Addition to the previous remarks:
ΔQ/T is useful only for heat and work, that is, at engines, for which thermodynamics was invented.
At chemical reactions, other processes occur than heat exchanges, expansion, compression and so on. Then, G and µ decide, rather than S.
And if you mix deuterium and protium (1H2), the process is very much irreversible despite ΔQ=0 and, with the usual definitions, ΔS=0. Other definitions of S would do that one better.