DISTILLATION OF STUMPWOOD. 59 



recovery of pine oil to the point where the oil constitutes but 5 

 per cent of the distillate coming over. The odor, color, refractive 

 index, density, where possible, and volume of each fraction thus ob- 

 tained by the different methods of treatment were noted in order to 

 determine by the comparison of these constants which process gives 

 the closest separation and best yield of high-grade product. 



As was to be expected the best results were obtained by the use of 

 caustic soda. With carbonate of soda, used in such proportion that 

 its hydroxyl strength was equivalent to that of the hydrate, the quan- 

 tity of the turpentine recovered from the crude oil was the same as 

 that obtained with caustic soda, but of inferior quality with respect 

 to odor. For commercial use, moreover, the fact of its being cheaper 

 than the hydrate is offset by its greater equivalent weight and the 

 correspondingly larger quantity required to produce the effect of an 

 equivalent amount of sodium hydrate. Milk of lime has only low 

 cost to recommend it. The calcium resinate or lime soap formed, 

 being insoluble, does not form the pine-oil emulsion that helps ma- 

 terially to effect a sharp separation of the turpentine. The yield of 

 the turpentine is lower by 10 per cent than when sodium hydrate is 

 used, and the product is inferior in odor. Moreover, the lime soap 

 seriously fouls the apparatus with an incrustation difficult to remove. 



It was found that the quality improved and the percentage of tur- 

 pentine recovered increased with increasing amounts of alkali up to 

 75 cc. of 20 per cent caustic-soda solution per 500 cc. of crude oil, or, 

 in industrial terms, 75 gallons of 20 per cent caustic-soda solution 

 (containing 20 parts per hundred of actual sodium hydroxid) to 500 

 gallons of crude turpentine. This proportion was found to be satis- 

 factory for refining' the crude second turpentine. For the crude first 

 turpentine the amount of alkali probably could be diminished. The 

 concentration of the alkali solution is not so important, since the use 

 of half this quantity of 40 per cent alkali solution does not materially 

 affect the results. The duration of the chemical treatment before be- 

 ginning the distillation is of great importance, and at least 30 min- 

 utes after the mixture reaches the boiling point should be allowed for 

 the completion of the reactions involved. Separation of the alkali 

 solution from the turpentine before distilling has a profound effect. 

 Not only is the quality of the turpentine much inferior to that of the 

 turpentine obtained when the distillation is made in the presence of 

 the alkali solution, but the yield is lower by 20 per cent, with a corre- 

 sponding increase in the second turpentine and pine-oil fractions, 

 showing that, however brought about, the soap solution exerts a re- 

 straining influence over the pine oil, the complete separation of which 

 from the turpentine is most essential to the production of an article 

 possessing the properties demanded by the trade. Passing air through 



