Triboelectric Effect

Xu, W. et al. (2020). A Droplet-based Electricity Generator with High Instantaneous Power Density. Nature, 578(7795), 392–396. doi:10.1038/s41586-020-1985-6.

From the abstract, “Extensive efforts have been made to harvest energy from water in the form of raindrops, river and ocean waves, tides and others. However, achieving a high density of electrical power generation is challenging. Traditional hydraulic power generation mainly uses electromagnetic generators that are heavy, bulky, and become inefficient with low water supply.

“An alternative, the water-droplet/solid-based triboelectric nanogenerator, has so far generated peak power densities of less than one watt per square metre, owing to the limitations imposed by interfacial effects — as seen in characterizations of the charge generation and transfer that occur at solid–liquid or liquid–liquid interfaces.

“Here we develop a device to harvest energy from impinging water droplets by using an architecture that comprises a polytetrafluoroethylene (PTFE) film on an indium tin oxide substrate plus an aluminium electrode. We show that spreading of an impinged water droplet on the device bridges the originally disconnected components into a closed-loop electrical system, transforming the conventional interfacial effect into a bulk effect, and so enhancing the instantaneous power density by several orders of magnitude over equivalent devices that are limited by interfacial effects.”


Robert Murray-Smith. 2022. Rain Power: A Major Breakthrough. Video. Retrieved from youtube.com/watch?v=ee3WW0OnBL0

Robert Murray-Smith. 2022. 1459: How To Make A Rain Fuel Generator. Video. Retrieved from youtube.com/watch?v=JWHl4TLM9TY.

In these videos, the basic design of the triboelectric generator described in Xu, W. et al. (2020) is presented. The use of a plastic panel — aluminium substrate — PTFE / plumbers' tape — aluminium electrode, the 'getter' (in place of the glass panel — indium tin oxide substrate — teflon coating – aluminium electrode of the original paper) is also investigated and found to be a viable alternative.

Problems with this approach, which were mentioned in subsequent discussions, included the potential corrosion of the aluminium, the potential difficulty of bonding the teflon tape to the substrate, and the relatively low voltage spikes (on the order of 60mV).


cayrex2. 2022. Rain Fuel Generator from Conductive ink. Video. Retrieved from youtube.com/watch?v=334eNJG3EAI.

Robert Murray-Smith. 2022. 1460: Improving The Rain Fuel Generator. Video. Retrieved from youtube.com/watch?v=Mp8mMu33mNk.

In these videos, the design is further refined to use a plastic panel — aluminium substrate — conductive ink — film of (outdoor) polyurethane varnish — aluminium electrode. The varnish is applied by submerging the assemblage (minus the electrode) into an outdoor varnish and allowing it to 'drip dry', before mounting the electrode on top of the varnish film (which now serves the same purpose as the PTFE tape in addition to protecting the conductive ink from the elements).

Not only is the resulting device easier to construct, a similarly sized harvester generated on the order of 250mV (compared to the 60mV of the teflon tape approach).

Information on how to make the conductive ink used in Murray-Smith (2022) can be found in:

Robert Murray-Smith. 2022. 1381: How To Make A Conductive Ink. Video. Retrieved from youtube.com/watch?v=q32Fr0Hh84o.


Robert Murray-Smith. 2022. 1463: Large Rain Fuel Generator in Parallel: An Exploration. Video. Retrieved from youtube.com/watch?v=jIcK5CBue4M.

In this follow-up video, a simple 'parallel' generator is made using a wide arrangement of PTFE tape and strips of Al tape; results from a more 'dribbly' than showering watering rose were approximately 0.2uA short-circuit current / 110mV. Note sure what this would be like under actual rain-like conditions, though.


  1. Curiously, in all of the replications, the water was only dropped from a height of a few centimetres (maximum), one drop at a time; even the original study used an an array, then a matrix of droppers. How might the results vary if, e.g., droplets coming from a raised watering can (~1–2m), or actual rain (hundreds to thousands of metres) were used instead? How would this affect the 'harvested' charge?
  • Last modified: 2022-03-08 22:13
  • by Peter