The selective hydrogenation of conjugated dienes in monoolefine-rich
steam is an important process to eliminate dienes from monoolefins
in petroleum refining, where the discovery of a highly active, selective,
and stable Pd-based alloy is beneficial to its large-scale application.
Herein, we report an experimentally validated theoretical framework
to discover promising Pd-based bimetal catalysts for 1,3-butadiene
selective hydrogenation to butene, which combines density functional
calculation, descriptor-based microkinetic modeling, and Wulff construction
principle. Since the activity and selectivity of 1,3-butadiene hydrogenation
on Pd-based bimetal surface could be expressed by H and CH3 adsorption energy, our theoretical framework efficiently predicts
the desorption rate of butene on equilibrium-state Pd-based bimetal
nanoparticles. After high-throughput screening on the second component,
the PdW nanoparticle, of which butene production is mainly contributed
by PdW(100), is predicted to be promising and experimentally proven
to outperform Pd in selective hydrogenation of the butene-rich C4
fraction, including butane yield and long-term stability. The work
demonstrates the importance of comprehensively considering the contribution
of diverse crystal surfaces to catalytic performance when high-throughput
screening on an alloy component. The outlined approach is general
and applicable to a range of different hydrogenation reactions over
alloy catalysts.