7...... ~32 5.. $1.12... f. gl‘ilciul‘ as. 11.01.13.133? . .Q; «n.3,. FLQL V {‘00 T1. 1; A I‘d nub/1H... .lLrL WORM E] U r l GARBU BY 0F TlmlESlS FOR DEGREE OF M. 8. ‘1 9 .2 8 1 I a L \ST 4T.._..V_-.sx E: n: agitg :2::__r:__::_::._:= mmm <12 3,3 .0 I “3“ I'm mthflm W m I.“ a"? ; .‘1 u" rial. .\ 1..“ ‘ ,r ‘ . 3 . ’5:- ‘i‘u: . . -'-vw 1'7“” .13)? .:.-'.‘11\I 1‘ "‘ ,‘,;,.~-“‘,; ‘.f'~._:(‘n - ‘\_';. -.,'-.‘L'.\o _,_ bi’sj‘afazrsz‘i.” 4;..- ‘f 9 -1. 4-._ ii A STUDY OF CARBOHYDRATE DERIVATIVES. A STUDY OF CARBDHYDRATE DERIVATIVES A THESIS by Orson D. Bird Submitted to the Faculty of Hichlgan State College In partial fulfillment of the requirements for the degree of Hester of Science June 1928. ACKNOWLEDGEMENT. The writer wishes to express his gratitude to Dr. R. C. Huston for assistance and inspiration receiv- ed during the course of this problem. 331610 TABLE OF COETENTS. Review of literature Experimental part Methylation of glucose Hydrolysis of methylated glucose and determination of methoxal content Anilide of tetramethylglucose Hethylation of sucrose (once) Distillation of methyl sucrose Hydrolysis of methyl sucrose Methylation of sucrose (three times) methylation of starch methylation of glycogen Summary '2) 10 REVIEW OF LITERATURE. The first recorded attempt to prepare alkyl ethers of the sugars was performed by Berthelot, in 1860, (Ann. Chim. Phys.. 60. 103) who heated sugar with caustic potash and ethyl bromide. obtaining a substance which he called diethylglucosan ether. Fischer. in 1893. (Ber.. 26. 2400) devised a practical method for alkylating sugars by the direct action of alcohol on the sugar in the pre- sence of hydrochloric acid. But by this method only the terminal carbinol group of the sugar was etherified and the products. which were of a glucosidic nature. easily lost this alkyl group by hydrolysis. Purdie and Irvine (J. Chem. Soc.. 83. 1021) were the first to prepare fully alkylated methyl ethers of the sugars which retained the aldehydic and ketonic properties of the parent sugars and resisted the action of hydrolyzing agents. This was accomplished in 1903, and really began the study of the methylated derivatives of sugars about which there has since been much interest shown. Purdie and Irvine adopted a mixture of methyl iodide and silver oxide for their methylating reagent. Applying this to glucose directly they got no results since the aldehydic group of the glucose acted as a re- ducing agent and reduced the silver oxide to metallic silver. To avoid oxidation of the aldehydic group of the sugar they then began with a-methylglucoside, pre- pared after the method of Fischer, and applied their .MMH * Luisa stream: .03 cm”: sou «03.31 at unions-us 16 00810th onions this 1.330s 1:91am. one: (61 as mom someone 3 3331115140 . ._.1fl3 Asa-(u Juicer! and. m fimfiis 101 Men! ism , . ~s‘Is SM £14 1339; and ..a lexical w mw‘usid: w Sufi .1: 5“ Ewan no 115313.011: 1 - ”t”. .m:«:: QW’M s ‘10 '- ' ’ m 010‘; meadows! Jam; 01 113112 am new ’3: “wan-toe new. um and r ;"‘ A W Was Ma 10, causes : saw 11.14163: gnaw-01:11:! ‘t i _ as; ‘wamu w.” aim ' wmmdslm ¢u.od),_s1mr m1» Mm.» mam fimfins M'usomib «when “cool: 9d: '10 mu m‘ Wills 99;.) B “W Cd: is n: .a‘ Wes-vs ans ‘1” mu” m dado i1 1 < A T.——‘—. v 2. methylation process to this. They obtained a trimethyl methylglucoside. bailing at 167-1700 (17 mum.) 0n hydrolysis three of the four methyl groups remained, the product being a trimethyl glucose. A further methylation of this substance in the same manner gave tetramethyl methylglucoside. boiling at 144-1450 (17 m.m.3. Dn-hx- drolysis with hydrochloric acid tetramethylglucose was obtained which distilled at lee-123:.o (17 m.m.3. Re- crystallization from ligrcin gave crystals which melted at 81-d50. Sucrose was also methylated by these workers and gave a thick. viscous syrup which on hydrolysis gave a substance identical with the tetra methyl glucose ob- tained from glucose. as shown by its melting point and analysis. This showed. they pointed out, that the con- stitution and linkage of the glucose group in sucrose