These simulations are quite accurate reproductions of the dense cloud interstellar environment. We use big pumps to reproduce the vacuum (10^-8 torr = 10^-12 atmospheres), and we have a special fridge that uses helium in place of freon that allows us to cool our samples to just 10 to 15 degrees above absolute zero. Finally, we use a hydrogen plasma as our source of UV light, so we make it the same way that its made out there. Click here to see what these machines look like. To see an artists representation of how the experiment is performed visit the scientific american web site here.
Finally, our ice is made of simple molecules that are well known to be present in the interstellar medium at the same proportions that are seen (by infrared astronomy) on ice grains. So our water rich ices are very reasonable. We cite specific examples in the paper...
"These solid mixtures are representative of the composition of interstellar ice mantles in dense clouds and towards forming stars (protostars). For example, relative to H2O, NH3 has been observed at 10-30% in the ice around the massive protostars NGC 753817 and GCS3,18 and CH3OH has been commonly observed in clouds around forming stars.19 Thus, H2O, CH3OH, and NH3 are reasonable starting materials because they are among the most abundant molecules frozen onto grains in the dense ISM. In addition, it is reasonable to include HCN since it is abundant in cometary coma and the dense ISM,20 where the majority of HCN should be frozen onto grains. Other simple molecules present in interstellar ices at comparable or greater abundances (CO, CO2) rapidly form in-situ in our experiments as a result of UV photolysis.13,21 "
The fact that these water rich ices produce smaller amino acids is OK with us. these are the amino acids that are deuterium enriched in meteorites. More fancy amino acids often are not seen or are not enriched, so we are happy with what we have.