Surface photovoltage is usually presented as relative: this sample's signal is larger than that one's. That comparison suffices for screening, but mechanistic work eventually needs the question answered numerically: what fraction of photogenerated charge actually separates? Quantifying SPV is possible, but it demands deliberate normalization.
The measured photovoltage depends on the amount of light absorbed, the capacitance of the measurement geometry and the measurement technique itself, in addition to the separation behavior of interest. Two samples with identical separation but different absorption report different raw signals.
Absorb first: measure the sample's absorption at the SPV excitation wavelength and normalize the signal by absorbed photon flux, isolating separation per absorbed photon. Reference second: measure a well-characterized reference material with the same geometry to factor out the apparatus response. Both together turn raw volts into a comparable, defensible figure of merit.
Choose a stable reference with known behavior, run it in every session, and express sample signals relative to it. Session-relative reporting cancels apparatus drift and makes long campaigns comparable. State the reference and the normalization procedure explicitly in publications.
Report normalized SPV with its measurement conditions (wavelength, flux, geometry) rather than pretending to an absolute efficiency. A well-defined relative metric honestly labeled beats an absolute number without provenance.
See Perfectlight's SPV systems and mechanism research solutions for quantification workflows.
