The denser liquid might perhaps be constituted by formic acid (1.22 g/cm3; 20 ºC), ammonium formate, and possibly, a minor amount of water; if the combined effect of low temperature and salting-out might explain phase separation, or at least, a density gradient effect for mixture with formamide (1.13 g/cm3). Commercial formamide is often contaminated with formic acid. Formamide is very hygroscopic. The mentioned contaminants could, in principle, chiefly result from formamide hydrolysis, maybe acid-catalyzed. Formic acid is quite corrosive, and could perhaps lixiviate ferrous iron from the mentioned amber glass container, which should then tend to oxidize to ferric iron, particularly upon air exposure. Please note that the flask taken from the refrigerator (4 ºC) should not be opened while still cold, namely below the dew point of room air.