is the same as that of the simple glucoside. They also oxidized tetramethylglucose with bromine water and ob- tained tetramethyl gluconic acid which had a lactone structure as evidenced by its characteristic reaction toward alkalie. Now lactone formation in the sugar mol- ecule had been assumed by Fischer to take place on the gamma carbon atom so Purdie and Irvine reasoned that their tetramethylglucose formed a lactone. the unmethyl- ated hydroxyl group must be in the gamma position, and therefore that the butylene oxide structure for the glu- cose residue in sucrose was correct. Hudson, in 1910. (J. A. C. 3.. 32, 345) working with lactones derived 3. from sugars. proved beyond a doubt by a theoretical con- sideration of polarimetric data that the butylene oxide structure was correct. thus further substantiating Pur- die and Irvine's views. This mass of evidence caused the following formula to be generally accepted for nor- mal glucose: HOP! Héon HocH O 3 ”$11 HC'ZOH CHmOH Before the study of methylated sugars could go very far some reference compounds had to be prepared by which they could be isolated and recognized. Therefore Irvine and hoodie. in 1908, (J. Chem. Soc.. 93. 93) pre- pared the oxime. anilide and chloro- derivatives of tet- ramethylglucoss and also reached the conclusion that tetramethylglucose entered the same reactions as the par- ent sugar and as the simple gaucoside. In all the first work done on methylation of carbohydrates the methyl iodide. silver oxide method of alkylation was used but in 1915 Haworth (J. Chem. 800.. 107. ll) develOped a much simpler and less expensive process. using dimethylsulphats and sodium hydroxide as methylating agents. This method use applicable to both the reducing and non-reducing sugars directly. In the case of reducing sugars the reaction was carried out at a very low temperature at the start, during which time the glucoside was formed, and later the temperature could 4. be raised and the remaining hydroxyl groups alkylated. Haworth and Leitch. in 1918. (J. Chem. 30s.. 116, 188) used this method of alkylation for an investigation of lactose and melibiose. In 1914 Fischer prepared (Ber., 47, 1980) a new methyl glucoside very much greater in activity than the a and b forms previously known. He called this gamma methyl glucoside and concluded that it must have a different internal linking than either a or b forms. Almost simultaneously. that is in 1915. (J. Chem, Soc., 107. 524) Irvine. Fyfe and H033 reported a similar re- active methyl glucoside which they had discovered while treating glucose with methyl alcohol and hydrogen chlor- ide to form the ordinary glucoside. The compound reacted with acetone while neither a or b forms did so. Also it was very easily hydrolized. They methylated this gamma methyl glucoside with methyl iodide and silver oxide and obtained a tetramethyl gamma methyl glucoside which was purified by distillation. This compound reduced alkaline permanganate in the cold and was hydrolized by 3/100 HCl at 40°. conditions which did not affect the normal al- kylated glucosides. The product of hydrolysis. called tetramethyl gamma glucose. was not crystalline as or- dinary tetramethylglucose. but a liquid, and was laevo- rotatory. Continuing the work on active or gamma sugars, Hudson. in 1916 (J. A. C. 8.. 58, 1223) reported the is- olation of an active form of galactose. Haworth, in P 0. 1925, (J. Chem. 300.. 185, 294) isolated tetramethyl gamma fructose thru the hydrolysis of methyl sucrose and fractional distillation thereof. and attempted to prove an amylene oxide structure for it. He argued that it should have the structure I (below) since on oxidation it gave trimethoxyvalerolactone II, and then trimethoxy- glutaric acid III, which latter was identified as the the anhydride IV, and the dimethyl ester V: ($1.0M: '0 coves o C O'OMG’ l I é-OH C H'OMe CH-OMC CH-OMe é HcOP’le l éu-am o owl-om éH-OMC [two/we A H 'OMe o 'cH-aMe —> [Kw-own: 1" cH-ONIc ""2" Len-om “‘9 eta-0M: I l l I Les-0M: CH1, coo» o c acme I CH3. 