Structure confirmed: one workbook `Book1`, X-axis in eV (col A), and 15 power spectra already merged as columns (5→100 µW). No separate Si/reference folder exists, so no background subtraction is needed. Suggested peak window is **1.23–1.45 eV**.

Despiking first (mandatory before any fit):



![Power Fit]([file link removed])

## Power-dependent PL analysis — `Book1` (15 spectra, 5→100 µW)

**Pipeline:** inspect → despike (σ=4, 130 spikes removed) → Lorentzian fit per spectrum → allometric fit I = a·Pᵇ. No Si/reference folder existed, so no background subtraction was applied. Fit window 1.23–1.45 eV.

**Lorentzian fits:** All 15 peaks converged at HIGH quality (R² = 0.887–0.984, median 0.981; SNR 7.6–24.0).

| Quantity | Result |
|---|---|
| Peak position xc | 1.3206 eV (5 µW) → 1.3356 eV (100 µW), **blueshift** ≈ +5.0 meV/decade |
| FWHM | 62.4 → 54.4 meV (narrowing with power) |
| Intensity I | 98.8 → 1714.6 counts |

**Allometric fit (left panel, I = a·Pᵇ):**
- a = 16.28 ± 1.81
- **b = 1.0048 ± 0.0261**
- R² = 0.995 → essentially linear (b ≈ 1), consistent with a single excitonic recombination channel that is not saturating over this power range.

The right panel (xc vs P) shows the monotonic blueshift; it follows a sub-linear saturation curve rather than a clean power law, so its "allometric" character is weak compared with the intensity fit.

📥 Full per-power table: [power_fit_results.csv]([file link removed])

RESULT: b = 1.0048 ± 0.0261