Scientists recently took a closer look at Stanley Miller’s samples obtained from his famous spark-discharge experiments.
Because our instruments are much more sensitive than the tests Miller used, it was determined these experiments produced a wider variety of compounds than appreciated.
Here are some of the data:
In this figure, the underlined compound are those that were detected with the more sophisticated testing.
As I see it, these data actually underscore the serious problems associated with abiogenesis. The problem is not in the origin of these compounds. Neither is it the origin of the complexity, as the mixture detected was clearly quite complex.
Life universally employs a specific set of 20 amino acids. If you look at the figure, you’ll notice 27 amino acids and other organic compounds. But in that list, only seven are the biological amino acids. So the problem is how these seven were picked out from a mixture of 27. And if you consider only the newly detected compounds, it highlights this problem, as only one out of ten is a biological amino acid.
To drive this point home, consider the following figure.
Valine is one of the twenty biological amino acids, but Miller’s methods also produced roughly the same amount of two of its isomers – norvaline and isovaline – that are not found in natural proteins. So how was valine picked instead of norvaline or isovaline?
Others have long noticed this to be a serious problem. After reviewing origin of life research, Simon Conway Morris, approaches this core problem:
Not only are the yields often disappointingly low, even miniscule, but typically (and unsurprisingly) the experiments produce a wide range of other chemicals that seemingly have no relevance to the origin of life. In some instances a substantial quantity of the organic material synthesized forms a tar-like ‘gunk’, reminiscent of the heterogeneous ‘goo’ found in carbonaceous meteorites. (emphasis added) 
Gerald Joyce, one of the leading researchers of the RNA world, encounters the same problem:
If the building blocks of RNA were available in the prebiotic environment, if these combined to form polynucleotides, and if some of the polynucleotides began to self-replicate, then the RNA world may have emerged as the first form of life on Earth. But based on current knowledge of prebiotic chemistry, this is unlikely to have been the case. Ribose, phosphate, purines and pyrimidines all may have been available, although the case for pyrimidines is less compelling. These may have combined to form nucleotides in very low yield, complicated by the presence of a much larger amount of various nucleotide analogues. The nucleotides (and their analogues) may even have joined to form polymers, with a combinatorial mixture of 2’,5’-, 3’,5’- and 5’,5’-phosphodiester linkages, a variable number of phosphates between the sugars, D- and L- stereoisomers of the sugars, a- and b-anomers at the glycosidic bond, and assorted modifications of the sugars, phosphates and bases. It is difficult to visualize a mechanism for self-replication that either would be impartial to these compositional differences or would treat them as sequence information in a broader sense and maintain them as heritable features. The chief obstacle to understanding the origin of RNA-based life is identifying a plausible mechanism for overcoming the clutter wrought by prebiotic chemistry. (emphasis added) 
In other words, the prebiotic world envisioned as a result of prebiotic chemistry is a very complex state itself, filled not only with some of life’s parts, but even more so with a countless other pieces and parts that are not used by life. Steven Benner, from the Departments of Chemistry, Anatomy, and Cell Biology at the University of Florida, spells out this problem even more bluntly in his review of a book on origin of life research:
Courageous groups began efforts to get RNA to reproduce itself. Despite their sophistication, these–dare we call them “classical”?–approaches did not solve the problem surrounding life’s origin. Prebiotic chemistry could produce a wealth of biomolecules from nonliving precursors. But the wealth soon became overwhelming, with the “prebiotic soups” having the chemical complexity of asphalt (useful, perhaps, for paving roads but not particularly promising as a wellspring for life). Classical prebiotic chemistry not only failed to constrain the contents of the prebiotic soup, but also raised a new paradox: How would life (or any organized chemical process) emerge from such a mess? Searches of quadrillions of randomly generated RNA sequences have failed to yield a spontaneous RNA replicator. This failure raises new questions: Will the elusive replicator emerge after we examine quintillions of random sequences? Or must we add something to RNA to be successful? Researchers have prepared a number of new nucleic acid analogs. With these has come the realization that for a single biopolymer, capable of both genetics and catalysis, to sustain Darwinian evolution, it must be Capable Of Searching Mutation-space Independent of Concern over Loss Of Properties Essential for Replication. This “COSMIC-LOPER” behavior is now known to be scarce in molecular systems. (emphasis added) 
The new discoveries concerning Miller’s samples simply indicate that the ‘gunk’, ‘clutter’, and ‘mess’ is more extensive than we thought.
1. Conway Morris, S. 2003. Life’s Solution: Inevitable Humans in a Lonely Universe. Cambridge University Press; Cambridge.
2. Joyce, GF. 2002. The antiquity of RNA-based evolution. Nature 418:214-221.
3. Benner, SA. 1999. Origins Of Life: Old Views of Ancient Events. Science 283: 2026.