Diarizines (and really, any N=N bonded molecules) are rather sensitive to thermal degradation, so perhaps even in the dry ice/acetone bath your reaction is getting too hot. Thats why you add your hydroxylamine so slowly in the lit procedure. Why dont you try adding by syringe hanging through a septum since its a small volume, and that will allow you to add literally a single drop at a time.
Just so everyone can see
2-(3-(But-3-ynyl)-3H-diazirin-3-yl)ethanol (16). Anhydrous ammonia (20 mL) was condensed into a round-bottomed flask containing (ketone precursor) (2.52 g, 20 mmol) in an acetone-dry ice bath. The mixture was stirred at 35-40 °C for 5 h. The solution was cooled with dry ice, and a solution of hydroxylamine-O-sulfonic acid (2.59 g,
23 mmol) in anhydrous methanol (6 mL) was added over a period of 30 min. The acetone-dry ice bath was removed, and the mixture was refluxed with stirring at 35 °C for 1 h. The remaining ammonia was then allowed to evaporate overnight. The resulting slurry was filtered and the filter cake washed with several portions of methanol.
The combined solution was evaporated in vacuo. The residue of the diaziridine derivative was dissolved in dichloromethane (10 mL) and treated with triethylamine (3.5 mL). A solution of iodine (3.7 g, 28.8 mmol) in dichloromethane (10 mL) was slowly added with stirring until the appearance of a persistent orange-brown coloration. The mixture was chromatographed on a column of silica gel equilibrated with dichloromethane hexane
(4:1 v/v). After successive washing of the column with this solvent and then with dichloromethane containing 5% ether to give 2-(3-(but-3-ynyl)-3H-diazirin-3-yl)ethanol 16 (1.92 g, 70%).