From a science standpoint, there is always the potential to expand knowledge about many subjects through detailed experimentation. One of the first challenges will be managing the considerable number of variables and conditions relative to the environment, chemical conditions, atmosphere, temperature, water conditions and sources of energy that existed on early Earth, but which are not well known.
Regardless of whether the initial conditions could even be perfectly replicated, there are no natural processes that can chemically assemble the required elements into increasingly more complex biochemical assemblies that could eventually lead to even the simplest initial organism.
As far as the the experimental design steps that "would best explain the origin of life," there are none within the context of a single experimental setup that could accomplish the task. More importantly, because no natural conditions suffice, it would take extreme levels of intelligent design in the configuring and processing at all phases and steps of the experiments, making no natural condition simulating primordial Earth realistic or possible.
The destruction of molecules (assumed to mean disassembly into elements and smaller molecules) is not systematically precise enough to produce the right concentrations of the right gases.
Using viruses grown from plants is not viable as it requires the use of living organisms to accomplish an outcome of dubious relevance.
Creating organisms is intelligent design, not natural working processes and is a totally independent activity whose comparability to the true initial conditions is undetermined.
Breaking down fossils is not viable because fossils do not contain all required materials and compounds required to emulate initial conditions.
The net is that the range of experimentation will demonstrate intelligent design, not emulate initial conditions, and require an extremely wide range of efforts with no guaranteed result.
