h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

I like this entry, particularly the bit that explained about transferring the entire weight of a person onto the tiny highest peaks that exist on surfaces of materials at microscopic levels and how that emlts the ice beneath.
Doesn't it also have to do with ice being more frictionless than other surfaces, for though as you rightly point out, even smooth surfaces will have minescule troughs and valleys and cratered, pock-marked surfaces, ice is smooth(er) and so there is is less reisitance between the two surfaces when they contact and the forces applied simply by moving means there is less opposing force to counteract that and we are more likely to fall because we are used to walking on surfaces that have more resistance and friction.
That is also why they grit the roads during winter - the sand provides some grip to the surface of the road covered in snow, until the snow melts from the heat generated by cars going over it, the heat being the by-product of the friction, as more tarmac is revelaed so there is more heat ( 'cos of more grip and more friction ) and eventually the road is cleared and the snow becomes that horrible grey mulch lining the pavements.

Just a neat point of interest. As we consider smoother and smoother surfaces, they become less and less slippery (assuming no additional lubrication as is the case with ice). In short, this is due to the rise in the number of contact points. Grip two beautifully smooth surfaces together, and they theoretically ought to dry weld to each other.

I wonder would that work ? Is there a catch ?

I suspect that at a molecular level the molecules of one surface would have to fit perfectly around or between the molecules of the other surface.

Well, it all brings to mind certain questions. For instance, what if the sole of the shoe were to land on the ice in, say, Greenland, where it's very, very cold. Would the microscopic peaks be unable to generate sufficient heat to melt the ice? And would the shoe therefore not slip? Also, why does the shoe that melts the ice not create suction, as a ski does in warm snow? What REALLY happens at the ice/shoe interface?

Recent experiments at the top of an extremely cold mountain (-26C before windchill) showed that snow at least was not slippery at all on the mad sprint to carry my frost-bitten friend into the nearest bar. I didn't fall over, and we saved his ears and cheeks.
As a further point of not quite relevent interest, if you are driving a snocat (one of those half-track things) and you stop for a break, in sufficiently severe conditions the snow that has melted to slush inside the tracks freezes solid and you have to chisel the ice out before you can get the thing moving again. Usually, of course, this happens when it is dark and you are tired and hungry...
And furthermore! I was always under the impression that dry-welding occurred in perfectly clean surfaces when the (usually) metal atoms on either side of the break form crystalline cross-links with each other. In order for them to do this, they have to get VERY close to each other, hence the requirement for a 'perfectly' flat (ie smooth) surface. I understand that it is possible to dry-weld pieces of steel that have been cut apart using a laser beam.

All this is fine and dandy, it makes perfect sense. but what about socks on a clean, smooth floor? why is that slippery?

I at one time worked in a calibration lab, where we had a great many tools which helped us to ensure any device required for day-to-day measuring purposes was properly doing its job.

In order to calibrate such tools as inside- and outside-micrometers, electronic levels, and dial calipers, we had to use a specific standard. The official name escapes me at the moment, as most of the time we just called them cal-blocks. Cal-blocks were precision cut blocks of steel alloy that were impossibly smooth on all sides, and had precise measurements for each block. In order to handle them, you had to wear special gloves as the oils and heat from your hands could theoretically warp the surface of the blocks, thereby giving an erroneous measurement.

Why do I bring this up? Because the method we had to employ in order to measure calibration tolerances on these devices was to verify their inner and outer limits, which meant using them to measure a standard at the smallest level (usually an impossibly thin sliver of steel, standardized to one micrometer thick), and the outer limits, which meant that most of the time, we would have to combine more than one block to accurately match that tool's highest measuring capacity.

To combine more than one block, we would hold their surfaces together and rub them ever so slightly. After a bit, they would stick together as if they were one, even if they were shaken vigourously. We then completed our measurement and slid them apart before putting them away.

We also had a large block of laser cut, polished marble that was a standard of flatness which exhibited the same properties.

So yes, long story short, not only have I seen "dry welding" I have gotten paid to do it daily!

The question about what happens when it is very cold has become clear in the last week or so, where the temperature has stayed significantly lower for a period. And walking on frozen snow, where it remains frozen, is far easier than when it starts to melt ( or nearly melt ) giving rise to the effects described.

This meant that I could walk easier at home, where the pavements were covered in snow, that at work where they were not, but there were occasionally icy patches.

Woah! Long dead thread activation! Don't we need like garlic or silver bullets or something?

One thing not mentioned is gravity. This is why small, mincing steps are less fatal on a slippery surface than large strides (gravity pulling you down meets propulsion throwing you forward). You won't necessarily fall over when moving at speed or people couldn't ice skate, so balance and awareness come into it (how well you control the situation, rather than the situation controls you).