Skip to content
PaperFren

Hot electrons · Plasmonics

Light on metal nanoparticles mostly makes heat, yet hot-electron transfer is still measured

Evidence: ContestedQualified studies on the same question disagree. What the labels mean

Study published Jan 1, 2019. PaperFren added this explanation Sep 26, 2026.

Save this development or follow its topic to track what changes.

Short answer

Under continuous illumination theory says absorbed energy mostly becomes heat, but ultrafast measurements in engineered structures still see percent-level electron injection.

What happened

Dubi and Sivan solved a steady-state Boltzmann equation for electrons in a silver sphere under continuous light, conserving energy between electrons, lattice and surroundings. At low intensity the power channelled into non-thermal carriers was more than eight orders of magnitude below the power heating the electrons, rising only to about 10^-5 of the total at higher fields. In a different system, a gold grating on monolayer MoS2 tuned for strong coupling between two plasmon types, pump-probe measurements showed electron injection in about 40 fs, with an external quantum yield peaking at 1.65%.

Why it matters

Many photocatalysis and photodetection results are credited to hot electrons. If the steady-state budget is overwhelmingly thermal, some of those results may be heating effects; if strong coupling and ultrafast injection behave differently, the answer depends on regime and structure rather than being one number for 'plasmonics'.

Evidence

Study type
Theoretical model versus a pump-probe experiment on a different structure
Sample
Numerical solutions for model silver; one experimental heterostructure family with several grating periods
Journal
Light: Science & Applications · peer reviewed
Replication
The two papers address different regimes (steady-state versus femtosecond pulses) and have not been tested against each other
Limitations
The model neglects interband transitions and non-equilibrium phonons and does not model charge transfer out of the particle. The experiment's injection time is deconvolved against a similar-length instrument response and its 'energy recycling' mechanism is not directly measured.

What this connects to

Sources

The 2 studies this explanation is built from, by the role each plays. Every source links to PaperFren’s explanation of it and to the original paper.

Primary study

  • When light hits metal nanoparticles, is it hot electrons or heat?

    Dubi Y, Sivan Y · 2019 · Light, science & applications · 81 citations

    Under steady illumination, almost all the light energy absorbed by a metal nanoparticle ends up heating it, and only a vanishingly small fraction creates the high-energy 'hot' electrons often credited for photocatalysis.

    What it does not show

    It is a purely theoretical study with parameters chosen for silver at one plasmon resonance; no new experiment tests the predictions. The model neglects interband transitions, non-equilibrium phonons and field inhomogeneity inside the particle, and does not treat the high-intensity regime where the metal's permittivity would itself change. It argues that some photocatalysis results are thermal, but does not model the actual chemistry or electron transfer out of the particle.

    PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by

Conflicting evidence

Asks the same question and reaches a different answer.

  • How fast do plasmon hot electrons jump into a 2D semiconductor?

    Shan H, Yu Y, Wang X, et al. · 2019 · Light, science & applications · 76 citations

    When two kinds of plasmon in a gold nanostructure are strongly coupled, hot electrons pass into an atom-thin semiconductor in about 40 femtoseconds and more efficiently than one plasmon type alone would allow.

    What it does not show

    The 'energy recycling' mechanism for why strong coupling helps is a proposed model supported by a rough efficiency estimate, not measured directly. The injection time comes from deconvolving a signal against an instrument response of similar duration, so it is an estimate. Only one material system and a small set of grating periods were studied, and the yield of about one to two percent is still low for practical devices.

    PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by

Before

Hot electrons were often treated as the main product of plasmon decay and the natural explanation for plasmon-driven chemistry.

Now

A careful energy-conserving model puts the non-thermal share at a tiny fraction under steady light, while a pulsed strong-coupling experiment reports measurable injection. The model is for silver without interband transitions, and the experiment is one material system whose efficiency mechanism is proposed rather than measured, so neither settles the other.