I recently ran a several RNA cleanup methods to find the one that worked best in my hands; they were as follows. All were precipitated at -20 overnight and had three wash steps:

• from a lab mate, chloroform for phase separation (originally part of a Trizol extraction, though I used mirVana which is column-based but still has an acid phenol step) followed by isopropanol, ethanol, then sodium acetate+ethanol;
• from LifeTech, ethanol + sodium acetate, two washes with just ethanol; initial spin down was 2-3x longer than other methods (30 min)
• from an RG poster, isopropanol at RT + sodium acetate, then two washes with ethanol;
• from an RG poster, ethanol + sodium acetate, then two washes with ethanol.

I tried each of these with and without glycogen for a total of 8 conditions. I only used sodium acetate, even though my lab mate's protocol actually called for ammonium acetate, and I found a resource that said that iso + ammonium is better than just ethanol at precipitating RNA: "While isopropanol is somewhat less efficient at precipitating RNA, isopropanol in the presence of NH4+ is better than ethanol at keeping free nucleotides in solution, and so separating them from precipitated RNA." (https://www.lifetechnologies.com/us/en/home/references/ambion-tech-support/nuclease-enzymes/general-articles/working-with-rna.html)

The RNA I was starting out with had a 260/280 of ~2.2 and a 260/230 of ~1.5, concentration was ~125 ng/ul.

In 3/4 of the bulleted conditions above, glycogen increased my yield 2-3 times, but also lowered 260/230 by 0.1-0.7 units. Yields ranged from 13-25 ng/ul (my labmate's protocol with and without glycogen) to 110-120 ng/ul (the LifeTech ethanol method with long spin, and the RG poster's isopropanol at RT method, both WITH glycogen).

The samples I will eventually extract have been in the -80 freezer for two years and were flash frozen in liquid nitrogen. I worry, therefore, about yield. The eventual application will be qPCR. I'm not sure to what degree the time-related degradation will affect everything.

Anyway, I'm interested in going with the highest-yield protocols, but the lowering of 260/230 with glycogen makes me wonder whether I can still rely on that metric for assessing contamination/purity, or if it will be less reliable once glycogen is added (but with the benefit of increased yield). I've read conflicting things about whether glycogen absorbs in Nanodrop, and am unsure if it interferes at all with downstream applications (at least in terms of kinetics). Many molecular bio sources say it's "biologically inert" but that doesn't necessarily mean it won't interfere with RT, qPCR, etc. Any guidance would be appreciated.