In a
2-l. round-bottomed, three-necked flask fitted with a
stirrer and
two large-bore condensers are placed
200 cc. of 50 per cent nitric acid and
0.25 g. of vanadium pentoxide. The flask is heated to 65–70° in a
water bath (
thermometer in the water), and
1 cc. of cyclopentanone is added. Oxidation is indicated by the production of brown fumes. The water bath is removed, and
42 g. (less the 1 cc.) of the cyclic ketone is added from a
dropping funnel through the condenser at the rate of a drop every three seconds. The heat of the reaction keeps the flask at about 70°. If the temperature drops, oxidation ceases until the ketone has accumulated, when it may proceed almost explosively. In such a case, or if the temperature is higher, much
succinic acid is formed. After addition has been completed, the water bath is replaced and the mixture heated to boiling. The contents of the flask are poured into an
evaporating dish (
Hood), and the volume reduced one-half. When cold, the acid is filtered and the operation repeated twice; the last time the acid is yellowish, and the color is removed by washing with dilute
hydrochloric acid. The crude
glutaric acid is white and weighs
50–55 g. (
80–85 per cent); m.p.
92–94°. If any
succinic acid is present owing to lack of proper control, it separates in the first crop. It is more convenient to allow the mother liquors from several runs to accumulate and work them up separately; from each run
2–3 g. more of
glutaric acid may be so obtained. Further purification is accomplished, if desired, by crystallization from
benzene. The acid as prepared above always contains traces of
nitric acid, but is satisfactory for conversion into the anhydride. If the catalyst is omitted, the yield is less by
10 per cent. (C. F. H. Allen and W. L. Ball, private communication).