I believe the CFI where I was based said that descending quick (cruise speed almost) should generate enough friction on the leading edge to keep it from icing in those conditions.
When I saw this posted I was itching to respond too, until I realized I don't have enough data. But my gut feeling told me that it was wildly inaccurate.
As far as I know, water can exist in supercooled, liquid form, up to -15C. So in order to be safe from icing the friction has to generate at least 15C of heat. As far as I know, that only happens if you get into the 500 knot+ region. In other words: airliners that routinely fly at mach 0.8 or thereabouts. Not in the spamcans we generally fly in this forum.
But I'd be very interested in knowing the calculations behind what Beagle wrote, above. It's not something that was covered in my PPL theory, at least.
Anyway, supporting my case, if you look at propellors of de-iced aircraft you'll find that the inner half or third is heated or TKS de-iced. Only the outermost half or two thirds is not de-iced. Assuming that the tips rotate at mach 0.8 or thereabouts (which is the normal propellor limit - above this you will get mach effects which leads to a reduction in efficiency, and a significant increase in noise), that means that anything below mach 0.27 or thereabouts needs to be forcibly de-iced, and cannot be safe from icing by friction effects alone.