The localisation of the ring $R$ at the element $s \in R$ is the ring of sums of 'words' in the language $R \cup \{x\}$ modulo all relations that come from $R$ and the relations $xs=sx=1$ and $1x=x1=x$. Elements look like the following sort of thing. . . .
$xax^2b$
$x^3 + axbxc$
$abxcx^2dx + b$
. . . where $a,b,c$ and $d$ are elements of $R$.
Modulo our equivalences we have $xax^2b = xax^2bsx$ for example.
In case $s$ has some nice properties, the localisation has nice properties. For example the element $s$ is invertible and the localisation is a nontrivial ring (Most authors only define the localisation when $s$ is this nice).
You want to show the ring defined above is in fact the zero ring for your example. You can do this using the equivalences. It is enough to show each 'word' is equivelent to zero. That can be done using how your $s$ is a right and a left zero divisor, and performing some symbol manipulation.
Edit: It is even enough to prove the subring $R \subset R_S$ is the zero ring.