1’ 15' 1E? JIT' 3E Work on the methyl derivatLVes of the pentoses was begun by Hirst and Purves in 1923(J. Chem. $00., 126, 1352). They assigned the amylene oxide structure to the normal form of xylose. CHOH f H cor! o 'HocH Héou é”; They proved this rather conclusively since the oxidation of normal trimethyl xylose with nitric acid gave a tri- methoxyglutarie “Cid which would be possible only in the presence of an amylene oxide ring. The normal form of arabinose was methylated by Hirst and Robertson in 1925(J. Chem. 300., 127, 359). They ascribed the amylene oxide structure to it: 6. C HOH c HCOH l O H-OCH I Hc>c+i I CH3" 7 This argument was based on the fact that oxidation of normal trimethylarabinose with nitric acid gave almost quantitative yields of trimethoxy glutaric acid. The trimethyl derivative of gamma arabinose “as prepared by Baker and Haworth in 1925, (J. Chem. 800.. 127. 36s) and gave on oxidation with nitric acid a laevorotary lactone, which on further oxidation gave a dibasic acid with only two methoxyl groups. This. on complete methylation gave a product identical with Hirst's dibasic methoxy acid: CHOH Co Co OH ('Z‘ooH l I . Héofle 'HGOMc Hcmwm H%Oflk ' ——-> Me 04': H -——>- MeO-c': H ——->- MeO-C H WO-C H ' ' I ' CH HocH rMOCH C H ’ t (‘1 éHg'OMC (Hg-0M: (OOH OOH These reactions could be accounted for only by accepting for the active or gamma arabinose the following butylene r—C'EHOH O HCEOH HOCH l I CH I CHzOH oxide structure: Charlton, Haaorth and Peat in 1926. (J. Chem. 800.. 128. 96) summed up all the recent work done on methyl derivatives of the normal and gamma sugars and arrived at a new structural formula for normal glucose, 7. such as is a constituent of sucrose. They observed a parellelism between the change in rotation on hydrolysis of the lactone from normal tetramethylglucose and those from the normal or amylene oxidic forms of galactose. xylose and arabinose. Therefore they ascribed to nor- mal glucose the following amylene oxidic structure: 'HOH HCOH ( HOCH 0 HC OH I He—J CHzOH On the other hand they found that the lactone of tetra- methyl gamma glucose corresponded in change of rotation to the lactones of gamma galactose. gamma xylose and gamma arabinose -- all of which have the butylene oxide structure. Therefore they assigned the butylene oxide structure to active or gamma glucose: EHOH HéoH HoéH Héh-d l HCOH ' CHon The presence of the amylene oxide ring in nor- mal glucose was demonstrated beyond a doubt by Hirst (J. Chem. 300.. 128. 350) who oxidized tetramethylglucose with nitric acid to give xyloltrinethoxyglutaric acid which was identified as the crystalline dianide: .HOH C'OOH (I50 0 H H some He-OMC H C-OMe MEG-CH H) ”‘0'?” --—?- ”90¢." H i... H :0“ l HC H CH C OOH a Ha-OMe (LungNle 2:31426 tdra- icframcflwL ‘Y‘d {rgmethavy- Methqulucose. qlucom'c acid. qhdsrc'c acid. 8. The formation of this trimethoxy acid could only be poss- ible assuming an amylene oxide structure for normal glu- cose. Normal crystalline tetramethylfructose was oxidized by Haworth and Hirst in 1926 (J. Chem. 800., 128. 1858) using nitric acid. The product they ootained was a mixture of the following: ‘ #7806 H H C'OMC I ' HC-OlV/e - M H? 0 e HC’L Oll’le C00” éOOH L- d! meflvoxy- q]- flf‘fléo {fibre/Half- Sacc/n/c (Kid, q/affir-I'c acid. which was possible only upon assuming an amylenc oxide structure for normal fructose. This being the case. the authors considered it imperative that the gamma fructose, which is found in the sucrose molecule. must contain the butylene oxide linking, thereby disproving Hasroth's ear- lier work on gamma fructose when he assigned the amylene oxide ring to gamma fructose. 0n the basis of this work the correct formula for sucrose would be: ('2H‘LOH H O k; ”eta” HOCH ’ q 0 0 HOCJI H$OH I .____J HCOH H? l __éH CHaOH éHzOH A study of the methyl derivatives of the poly- saccharides was begun in 1913 by Denham.and Woodhouse (J. Chem. Soc., 103. 1755) who methylated cellulose by adding 9. methyl sulphate to cellulose suspended in a 15% solution of sodium hydroxide. After three successive treatments in this manner a product was obtained haVLng a methoxal content of 23.11%, which gave trimethylglucose on hydrol- ysis. In a later article (J. Chem. Soc.. 111. 