crystallization from 60% methanol. The average ch,tiii length from reducing value was 14.4 glucoie units and the material was completely converted t o maltose by @-amylase. Measured amounts of standardized solutions of the dextrins were allowed t o react with measured quantities of periodate under various conditions. Analyses for the concentration of unreacted periodate were carried out by titration of the iodine liberated on addition of excess potassium iodide and standard arsenlte t o the bicarbonate buffered solution. Tests for formic acid were made by destroying the periodate with an excess of propylene glycol and titrating with 0.1 N sodium hydroxide to the phenolphthalein end-point. The alkali titrations of 15-inl. aliquots of the oxidation mixtures were equivalent to those obtained on periodate blanks, indicating that no formic x i d was produced. The dimedon test12 did not reveal the presence of formaldehyde in any of the digests. Total Consumption of Periodate and Rotational Changes. -Conditions necessary to avoid over-ouidation were ascertained by treating 0.01 JI @-dextrin with varying amounts of sodium periodate ,it room temperature. The reaction WdS followed by measuring the periodate consumption (Fig. 1) as well rlb the change in optical rotation; s: present the apparent r0t.twhen an excess of oxidant w tiom dropped to about -11 , then very gradually ro,c -3
t o 4". The addition of buffers to the oxidation mixture3 caused wide differences in the optical rotations ohwrved. With either sodium or potassium periodate, no added bnf fer, a i d pH about 4, the rotation dropped to - 11 '. In the presence of a pH 6.5 potaisium phosphate buffer the final rotation was 75-75", and with sodium periodate buffered with acetic acidTodium acetate t o pH 4.1 or 5.5 the final rotation was 50 . Since no appreciable amount of acid was produced in the reaction, there was very little shift in the pH of reaction mixtures without additional buffer, This cffcct of buffers on rotation was not investigated further. Following the prcliiiiinary experiments with @-dcxtrin, solutions of CY-, 13- . I T I ~ y-dextrins were oxidized under identical condition\ with 1.1 moles of sodium metapcriodate/glucose residue. Thc oxidations were carried out at 28' without added huffLr. The rotational change., .tiid periodate consumption tire plotted in Figs. 2 and 3. After four days, the consumption of periodate was not greater than 1.03 nioles/glucose residue. Kinetic Analysis.-The order of the initial p'srt of the reaction a t 4' was determined in 25-minute oxidatiotis with beta dextrill:
Po
.-
APi2.i miu.
0.00037 .00354 ,04975 ,00100 Order in periodate = 0.90 .OO,X)6 ,OtcifN .0007O Order in Schardinger clextrin = 0 . 9 2 Oxidation of dextrin-pcriodate mixtures (dextrins cu . 0.04 [I. with respect t o glucose residues) was then followed a t 4" over 2-3 days, the amount of periodate consumed determined by titrating aliquots, and the data used for the determination of the rate constants as explained ;tbove. The values of the second order rate constants kI and k2 so obtained are: a,kt = 9.0, kr = 5.4; B, k L = 33, kd = 8.3; y , kI = 105, k2 = 21.0; all expressed as liters X mole-1 X see.-' X 10-5 a t 4", with an uncertainty of about 5% of the stated values for each constant. The curves calculated for these constants together with the experimental points appear in Fig. 4. Cornparuble data for a Nageli-type amylodextrin of average chain length approximately 14.4 glucose residues are compared (Fig. 4) with a ctirve calculated on the assumption that all the glucose residues are oxidized a t the same rate; &, = ti5 X 10 -5 liters X moles-' X set.-'. Effect of Iodate on Reaction Velocity.-To each of five flasks was added sufficient sodium periodate and @-dextrin so that the initial concentration after dilution was 0.0133 JI: and O.OO1fit 121, respectively, together with sodium iodatc to the crtent of 0, 0.0056, 0.0140, 0.0283 awtl O.O-L20 Af. After 61 ruin. a t 4' the remaining periodate was determined to be the same in each flask: 0.0126 =t 0.0001 ,If.
Summary Oxidation of the a-, 0- and y-dextrins with sodium periodate has been characterized as producing no formic acid or formaldehyde and consuming one mole of periodate per glucose residue. The kinetics of the reaction have been analyzed by an approximate method and up to about 40% total oxidation agree with curves calculated for an initially hindered reactioii followed by a more rapid oxidation. The rate of the initial oxidation increases in the order: alpha, beta, gamma. The initial oxidation of the y-dextrin is still subject to an inhibition not characteristic of a low molecular weight amylodextrin. AMES, IO\VA
(12) Vorlander, %. anal. Chciir , 17, 241 (1929).
I C O C I X I B U l I O N FROM THE
5.
O.I)c)ZS4 U , O l f j t i O
RECEIVED APRIL 28, 193U
IOIVA AGRICULTURAL E X P E R I M E N T Sl'Al I O V ]
Studies on the Schardinger Dextrins. VI. The Molecular Size and Structure of the 7-Dextrin BY 1)E:XTEK I ~ K I & N C Ino~rs I, I$-.KN.\PP .\ND J. €1. I ' A L L i R The application 01 crystallogrraphic procedures for the determination of molecular size to the a- and p-dextrins indicated that these are composed of six and seven glucose residues, respectively.? A similar study has now been directed toward the establishment of the size of the yd e ~ t r i n ,previously ~ regarded by Freudenberg4 (1) Journal Paper N o J 1782 of the Iowa Agricultural Expenment Station, Ames, Iowa. Pro] 111G, supported in part by a grant from the Corn Industries Research Foundation (2) French and Rundle, THIS JOURNAL, 64, 1661 (1'343) (3) French, Levine, Pazur and Norbcrg, ibid , 71, 333 ( 1 9 1 ' 1 ) (1) (a) Preudenherg and Jacobi, A w , 618, 102 (1' l'rcurlenbeig, Pldukenhorn and Knauber, C h e ~a i ~d tiid ~ ,7
hepiasaccharide. I n the present study it was not found possible to determine unanibiguously the number of glucose units per molecule of y-dextrin by crystallographic procedures alone, but in addition it was necessary to examine ;is it cyclic
(c) A n n , 668, 1 f l R 4 7 ) ; (d) F I A T Report No. 1096 (duplicate publications). (e) A Iztte pablication, Borchert, 2. Nafurfovsch, Sb, 4ti-$ (194B), presents X-ray evidence that the r-dextrin is either P tetrasaccharide or an octasaccliaride. ( f ) A tetrasaccharide structare i.i not favored h y Frrudenberi: :rnd Cramer, ihid., ab, 46.1 (1948\, as being sterically unlikely as well a i being out of line with the trend in optical rot-itions cstabli.;he