1：Experimental Design and Method Selection:

The study involved growing InP-InAs-InP multi-shell nanowires using Au-assisted Chemical Beam Epitaxy (CBE) on InP(111)B and InP(100) substrates to control crystal phase (wurtzite or zincblende). Photoluminescence (PL) measurements were conducted at low temperature (≈6K) with non-resonant excitation using a 532 nm laser, and numerical simulations were performed to model electron and hole states.

2：Sample Selection and Data Sources:

Nanowires were grown with specific dimensions: InP stem diameter of 30 nm, InAs shell thickness of 2-8 monolayers, InAs island size of approximately 30 nm wide and 8-10 nm high, and InP outer shell thickness of 15 nm. Samples included ensembles of vertical nanowires on InP substrates and individual nanowires drop-casted onto SiO2 substrates.

3：List of Experimental Equipment and Materials:

Equipment included a Riber Compact-21 CBE system for growth, Zeiss field-emission SEM for morphology, Titan Themis TEM for structural analysis, confocal microscopy with NA=0.75 objective for micro-PL, 0.55 m spectrometer, CCD and InGaAs cameras for detection, half-wave plate polarizer for polarization measurements, and a cryostat with xyz-piezoelectric stage. Materials included InP and InAs semiconductors, gold colloids as catalysts, and precursors such as TMIn, TBAs, and TBP.

4：75 objective for micro-PL, 55 m spectrometer, CCD and InGaAs cameras for detection, half-wave plate polarizer for polarization measurements, and a cryostat with xyz-piezoelectric stage. Materials included InP and InAs semiconductors, gold colloids as catalysts, and precursors such as TMIn, TBAs, and TBP. Experimental Procedures and Operational Workflow:

4. Experimental Procedures and Operational Workflow: Nanowires were grown via CBE at 390°C, with controlled precursor fluxes. PL measurements were performed by exciting samples with a 532 nm laser at varying powers and polarizations, collecting and dispersing light through a spectrometer, and detecting signals with cameras. Data were analyzed for energy peaks and polarization dependence.

5：Data Analysis Methods:

PL spectra were analyzed to identify emission peaks. Numerical simulations used a finite-volume integration algorithm with a grid of 57500 nodes, assuming parabolic bands and translational invariance, to calculate emission energies from quantum-confined structures.