244) they report almost quantitative yields of trimethylglucose on hydrolysis of fully methylated cellulose. thus showing the cellulose molecule to be made up entirely of glucose residues. Glycogen was partially methylated by uhcbeth and Hackay in 1924. (J. Chem. 800.. 125. 513) This was accomplished by repeated treatments With sodium hydrOXLde and methyl sulphate and yielded a product of 37% methoxal content. Further treatments by either methyl sulphate or methyl iodide did not increase the methoxal content. 0n hydrolysis this partially methylated glycogen yielded trimethylglucose identical with that obtained from methy- lated cellulose. Starch was fully methylated by irvine and mac- donald in 1926. (J. Chem. 800.. 128, 1502) after twenty- four treatments with methyl sulphate and alkali. The final product had a methoxal content of 43.7% which agreed closely with that required for a trimethyl starch. This product on hydrolysis with methyl alcohol and hydro- gen chloride gave 2:3:6 trimethyl methylglucoside melt- ing at 57.50. 10. EX?ERIMENTAL. Methylation 93 glucose. The method of Haworth. (J. Chem. Soc.. 107. ll) employing methyl sulphate and a solution of sodium hydrox- ide as methylating agents. was used exclusively. However. the temperature and time of the reaction and the relative proportions of methyl sulphate and sodium hydroxide were varied. -Haworth‘s general method consists in adding three times the theoretical quantity of methyl sulphate. and sodium hydroxide in excess thru dropping funnels to a sol- ution of the sugar in the minimum of water. the solution being stirred constantly by a mechanical stirrer during the operation. The vessel containing the sugar solution is surrounded by a water bath kept at 70° thruout the re- action. which lasts one hour. then the temperature is raised to 100° for thirty minutes to destroy the excess of methyl sulphate. The product. on cooling. is extracted twice with chloroform. the chloroform distilled off and the residue dried. In methylating the reducing sugars the temperature is kept at 50° during the initial stages of the reaction. First methylation. 200 cc. of 50% sodium hydroxide and 100 cc. methyl sulphate were added slowly thru drOpping funnels to 50 grams of glucose dissolved in the minimum of water. the reaction being kept at 35° until all the reagents had been 12. added. then it was raised to 100° for 50 minutes. The re- action mixture was distinctly alkaline at the completion of the reaction and some caromelization had taken place as evidenced by the burnt odor. The product was extracted with chloroform, the chloroform.distilled 6ft and the re- sulting dark brown syrup distilled ‘en vacuo‘. Approx- imately 2 cc. of a straw colored liquid was obtained which boiled at 144—1460 (17 m...), which proved it to be tetra- methyl methylglucoside. Second methylation. 75 grams of sodium.hydroxide in 250 cc. of water and 175 cc..methyl sulphate were added in the same manner as before to 50 grams of glucose in water solution. the temperature being maintained at 55° until all had been add- ed then raised to 100° for half an hour. The reaction mixture remained neutral to litmus until after the final heating when it became just slightly acid. There was no caromelization and the resulting syrup after extracting with chloroform was a light golden color. After distill- ing off the chloroform.54.5 grams of the syrup remained. This was distilled at 24 mam. and came over at 150-1800, there being no definite boiling point. Eldrolysis 2; methylated glucose and determination 2; methoxal content. The methylated methylglucoside obtained in the second methylation was hydrolized after the method used 15. by Purdie and Irvine (J. Chem. Soc., 85, 1021). 