Life on earth is a tremendously complex process and, independent of whether one believes in one god or the other or not, evidence accumulates that this complexity originates from an evolutionary process. [1-3] According to the dogma of molecular biology DNA is to be transcribed into messenger RNA, a rather transient active copy, which is translated into proteins by using transfer RNAs (tRNA) as adaptors.  This dogma states that proteins are the only class of macromolecules that carry out catalytic functions. But how could such a complex system have evolved from a ‘primordial soup’? A significant modification of the central dogma of molecular biology was caused by the Nobel prize winning discovery of RNAs that are capable to catalyze biochemical reactions without the need of a protein component [5, 6] , giving rise to a theory of molecular evolution based on an RNA world. [7, 8] But even if you go for the RNA world hypothesis the initial question is only transferred from proteins to RNA: How could large, complex RNAs evolve from the vast ocean? This question leads to
The concentration problem:
Any kind of (bio) chemical reaction can only proceed to high yield if the concentration of starting material is high and best yields are achieved when the products either catalyze further reactions or when product and starting material get separated, resulting in a constant pull due to the continuing disruption of any equilibrium. Alas the concentration of organic molecules in the primordial ocean is thought to be similar to the one in the cotemporary ocean, which means too low to support life. 
The Contemporary solution to the concentration problem:
In contemporary life forms the concentration problem is solved by compartmentalization: Bacteria and
Archaea are equipped with a cell membrane that forms there outer hull, while Eukarya possess even additional sub compartments. Consequently the first life forms, so-called protocells, would consist of an outer membrane and an informational and functional biopolymer, i. e. RNA.  It could be shown that fatty acids (the ‘ancestors’ of present phospholipids) can self-assemble into lipid membranes and thereby form compartments that are capable to internalize new nucleic acid building blocks (nucleotides), while retaining the copied biopolymer. 
Thermophoresis as a possible primordial solution of the concentration problem:
This ‘membrane first’ approach would introduce an additional quite unlikely event in the schedule that would ultimately result in the evolution of life. As with basically any unlikely event there of course exists a competing theory, which in this case relies on inorganic compartmentalization as the cradle of life.  Several years ago Baaske et al. proposed an especially elegant approach to inorganic compartmentalization in the RNA world: The authors applied their recently developed theory of thermophoresis in aqueous solutions  in simulations on nucleotide diffusion in pore systems of hydrothermal vents at the bottom of the sea.  Thermophoresis describes movement of molecules in a temperature gradient: Heat of a specific source (here a hydrothermal vent) dissipates in solution and the resulting temperature gradient facilitates molecule accumulation or depletion in the heat source, depending on the nature of the molecule investigated.  The pore system of the vent would not only supply compartmentalization is this scenario but a whole network of compartments that are connected by thermophoresis.
The result of the simulations of Basske et al. was: By an interplay of solvent transport by convection and thermophoresis single nucleotides could be accumulated more than 10 8 -fold, while polynucleotides were concentrated even more, depending on their length and the pore geometry.  The authors note that their model already supplies a setting of temperature oscillation like it is used in exponential DNA amplification by the Polymerase Chain Reaction (PCR). Herewith a possible mechanism of mono- and polymer concentration was developed but an important question remained: It was unclear whether any self-replication of nucleic acids would be possible in the hydrothermal pore system. Obermayer et al. could address this question by a theoretical approach  , while Mast et al. succeeded recently in addressing this system experimentally in a DNA system. 
It seems like the thermophoresis model is indeed capable to compete with other model of evolution and we can be looking forward to the studies to come.
 J. E. Barrick et al., Nature 2009, 461, 1243-1247.
 D. Brawand et al., Nature 2011, 478, 343-348.
 F. C. Jones et al., Nature 2012 484, 55-61.
 F. Crick, Nature 1970, 227, 561-563.
 C. Guerrier-Takada, K. Gardiner, T. Marsh, N. Pace, S. Altman, Cell 35, 3, 1983, 849–857.
 K. K. Kruger et al., Cell 1982, 31, 1, 147-157.
 W. Gilbert, Nature 1986, 319, 618.
 G. F. Joyce, Nature 1989, 338, 217-224.
E. V. Koonin Proc Natl Acad Sci USA 2007, 104, 9105-9106.
J. P. Schrum et al. Cold Spring Harb Perspect Biol 2010, 2, a002212.
Mansy et al. Nature 2008, 454, 122-5.
S. E. McGlynn et al., Phil. Trans. R. Soc. A 2012, 37, 1969, 3007-3022.
 S. Duhr, and D. Braun, Proc Natl Acad Sci USA 2006, 104, 22, 9346-51.
 P. Baaske, F. M. Weinert, S. Duhr, K. H. Lemke, M. J. Rusell and D. Braun,s Proc Natl Acad Sci USA 2007, 104, 22, 9346-51.
 B. Obermayer, H. Krammer, D. Braun and U. Gerland PRL 2011, 107, 018101-1-4.
C. B. Mast et al. , Proc Natl Acad Sci USA 2013, 110, 20, 8030-5.