There are some attempts to do the math on whether just the fuel could have done it in the comments to this video (https://www.youtube.com/watch?v=UVzLV-R_-kk ht

There are some attempts to do the math on whether just the fuel could have done it in the comments to this video (https://www.youtube.com/watch?v=UVzLV-R_-kk https://www.youtube.com/watch?v=UVzLV-R_-kk). Example: The thrust from a propeller blade is typically proportional to the density of the medium it is operating in (plus other factors of course). The density of air at altitude will be somewhat less than 1 kg/m^3. Density of pure liquid kerosene is around 800 kg/m^3, but in this case the kerosene is dispersed as droplets in the air. If we take a hypothetical guess that the air/fuel mixture is around 25% fuel, 75% air, the net density of the mixture will be around 200 kg/m^3. On encountering the fuel mixture of this density, the propeller blade would experience a sudden increase in load by a factor of about 200. A sudden overload of this magnitude would be enough to bend most things. Of course we don't know the actual air/fuel ratio. It could be less or could be more. The velocity of the dumped fuel relative to the drone's airspeed could also be significant. Too many unknowns to be anything like accurate, but appears not impossible for a fuel cloud to bend a propeller blade.