3.5 grams of the material was dissolved in 50 cc. of 4% aqueous HCl. to which was added a little charcoal, and this was boiled under the reflux for one hour. The mix- ture was neutralized with barium carbonate and evaporat- ed to dryness on the water bath. The residue was ex- tracted with ether and the syrup resulting, after evap- orating the ether. distilled at 160-1900 (17 m.m.). no definite boiling point being evident. Three grams of a straw colored. viscous syrup was obtained. Determination of the methoxal content of the above syrup was made according to Perkin's modification of the original Zeisel method (J. Chem. Soc.. 83. 1367). The hydriodic acid us- ed was obtained from.Merck 5 Co.. of a sp. gr. 1.72. The alcoholic solution of silver nitrate was prepared accord- ing to the method of Pregl. (Die Quantitative Organische Microanalyse. p. 150) by boiling a solution of 20 g. sil- ver nitrate in 500 cc. 95% alcohol under a reflux conden- ser for three hours and filtering off the separated sil- ver oxide. Results of two determinations were as follows: 1. .0726 gave .250? A51 ; 4l.9£ OCH; The calculated OCH3 content of trimethylglucose, C5H903- (OCH3)3 is 41.89%. which corresponds closely with the above and identifies it as such. 14. ' Third methylation. 30 grams of the material from the second methylation above was remetnylated using 37 grams sod- ium hydroxide in 225 cc. mater and 87 cc. methyl sulphate. The temperature was maintained at 50-55° thruout the methylation and then raised to 103° for half an hour as before. The resulting product. after having been ex- tracted with chloroform. dried and distilled, gave 27 g. of a water clear syrup boiling at 144° (17 m.m.). Keth- oxal determination gave the following results: .1024 gave .4529 Ag: . 58.53% cons The calculated OCHs content for tetramethyl methyl- glucoside is 62%. 27.5 g. of this material was hydrol- ized by heating with 200 cc. 4% HCl for one hour under a reflux. neutralized with barium carbonate. and evaporat- ed to dryness on the water bath. After extracting with ether and removing the solvent by evaporation, a syrup was obtained which distilled at 180-1850 (18 m.m.) to give 15.7 grams of a lnght straw colored, viscous mater- ial. which crystallized out almost completely on stand- ing over night. Recrystallization from.ligroin gave 7.85 grams of needle like crystals melting at 79-80°. This identifies the material as tetramethylglucose. Preparation 3; the anilide g: tetramethylfilucose. The anilide of tetramethylglucose was prepar- 15. ed according to the method of Wolfrom and Lewis (J. A. C. 8.. 50, 857). 15 cc. of absolute alcohol and 15 cc. redistilled aniline were added to 4.5 g. of the tetra- methylglucose crystals and boiled gently under a reflux for two hours. On standing over night most of the water- ial had crystallized out. The crystals were filtered out. washed with a little aniline and the filtrate evaporated “en vacuo" when more crystals separated. The combined crystals were recrystallized from absolute alcohol. 2.8 grams of crystals were obtained which melted at 124-1250. Methylation g£.sucrosc. First methylation. The.method of MoOwan (J. Chem. Soc.. 128, 1737) was used for the methylation of sucrose. This is a modification of Haworth's general method emplqying more dilute solutions. The method is described as fol- lows: “165 grams sodium.hydroxide dissolved in 550 cc. water and 183 cc. methyl sulphate are added simultan- eously to 61 grams sucrose dissolved in the minimum of water. The reaction is kept at 35° during the addition of the reagents then raised to 60° for half an hour, 75° for half an hour. and finally boiled for half an hour. After cooling. the mixture is extracted with chloroform and the chloroform.distilled off". Thirty grams of a very viscous. straw colored syrup was obtained after one methylation by the above method. 