Denote by $X := [\mathbb{N}]^\infty$ the set of infinite subsets of $\mathbb{N}$. Recall that the Ellentuck topology is a topology on $X$ generated by sets of the form $\{A\text{ infinite} \mid s\text{ is an initial segment of }A\text{ and }A\setminus s\subset B\}$ for some finite $s\subset \mathbb{N}$ and (infinite) $B\subset \mathbb{N}$.
In the paper On completely Ramsey sets by Szymon Plewik, it was proved that the Ellentuck topology is not normal. The argument essentially reduced to a construction of a closed separable subspace, say $Y\subset X$, containing a discrete closed subset $Z$ of cardinality $2^{\aleph_0}$. Suppose that $X$ is normal. Because $Y$ is a closed subset of $X$, $Y$ is normal as well. On the one hand, since $Y$ is separable, the set of continuous function on $Y$, denoted by $C(Y)$, is of cardinality $\le 2^{\aleph_0}$. On the other hand, by the Tietze extension theorem, every (continuous) function on $Z$ can be extended to a continuous function on $Y$, and so $C(Y)$ is of cardinality at least $2^{2^{\aleph_0}}$. A contradiction.
I am wondering if one can demonstrate an explicit construction of two disjoint closed subsets of $X$ that cannot be separated by two disjoint open neighborhoods?