If $x > 1$ then does the set $ \lbrace{ \lbrace{ x^n\rbrace}: n\in\mathbb{N} \rbrace}\ $ always contain either $0$ or $1$ as a limit point?

55 Views Asked by At

This question is somewhat related to my previous question here.

Here, $\lbrace{ \cdot \rbrace}$ means fractional part.

Is it true that if $x > 1\ $ then either $0$ or $1$ (or both) are accumulation points of the set $ A_x:= \lbrace{\ \lbrace{\ x^n \rbrace}: n\in\mathbb{N}\ \rbrace}\ ?$

I think this is true for Pisot Numbers at least, because, for example, $(3+\sqrt{5})^n + (3-\sqrt{5})^n \in\mathbb{Z}.\ $ and $\ \vert 3-\sqrt{5}\vert < 1.$

Also relevant is here, where it says,

The sequence $\displaystyle (\alpha ,\alpha ^{2},\alpha ^{3},\dots )$ is completely uniformly distributed for almost all $\alpha >1.$

Unfortunately this does not decide whether or not $A_x$ contains either $0$ or $1$ as a limit point for every $x>1.$ And also the above fact does not tell us that it is not known whether or not $0$ or $1$ is a limit point of $A_x$ for every $x>1.$ So the question remains.

I have some vague idea about considering the interval $(\liminf A_x, \limsup A_x)$ and coming to a contradiction, but I don't have anything concrete yet.