16. Distillation of methylated sucrose. As the methylated sucrose could not be dis- tilled without decomposition in the ordinary vaccum dis- tilling apparatus used for most organic compounds. a spe- cial. all-glass apparatus was constructed from Pyrex glass. All the joints were sealed except one where the material was admitted and a thermometer was introduced. This was closed by an inverted test tube. ground to fit the neck of the flask in which it was set. and contain- ing a suspended thermometer which extended downward into the distilling flask so as to record the temperature at which the material distilled. The ground glass Joint was made more air tight by a mercury seal. A McLeod gauge was incorporated in the apparatus, which made it possible to measure pressures below 3 m.m. accurately to 0.1 m.m. The methylated sucrose described above was distilled in this apparatus and gave 15 grams of syrup distilling at 240-2500 (1.6 m.m.). Methoxal determin- ation gave the following results: 1. .1087 gave .358? AgI a 41.15% OCH3 2. .1063 gave .3517 Agi . 41.18; sea: This indicated that the methylated sucrose was methylat- ed beyond the penta-methyl stage, which required 37.6% OCH3, but not quite to the hexa-methyl stage which call- ed for 43.6! OCH3. 17. Hydrolysis of partially methylated sucrose. Twelve grams of the partially methylated suc- rose was hydrolized according to the method of Haworth, (J. Chem. 800., 11?. 139) by dissolving in 110 cc. .43 HCl and heating in a water bath at 60° for six hours. Then it was neutralized with BaCOg. evaporated to dry- ness, extracted with chlorofornand dried. Right and five tenths grams of syrup resulted which was distilled in three fractions as follows: 1. 160-1700 (22 u.r.) .97 g. 41.31; OCHz. 2. 17o-1so° (19 s.s.) 5.65 g. 42.59; . s. 180-1900 (16 u.s.) 1.08 g. 42.57: . These figures indicate the correct 0H03 content for a trimethyl hexose. which proves tint the original once methylated sucrose was hexamethyl sucrose. The large fraction of 3.65 g. was evidently largely tri- methylglucose as its boiling point corresponds closely to that recorded for trimethylglucose, namely 180-1850 (20 m.m.) Second methylation g£_sucrgse. Fully methylated, octamethyl sucrose aas pre- pared by three successive methylations of the sugar, us- ing the same method as outlined above except that in the second methylation the sodium hydroxide was dissolved in 750 cc. of water and in the third methylation in 1000 cc. 18. cc. of water. After the three methylations 41 grams of syrup was obtained, 61 grams of sucrose having been used. 0n distillation, 18.3 grams of a straw colored, viscous syrup boiling at 1770 (1.5 m.m.) was obtained. Methoxal determinations were as follows: 1. .1258 gave .5036 A31 3 52.54% OCH; 2. .1236 gave .4911 Agi . 53.c3% OCH3 The theoretical OCH; content for octamethyl sucrose is 54.6%. The success of this complete methylation of sucrose. using only sodium.hydroxide and methyl sulphate as methylating agents. evidently depended upon increas- ing the dilution of the solution for each succeeding methylation. This was accounted for by the fact that highly methylated sucrose was only slightly soluble in a solution of sodium.hydrox1de but if a large enough vol- ume of the alkalie solution were present. all would be dissolved and consequently acted upon by the methyl sul- phate. This was borne out by Haworth. in 1916. (J. Chem. Soc.. 117. 199) who reported that octamethyl sucrose could not be prepared by using sodium hydroxide and meth- yl sulphate alone. due to the insolubility of methylated sucrose in sodium hydroxide solution. Later(1920, J. Chem. Soc., 125, 509) he did prepare octamethyl sucrose with these reagents used in larger excess and in great- er dilutions. l3. Hethylatibn g£_starch. Starch was partially methylated by a method reported by Irvine and hacdonald, in 1326 (J. Chen. 806.. 128, 1302). The method was recorded by them as follows: "Thirty two grams starch were mixed with 150 cc. water and stirred slowly while 160 cc. of 12.5% sod- ium.hydroxide was added. Then 140 cc. of 50% sodium.hy- droxide and 80 cc. methyl sulphate were run in simultan-- eously, taking place over a period of three hours and adjusted so as to maintain the alkaline reaction. The liquid was kept at 35° and vigorously stirred during the reaction. After the addition of the above it was heated at 100° for 40 minutes. After cooling, an equal volume of rectified spirit was added, and carbon dioxide gas passed thru the solution for a prolonged period. The precipitated sodium salts mere removed by filtration thru linen, and the filtrate caustiously neutralized with dil- ute sulphuric acid. The liquid was evaporated to a thick syrup and the metholation repeated on this". These workers obtained fully methylated starch (0CH3 content: 43.7%) after twenty foun'methylations. They encountered three distinct stages during the methy— lations. which they thought almost saranted the existance of three compounds. ratio of OCHz %OCH3 to initial OH I. Dimethyl starch 32.? 