Analysis Workspace

Protocol-aware mobility extraction

Choose SCLC or FET/TFT, upload one clean CSV, provide protocol-specific metadata, and review apparent mobility with audit details.

Method Guide

What do you want to do?

Choose method

Choose extraction mode

Required data: VG, ID. Required metadata: W, L, Ci, VDS.

Upload, template, or sample

Upload CSV / XLSXDownload CSV template

Single sweep: VG, ID. Paired sweeps: VG + ID_forward + ID_reverse.

Data sourceThis changes provenance and reliability guards, never the formula.

Numeric CSV/XLSX rows → local μ(VG)

An isolated row has no mobility because mobility requires a slope. At each VG, the platform fits nearby rows over at least 8% of the full VG span and at least 5 rows, then assigns that regression-derived apparent μ to the centre VG.

Rows are used directly with symmetric local quadratic least squares on their actual VG positions (Savitzky–Golay geometry when uniformly spaced). No pixel tracing or image-density resampling is applied.

Advanced input options

Digitize from figure is an advanced input path and is not part of the main workflow. Use it only when a clean CSV is unavailable.

Open digitize figure tool

Fill metadata

Channel width W *

Channel length L *

Gate dielectric Ci *

Enter Ci as reported (papers usually use nF/cm²). Internally normalized to F/m² for calculation.

Drain voltage VDS (V) *

Required by μlin = L·|dID/dVG|/(W·Ci·|VDS|); changing VDS changes the calculated mobility.

Fitting window

Reliability-anchored, not user-tuned

The window is selected automatically and reported explicitly (start/end VG + neighboring-window stability). We deliberately do not offer a “pick your own points to raise the number” control: hand-narrowing the fit to the steepest segment is the main cause of mobility overestimation.

Instead the result reports a reliability factor r (steepest vs zero-threshold average slope) and a μ(VG) diagnostic plot — a flat curve means the number is well-defined; a peak means it is regime-dependent. Reference: Choi et al., Nat. Mater. 2018.