experimental Section Synthesis
All manipulations were carried out in a Glovebox Technology argon-filled dry box with an O2 and H2O content below 1 ppm or on a Schlenk line. The parent compound Bi2Se3 was synthesized by mixing together high-purity Bi pieces (Sigma Aldrich; 99.999%) and Se powder (Alfa Aeser; 99.999%) and heating them in a sealed silica tube to 720 °C at 2 °C min -1 , holding the temperature for 48 h, cooling to room temperature at the natural rate of the furnace, grinding the product into a fine powder and repeating the heating procedure with the same ramp rate and holding at 720 °C for 72 h before cooling to room temperature.
The intercalates, Lix(NH2/NH3)yBi2Se3 and Lix(ND2/ND3)yBi2Se3 with targeted compositions 0.2 ≤ x ≤ 1.0, were synthesised by placing the appropriate stoichiometric quantities of Li metal (Sigma Aldrich; 99%) and Bi2Se3 in a Schlenk tube with a magnetic stirrer bar. The Schlenk tube and a cylinder of ammonia (NH3: BOC; 99.98 %, or ND3: Sigma Aldrich; 99% D) were attached to a Schlenk line. After evacuating all pipework between the ammonia cylinder and the sample space, the Schlenk tube was placed in a slush bath of dry ice and isopropanol (-78 °C) to allow condensation of ammonia onto the reactants. For 0.40 g of Bi2Se3, around 15 mL of ammonia was condensed into the Schlenk tube. The solution was observed to briefly turn blue, characteristic of solvated electrons, after which the solution decoloured. The Schlenk tube was allowed to warm with the slush bath to room temperature over the course of around 4 h while boiling off ammonia [safety note: ammonia is volatile and toxic. Therefore, the Schlenk line was constructed so that any ammonia pressure could be relieved at all times via a mercury bubbler]. After evacuating the Schlenk tube, a dry dark grey powder was obtained.

Powder neutron diffraction
Powder neutron diffraction (PND) experiments were performed at 5 K and room temperature on the POLARIS diffractometer 1 at the ISIS Facility, Rutherford Appleton Laboratory, UK. Samples with NH3 or ND3 were contained in 6-mm-diameter thin-walled vanadium cans that were sealed with indium gaskets. Rietveld analysis was performed with TOPAS-Academic V5. 2

S3
In situ X-ray diffraction The synthesis of the ammonia-rich Li-intercalated Bi2Se3 were performed on the I12 beamline 3 at the Diamond Light Source (UK) in an experiment conceived to investigate the structural changes during the intercalation process in situ and identify possible intermediate phases. The set-up was similar to that described previously for the intercalates of FeSe. 4 In an argon-filled glovebox, 3.2 (0.461 mmol) mg of Li and a small magnetic stirrer bar were loaded into the bottom of a 18 mm o.d × 4 mm wall Pyrex ampules sealed with high pressure Rotaflow Teflon valve [safety note: since these ampules were to be sealed with liquid ammonia at room temperature or below, they were professionally

Description of Rietveld refinements
A coupled Rietveld refinement for the highest resolution banks 3/8, 4/7 and 5/6 of the POLARIS diffractometers was performed on all four data sets in which a single structural model was used to fil all patterns. In this refinement, the lattice parameters for each data set were refined independently, to allow for small differences between the H-and Dcontaining samples and the changes in temperature, while the atomic positions and occupancies were coupled between each data set (i.e. the assumption was made that the structural model was temperature independent). Moreover, all atoms were refined with identical isotropic displacement parameters: one for the 5K data sets and one for the room temperature data sets. By assuming that substituting D for H does not impact the intercalations, and by benefitting from their different neutron scattering lengths, we forced the H and D occupancies to be the same in order to completely model the intercalated structure.  Table S1. Structural and refinement parameters of POLARIS neutron diffraction data (esd's in parentheses).   (Table S1).  5. The background is subtracted for clarity. The dotted framework corresponds to the range plotted in Fig. 4. The lattice parameters of each structure is given in Table 1.  peak as a function of temperature between 470 and 500 K indicating the formation of an additional crystalline phase with a d-spacing of ~ 11.8 Å formed by annealing, prior to decomposition at elevated temperatures.