Density functional theory (DET) calculations were used to study the mechanisms of dehydrogenation of formic acid catalyzed by two Ru complexes, including P(CH2CH2PPh2)(3)RuH+ (P(CH2CH2PPh2)(3) = tris[2(diphenylphosphino)ethyl]phosphine) and P((CH2CH2CH2PPr2)-Pr-i)(3)RuH+ (P((CH2CH2CH2PPr2)-Pr-i)(3) = tris[3(diisopropylphosphino)propyllphosphine). Two competing catalytic cycles (I and II) have been explored. In cycle I, the catalytic reaction starts with a direct hydride transfer from HCOO- to a Ru center, releasing CO2 in the first place, followed by H-2 production, while in cycle II, neutral formic acid approaches Ru catalyst to produce H-2 molecule prior to CO2 generation via beta-hydride elimination. The computational results show that cycle I is more accessible than cycle II, regardless of the ligands surrounding Ru center. The calculated overall Gibbs free energy barriers for formic acid dehydrogenation catalyzed by P(CH2CH2PPh2)(3)RuH+ (14.3 kcal/mol) is significantly lower than its iron analogue in the previous experimental and theoretical studies (18.9 kcal/mol) at the same level of theory. More interestingly, isopropyl group, a more electron-donating group than phenyl group, leads to an even lower reaction barrier of P(CH(2)CH(2)CH(2)PiPr(2))(3)RuH+ (12.9 kcal/mol) than P(CH2CH2PPh2)(3)RuH+, indicating a more positive role the ligand could potentially play in the performance of the catalysts. Our results pave a new way to design more efficient catalysts for formic acid dehydrogenation. (C) 2016 Elsevier B.V. All rights reserved.