N-Phenylanthrinilic acid (Aldrich, as received),
Phosphorus oxychloride (Aldrich, as received, or recycled from previous experiment)
Ammonium hydroxide,
DCM,
ice.

N-Phenylanthrinilic acid (70 g; 0.33 moles) was stirred for 2.5 hours at room temperature in phosphorus oxychloride in a large 3 necked flask fitted with an overhead stirrer and 2 reflux condensers. The mixture was warmed slowly using a heating mantle until the reaction temperature was self-maintained. The mixture was then refluxed gently for 2 hours, in which time the reaction mixture turned very dark. The reaction mixture was cooled, and the phosphorus oxychloride removed under reduced pressure. The residue was slowly added (CARE) to a mixture of ice (100 g); aqueous ammonia (200 mL) and DCM (200 mL). (Quantities may be increased if effervescence continues). The mixture was filtered through Celite, then the aqueous layer was extracted with DCM (3 x 200 mL), and the combined organic layers were dried (sodium sulfate) and concentrated. The solid product was recrystallised from acetone to give a yellow or brown solid.

The stirring for 2.5 hours prior to heating leads to a much calmer reaction. Otherwise, the reaction can get out of control even with close supervision. The reaction gives off a lot of HCl, so it is advisable to bubble this through water or aqueous sodium hydroxide. It is extremely important to ensure no suckback occurs, though. Destroying removed phosphorus oxychloride is extremely unpleasant and time-consuming (water or ice is added VERY slowly), and we found it preferable to bottle it and use it again on repeating the reaction. There seemed to be no detrimental effect on yield doing this. On one occasion when the reaction was performed, the reaction occurred very cleanly, with no need for filtering through Celite or other purification, giving yellow solid. However, most of the time a dark solid was obtained. On the occasion that the pure material was obtained there was a slight difference to the procedure: the black residue was dissolved in DCM prior to adding to the ice/ammonia mix. It is not know if this was responsible for the cleaner reaction, but may be that the more dilute addition prevents 'burning' through overheating. (Note that the clean reaction was performed using recycled phosphorus oxychloride, though) The reaction has been performed on a scale of up to 140 g, on a yield of up to 90+%, although lower yields sometimes occur, especially if the reaction has been sprayed liberally around the fume cupboard. The 'recrystallisation' was not a true recrystallisation, as not all the solid dissolved.

1H nmr (CDCl3) 7.7 (2H, td, 6.6, 1.3), 7.85 (2H, td, 6.6, 1.4), 8.25 (2H, d, 8.7), 8.5 (2H, dd, 8.7, 0.6).

M. J. Ellis and M. F. G. Stevens, J. Chem. Soc., Perkin Trans. 1, 2001, 3180.

G. D. Jaycox, G. W. Gribble, M. P. Hacker, J. Heterocyclic Chem., 1987, 24, 1405. A. Albert, The Acridines, Edward Arnold (Publishers) Ltd., London, UK, 2nd edn., 1966, pp. 50–51.