Multifunctional carbon‐nanotube supported catalyst for efficient glycerol hydrogenolysis to 1,2‐propanediol

Multiwall carbon nanotube (NC7000) supported mono-, bi-, and trimetallic catalysts were synthesized and tested in a multi- batch reactor setup supported by data-driven modeling for the chemical hydrogenolysis of glycerol to 1,2-Propanediol (1,2-PDO) at 220◦C under 30 bar hydrogen pressure in aqueous solution. ICP-OES, N2 -physisorption, scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy dispersive x-ray spectroscopy (EDX), powder x-ray diffraction (XRD) and x-ray photoelectron spectroscopy (XPS) were used to investigate the physico-chemical properties of the synthesized catalysts. Interesting structure-activity-selectivity correlations could be deduced from the different metal ratios and combinations. While Ru was by far the most active monometallic species ( X = 100%), followed by Pt ( X = 80%) and Ir ( X = 65%), it suffered from a very low 1,2-PDO selectivity ( S < 10%). Combining Ru with the more selective transition-metals RuX2 ( X = Fe, Co, or Ni) maintained the high activity ( X > 95%) but only moderately increased 1,2-PDO selectivity ( S = 20%–30%). Interestingly, combining Ru with Cu in a bimetallic RuCu2 catalyst slightly decreased activity ( X = 81%) but drastically improved 1,2-PDO selectivity ( S = 58%). Moreover, a trimetallic combination of RuMCu2 (M = Pt, Pd, Ir) did not further increase activity ( X = 50%–70%) but could push 1,2-PDO selectivity up to 79% for the trimetallic RuIrCu2 catalyst. Most prominently, a final metal ratio variation for the RuCux catalyst was derived with the help of the data-driven models producing the most promising RuCu3 /NC7000) catalyst combining both high activity ( X = 78%) with a very high 1,2-PDO selectivity ( S = 79%) resulting in the highest 1,2-PDO yield of 58% in a nearly closed carbon balance ( C > 95%). This lays the foundation for implementing this highly active and selective catalyst in a future SMART reactor for glycerol valorization.