Platinum seems to have higher kinetic friction coef. than static - how is this possible?

According to the Wiki Friction entry, Platinum on Platinum (dry and clean) has $$\mu_s=1.2$$ and $$\mu_k=3$$, which means that $$\mu_s<\mu_k$$. This seems impossible to me. The same phenomenon occurs in other materials, like Aluminium-Aluminium and Silver-Silver. Is this due to a large experimental error, or could this be a reality?

Why this seems strange: Lets take a Platinum box weighing $$1N$$ on a Platinum table. The maximal static friction would be $$1.2N$$. What happens when we push with $$F=1.2N$$ and slowly increase the force? By definition it should start moving (speeding up), but $$f_k=3N>F$$, so instead of accelerating, we are slowing down- from an initial velocity of $$0$$!

• It seems strange to me as well. But not for the reason that you wrote (that seems non sensical, honestly). You can't do such an analysis with empirical coefficient and using regimes for which no continuity can be assumed. More on the topic here physics.stackexchange.com/questions/541/…. Commented Nov 10, 2021 at 12:59
• And engineering.stackexchange.com/questions/8719/… has relevance. Commented Nov 10, 2021 at 13:42
• @Alchimista If the reason I wrote seems nonsensical, I'll rephrase. What happens when we apply 2N of force on that piece of Platinum (e.g. by putting it on an inclined plane)? Say it had some initial velocity. It should slow down. Does it stop? Once it stops it should start moving again. Commented Nov 10, 2021 at 13:44
• This is yet another case of what I call "That's Wikipedia for ya." There are a huge number Wikipedia pages that are inconsistent with one another, and even worse, there are many Wikipedia pages that are internally inconsistent. This appears to be one of those. The table in question takes data from multiple sources. Commented Nov 10, 2021 at 13:47
• @RdBasha I think that the linked Wikipedia article used different sources for static vs dynamic friction for the same substance is one of the many things that makes that article fall into my "That's Wikipedia for ya" category. (I also agree with the accepted answer that the value of 3.0 for metal on metal is highly suspect.) Commented Nov 10, 2021 at 15:39