Some observations on the chemistry of KO2—DMSO solutions