The great challenge in explaining the origin of life is the chicken vs. egg problem. The most primitive life we know of is utterly dependent on a complex web of chemical reactions that depends on accurate transcription of detailed blueprints for the construction of the molecules that make metabolism possible. All known life requires a cell membrane, an elaborate set of metabolic pathways guided by carefully tailored enzymes, DNA blueprints that preserve genetic information, and complex ribosomal factories for manufacturing the proteins specified by the blueprints. It's very difficult to imagine how such a complex, interdependent mechanism could have evolved by selection, since it's very difficult to imagine intermediate states that function well enough to support the degree of faithful reproduction required for evolution.
Nick Lane, writing in New Scientist, explains a new idea that bridges some of the difficult steps required for development of living cells.
My summary can't do justice to the rather extensive article, but the key ideas as I see it are: (1)initial evolution didn't require cell membranes because it took place in natural pores in rocks by hydrothermal vents, (2)the first metabolism was based on proton transport, which is used in all cells, (3)In the presence of free hydrogen (as occurs near such hydrothermal vents), metabolism can occur without elaborate enzymes and energy transport, (4)the pores form a natural environment favorable to the synthesis of nucleic acid precursors.
This far from being a compete story, but it looks to me to come closer than many earlier attempts. Good theories suggest good experiments, and this one suggests a lot.