To control the chemically active boron (B) in 9CrMoCoB within a narrow target range, thermodynamic calculations using FactSage 7.2 and equilibrium reaction experiments were conducted between slag and steel. First, equilibrium reaction experiments between 9CrMoCoB and slag of 55%CaF2–20%CaO–3%MgO–22%Al2O3–x%B2O3 (x ≤ 3) were conducted to supplement and optimize the Henrian activity coefficient γ (lgγ = A/T + C, where A and C are constants) of B2O3 in the solution phase of FToxid-OXYL in FactSage 7.2. Second, the equilibrium reactions between 9CrMoCoB steel and slag of 55%CaF2–20%CaO–3%MgO–22%Al2O3–x%SiO2–y%B2O3 (x ≤ 3, y ≤ 3) were calculated by using FactSage 7.2 to study the effects of B2O3, SiO2, and O2 in the atmosphere on the contents of B, Si, and O in steel. Then, equilibrium reaction experiments were conducted to validate the calculated results. Finally, suitable slag was designed, and electroslag remelting (ESR) experiments were conducted for verification. Results showed that A = −6582 and C = 0 in the Henrian activity coefficient equation were suitable to the slag of 55%CaF2–20%CaO–3%MgO–22%Al2O3–x%SiO2–y%B2O3 (x ≤ 3, y ≤ 3), and the appropriate slag for ESR remelting the 9CrMoCoB steel was 55%CaF2–20%CaO–3%MgO–22%Al2O3–1%SiO2–1%B2O3. The volume percent of O2 had a stronger influence on the Si content, compared with B content. The volume percent of O2 should be less than 0.05% to produce the qualified 9CrMoCoB ingot when using the designed proper slag.