234 PRACTICAL ORGANIC AND BIO-CHEMISTRY 



ESTIMATION OF PENTOSES. 



Like the hexoses, the pentoses reduce Fehling's solution and can be 

 estimated in the same way, if they are present in the solution alone ; generally 

 a mixture of pentose and other carbohydrates is present ; under these con- 

 ditions the pentose can be estimated by converting it into furfural (p. 303) 

 by distillation with hydrochloric acid, and combining the furfural with 

 phloroglucinol and weighing the compound. 



The estimation is usually carried out in the following way : 



A quantity of material is taken which will yield an amount of phloroglucide 

 varying from "03 but not exceeding '3 gm. It i> distilled from a flask, which 

 is provided with a tap funnel and connecting tube with a trap leading to a. con- 

 denser, with looc.c. of 12 per cent, hydrochloric acid at such a rate that 30 c.c. 

 pass over in 10 minutes. The distillate is filtered through a small filter 

 paper as it is collected. When 30 c.c. have distilled over, 30 c.c. of dilute 

 acid are passed through the tap funnel into the flask, and the distillation 

 continued for another 10 minutes. This addition is continued 12 times or 

 until about 360 c.c. of distillate have been collected. 1 To the whole distillate 

 about double the amount of pure phloroglucinol 2 required dissolved in 12 per 

 cent, hydrochloric acid is added and the mixture well stirred. The solution 

 turns yellow, then green, and a precipitate at first green in colour, but ulti- 

 mately black, separates out. The volume is made up to 400 c.c. and after 

 standing for 12 hours the precipitate is collected on a weighed filter, washed 

 with 150 c.c. of water, dried for 4 hours at 100, cooled and weighed. 

 The amount of furfural is (a + -0052) x 0-5185 

 ,, pentose is (a + '0052) x 1*0075 

 ,, pentosan is (a + '0052) x cr8866 



where a is the amount of phloroglucide and -0052 is the quantity of phloro- 

 glucide not precipitated but remaining in solution under the above conditions. 



Methyl pentoses behave in the same way. The phloroglucide they pro- 

 duce is soluble in alcohol. The alcohol soluble portion of the precipitate 

 is returned as methyl pentose. 



Note* Cunningham and Doree 3 have shown that whydroxy-methyl furfur- 

 aldehyde is formed by the action of acids upon hexoses, starch and celluloses 

 in amounts varying from 1-2 per cent. It is slowly formed and does not 

 interfere in the estimation of pentosans, if aniline acetate be used as indicator. 

 It causes inaccuracies in the estimation of methyl pentosans. 



Flohil 4 showed that the furfural in the distillate could be estimated by 

 means of Fehling's solution, either gravi metrically or by the iodometric method. 

 Eynon and Lane 5 thoroughly tested this method and a further improvement 

 was introduced by Baker and Hulton. 6 The furfural-containing distillate is 

 not collected after it ceases to react with aniline acetate containing sufficient 

 sodium acetate, a period of 5 minutes being given to decide upon the negative 

 reaction. Usually all the furfural has passed over in 200-300 c.c. of distillate. 

 10 c.c. of distillate are titrated with ^N alkali. 50-75 c.c. in a 250 c.c. 

 conical flask are neutralised with saturated sodium hydroxide, the solution 

 being kept cold, 20 c.c. of Fehling's solution are added, the mixture diluted 

 to 100 c.c and heated under a reflux in a boiling water-bath for 40 minutes. 

 The cuprous oxide is estimated gravimetrically. A blank experiment with an 

 equivalent quantity of salt is also carried out ; the amount of cuprous oxide 

 from 2-4 mgm. is deducted. 



3 mgm. Cu = i mgm. furfural. 



1 The completion of the distillation may be tested by means of aniline acetate ; a drop 

 of reagent and a drop of distillate are placed side by side on a piece of filter paper; if no 

 red colour appears the distillation is complete. 



2 The purity of the phloroglucinol may be tested for by dissolving a small quantity in 

 a few drops of acetic anhydride, heating almost to boiling and adding a few drops of con- 

 centrated sulphuric acid. A violet colour shows the presence of diresorcinol. 



3 Biochem. J M 1914, 8, 438. 4 Chem. Weekblad, 1910, 7, 1057. 



6 Analyst, 1912, 37, 41. * Ibid., 1916, 41, 294. 



