There is at present no $\Sigma^0_1$ (= provably true if true) or $\Pi^0_1$ (= provably false if false) sentence which is known to be equivalent$^1$ to the Collatz conjecture.
Note that checking whether a sentence is $\Sigma^0_1$, or $\Pi^0_1$, is decidable: it's a purely syntactic property. The issue is the "equivalent" bit.
$^1$Equivalence of sentences isn't a meaningful notion without specifying a "base theory:" two sentences $p,q$ are equivalent over $T$ if $T$ proves $p\leftrightarrow q$. The naive "theory-free" definition of equivalence - "$p$ is equivalent to $q$ if $p\leftrightarrow q$ is true" - leads to all true statements being equivalent and all false statements being equivalent. When we replace "true" with "provable," we get something more interesting, but then we have to specify a theory: provable in what system?
I haven't specified a base theory above. However, there is no reasonable $T$ over which the Collatz conjecture is known to be equivalent to a $\Sigma^0_1$ or $\Pi^0_1$ sentence. If you like, take $T$ to be ZFC.