The properties of friction change from one situation to the next, so a universal explanation doesn’t exist.
Why is ice slippery? You might have heard that scientists don’t know the answer. If that surprises you, it could be because simple language can camouflage complex problems.
Take the question “why does spoiled milk smell bad?” as an example. It seems like a simple question, but to answer it at a fundamental level, you’d have to dig into topics ranging from the biochemistry of lactose and fat and bacteria to the physiology of our noses and the evolution of our psychology.
Just because something is an everyday phenomenon, that doesn’t mean it’s not extremely complex. It is arguably easier to detect the Higgs boson than to truly and completely understand why spoiled milk smells bad.
Or why ice is slippery.
Friction is something called an empirical property, which means that it is something that is observed or measured, rather than being deducible using fundamental theories. For example, the protective effects of COVID-19 vaccines are empirical measurements, which can change depending on the testing conditions.
Friction is the measure of how much force is needed to push a certain object across a certain surface. The results spring from a combination of effects such as how much the two surfaces stick to each other, how rough they are, how much they deform, and whether there is a lubricating film between them.
So, strictly speaking, “why is ice slippery” is not the right question. Nothing can be slippery by itself and ice is not always slippery. For example, if you have crampons on, or if you’re trying to scoot sideways wearing skates, ice is not that slippery.
Often research about ice slipperiness focuses on the physics of skates on ice, and what causes the presence of loose water molecules between the ice and the blade.
Notice the wording “loose water molecules,” and not just “water.”
That’s because the molecules might be described either as a liquid or a gas, depending on how thick the layer is. If the layer is only a single molecule thick, its behavior would be considered to belong in what’s called a 2D gas regime, where the mobility of the molecules is more important. However, if the layer is thicker, then it would belong to a liquid regime, where the viscosity of the liquid plays a more dominant role. The physics of each is different.
In the 2D gas regime, the water molecules would make the ice slippery by rolling and moving around, similar to marbles that roll out from under the feet of a hapless stooge in a slapstick comedy.
On the other hand, a film of liquid water would make the surface slippery by lifting the skate off the solid ice. Then the amount of friction would mostly be determined by the viscosity of the liquid: the lower the viscosity, the lower the friction.
“Some of the scientists still disagree on what regime this thin layer of water molecules belongs to,” said Rinse Liefferink from the University of Amsterdam in the Netherlands. He is an author of a Physical Review X paper published on the topic earlier this February.
Although, he added, “it may seem like a semantic debate.”
So, who’s right?
In science, the most unsatisfying answer is also often the truest: It depends.
“Different experiments often have different results because they are based on different conditions,” said Liefferink, “and because different conditions can result in different mechanisms. It makes it a bit hard to say something general about ice friction.”
Scientists have long come up with possible explanations for how the water molecules get separated from the ice below. After all, there are only a few ways this can happen, either from the pressure of the skate pressing on the ice, or from the heat of the skate rubbing against the ice, or from microscopically rough contact between the blade and the ice surface.
But the relative significance of each mechanism would depend on the conditions, such as the shape of the skate, the speed and weight of the skater, the path the skate is taking, the temperature of the ice and air, the presence of existing liquid water on the ice surface, and even if the ice has been melted and refrozen. So, one can have very different answers for the question of “why is ice slippery,” and none of them has to be wrong.
If friction is an empirical property, it may be more practical to figure out how to predict its property than to understand the fundamentals of the phenomenon.
In the “why does spoiled milk smell bad” example, it may be more practical to answer the more focused question of “how can we predict and control for milk spoilage?”
In the case of slippery ice, that question would be “how much do different conditions affect the friction experienced by a skater on ice?”
First, let’s look at the temperature of the ice.
When the ice is very cold, fewer water molecules can be knocked loose. When the ice is very warm, too many water molecules can be knocked loose, and the skate would carve out a trough in the ice and increase the friction. And so, in between the two temperature ranges lies a “sweet spot” where the skates would glide with the least resistance.
What about the shape of the skates?
The sharper the skates, the easier it would be for them to carve into the ice and form a trough, so the “sweet spot” temperature would be colder for a sharper skate, because the ice would need to be harder.
Okay, what else?
The speed of the skater also affects the friction. The ice would appear “harder” when the skates are gliding faster. Think of how hard water feels from a 6-foot drop compared to a poolside jump.
The weight of the skater also plays a part, since it changes the pressure the skates are applying to the ice.
Other factors that affect how slippery ice is include if there already is a layer of water on the ice, if the ice has been melted and refrozen, how smooth the ice surface is, what path the skates take when turning, what the cross section geometry of the skates looks like, what the skates are made of, and more.
While it is already difficult to control for each of these parameters during experiments, another layer of complexity is added to the problem when you consider that some of the parameters may also interact with others, and result in something more than the sum of the parts.
The dilemma of controlling for chaos
“Sliding situations are usually not controlled, so why are we studying it in a controlled manner?” said Karlis Gross, a materials scientist from Riga Technical University in Latvia.
Gross had been studying the influence of weather conditions on skeleton bobsleds, such as the effect of air temperature and humidity on the speed of a bobsled. “Skeletons, bobsleds and luges are very popular in Latvia,” he said.
“When we pushed a sled down the slope, we noticed that the sled was vibrating up and down and side to side the entire time,” said Gross. “The vertical vibrations mean that the blades were whapping down on the ice all the time, which would deform the ice in a way that isn’t accounted for in controlled experiments.”
In many of these “why is ice slippery” experiments, the setup involves a mechanical probe attached to a motorized arm moving across an ice surface. Because of the limitation of the instruments, some probes could only go in a circular pattern, tracing over the same track over and over, while some perhaps better ones could go in a spiral pattern to avoid retracing. However, Gross thinks both setups are too well controlled and the difference between them and real-life situations may be significant.
With the sheer number of things that may affect how a steel blade slides on ice, it is practically impossible to monitor and control for them all, especially given the limited experimental resources.
While some potential applications may come out from these studies, “it is mainly driven by curiosity, at least to me,” said Liefferink.
And we have only talked about the physics of a steel blade on ice. If you’re looking for an explanation for how rubber tires lose their grip on snowy roads, you’ll have to slide down a completely different rabbit hole.