Abstract: The effects of potassium and sodium on coke degradation are commonly thought to be similar by studying the influence of alkali carbonates on coke gasification, and the amounts of potassium and sodium into the blast furnace (BF) are controlled without considering the differences. But BF investigations indicate that alkali carbonates have decomposed and in coke the potassium content is always larger than the sodium content, where the enrichment of alkali metals is obviously aggravated. In this article it is found by thermodynamic calculations that alkali metals exist as simple substance vapors instead of carbonates or oxides in the alkali enriched regions. Based on that, experiments for testing the autonomic absorption and damage of potassium and sodium vapors on coke with or without carbon dioxide were designed to simulate the alkali enriched regions. Atomic absorption spectrometry (AAS), X-ray diffraction (XRD) and scanning electron microscope-energy dispersive spectrometry (SEM-EDS) analysis reveal that the absorbance and damage of potassium vapor on coke are much larger than those of sodium vapor because of easy combination with ash in coke to form kaliophilite, which induces volume expansion and crack propagation. So it is proposed that the ash content in coke should be low and potassium into BF should be strictly controlled. Coke gasification tests with different contents of potassium vapor show a steep ascent in coke reaction index (CRI) when the mass ratio of potassium vapor to coke in the gas-solid system is above 3%. According to the different absorption and damage effects of potassium and sodium on coke, quantificational control models are constructed for determining the upper limits of potassium and sodium as well as the total amount into BF.