6:9 II. Hethylated starch 36.3 7:9 III. Trimethyl starch 43.7 319 2‘). The methylation of starch was carried out according to the above procedure, using the same amounts of material and reagents. but a simpler method of sep- aration of the product was devised. Instead of precip- itating and filtering off the sodium salts. then evap- orating the filtrate which contained the partially meth- ylated starch, it was found that by neutralizing the sol- ution to litmus directly after the methylation process the methylated starch became insoluble and separated out as a curdy mass, leaving the salts in the solution which could be decanted off. This curdy mass was then emulsi- fied in 12.5% sodiun.hydroxide and methylation repeated on this. using the same amounts of reagents as before. After eight methylations in this manner a product was ob- tained which when dried was very brittle but on standing soon took up some water and became plastic. This was purified by dissolving in chloroform and filtering out .the insoluble material. After evaporating off the chlor- oform the residue was dissolved in ether. which upon evap- oration left a flaky white powder. Methoxal determinat- ions gave the following results: 1. .1064 gave .2974 A51 - 36.83% OCHj 2. .1164 gave .3193 A31 - 36.21% OCH3 These results indicate that the methylated starch obtained corresponds to fraction II. obtained by Irvine and Macdonald. which had an average meth0xa1 con- tent of 36.51. 21. fiethylation g£_glycogen. Glycogen was partially methylated by the met- hod of Irvine and macdonald (J. Chem. 800., 128, 1502) as applied to starch. The same changes were made in the method of separation of the product as in the case of metholated starch. Starting with 52 grams of the material. there resulted. after six methylations and purification of the material with chloroform and ether. 15 grams of a gray- ish powder which very much resenbled the original sub- stance. Methoxal determinations gave these results: 1. .1406 gave .3616 A31 35.94% eons 54.44% case 2. .l240 gave .3236 AgI During the last two methylations the material became so insoluble in sodium.hydroxide that is doubt- ful if any further methylation was taking place. l. Glucose was methylated to the tetramethyl stage in one methylation. using a distinctly alkaline solution and keeping the temperature down to 35°. but consider- able caromelization took place and the yield was very small. A good yield of tetramethyl glucose was ob- tained by two methylations. keeping the reaction just slightly alkaline and the temperature at 35° during the first methylation. and raising the temperature to 50- 600 and increasing the alkalinity during the second methylation. 2. Hexamethyl sucrose was prepared by one methylation with sodium hydroxide and methyl sulphate. The product was successfully distilled at a pressure of 1.6 m.m. and came over at 240-2500 with no appreciable decom- position. This product was hydrolized to a trimethyl hexose. 3. Sucrose was fully methylated after three methylat- ions with sodium hydroxide and methyl sulphate. by in- creasing the dilution of the sodium hydroxide solution for each successive methylation. This distilled smooth- ly at 177° under 1.5 m.m. pressure. 4. Starch and glycogen were partially methylated, using the same method for both. The starch had a methoxal con- tent of 36.5% after eitht methylations. By the same treatment glycogen had a 34.2% OCH; content after six methylations. RS ""iiiLflfirijMi[iiiujfflluifliflfflilm‘r‘fe’