Direct rocky planet formation without protostellar disk?

Question

Inspired by this question, which sort of asked what I'm asking, but emphasized a different point (and the answers accordingly addressed that).

Can a rocky planet form directly from the collapse of an interstellar gas cloud?

The picture I have of rocky planet formation is that dust and rocky material in a protostellar disk accumulates into a rocky planet. There must be some mechanism to prevent the accretion of significant amounts of (hydrogen/helium) gas, I'm guessing this is something to do with the gas either being unable to cool, reheated if it does cool, or outright removed from some region around the protostar as the stellar wind and/or radiation pressure kick in.

My intuition is that even the most metal enriched (astronomer metal, i.e. everything heavier than $\rm{He}$) cloud still by far hydrogen dominated, so that a cloud in isolation must always form a primary star/brown dwarf/gas giant. But has anyone ever heard of a mechanism where a cloud in the right environment (some radiation field that efficiently heats the gas but not the dust/metals perhaps?) could form a rocky body directly, with no associated star? This seems far fetched, but I'm curious to know whether anyone has ever looked at this; literature references appreciated.

Answer 1

A recent paper suggests a mechanism to create totally metal stars.

Some Stars are Totally Metal: A New Mechanism Driving Dust Across Star-Forming Clouds, and Consequences for Planets, Stars, and Galaxies http://arxiv.org/abs/1406.5509

Ordinary star formation produces planetary mass objects (sub-brown dwarfs), so perhaps this new mechanism would also do so.

Answer 2

Molecular clouds must have mass greater than the Jeans limit in order to collapse. Brown dwarfs resemble planets without an accompanying star, according to this NASA press report, but the more likely scenario is that a protostar forms and the rocky bodies form from the disc that surrounds it before the onset of fusion.