“…For simplicity, we also assume that there is no free pore water that would instantly convert to CO 2 hydrate in contact with the injected fluid, with probably detrimental consequences for the reservoir permeability. Such settings represent fluid flow between injectors and producers in a class-I GH reservoir. , The input parameters used in our long-term simulations are presented in Table , based on results given in Appendix D (see Table D1, Supporting Information); diffusion coefficients are Arrhenius extrapolations of the constants inferred in run 2 (6 MPa, 277 K) with the activation energies deduced in Appendix E (see Figures E1 and E2, Table E1, Supporting Information). In all tests, the thickness of the initially reformed layer is assumed to be Δ 0 ≈ 10 μm, while three different SSA-equivalent radii r 0 = 69, 242, and 830 μm (i.e., mean respective particle radii ⟨ r 0 ⟩ = 50, 175, and 600 μm at standard deviation γ 0 = 0.42) are explored.…”