1 Q . 1 flu".- ‘ I-.¢‘ I-c ‘ .‘ ., , . unqu that! G ,‘o w 31“? \‘38v. fa.“ \. x . ‘K 1"”.1‘. : Cum p-q i This is to certify that the thesis entitled The Prepa a+1on oi Ethyl 9-Broxxononanoate, and the 83nthesis oi lO,l9-Dioxo-octacosanediojC Acid, from Sebacic Acid presented by Kelly F arhat has been accepted towards fulfillment of the requirements for :2. o S o degree in Chemi S tr‘ .4» F" v . Major professor ’ r' I , ‘. Date October 22 , 19‘5" L] L ‘. it" 0-159 L‘"-"‘S‘ . (4:11 Jr 'o ‘Twrj'J “J T: 'r‘vx..\ t- ‘s ‘\—' ,7 -‘ H *TTI‘V'V.‘ 4‘ ”.3“; Ei :11?ng SWEE; ' "‘1‘: V \x: ‘3‘, t V_ gfit’j‘jjj it?" , 3‘ \‘ ‘ _ «(f-“L :lA“ L»; .4 , 'T‘ Fligtwf ‘ k. "‘ :21»! ‘,;-‘\' ‘ iii/15% ‘- 4"!” - :4. ‘6‘“ I, ’P Y; ‘3 1 1 ”w uk-TA’} ". [if f ‘.V.-- A17 ‘ A." la. 1; ‘1‘“? Y7“; 1'L_ ,r' 4 \’}"_ J 31’ ‘“ '\_, ‘ Fr; H». ‘ /‘\j" .‘g V‘ ‘(1 ' ',~ “1‘7“:’M ‘I '1 "‘A’t'fl‘i. l \ ‘l~ \_1 1s ‘ \ '7: "I -4 51-“ _. Jx‘ ‘*‘ ' if f',-‘;+} .L,‘ - " - 3' ‘ {31). {var -' . «a» M r‘w . ‘. - ‘ < w 2." F; v" ‘ — v " :36“ {1‘4" ‘55-. 4“ .1“ . "i4" ,3 ~ }\ I _ (k: “ 3-3.: -“'\~= ‘é ...- , '1 1.‘ L}, 5,. "._(‘ It I x . 5 ~ r‘ j v{ l/N‘ \ #1 r, 4}" A“ L361, v‘." \‘g 1‘;V J \3 , M . ‘ ' 5 ‘ ,(‘J’ ii'y‘t . 1-..." ‘. x l: t 3" 7;; x . 3".“ ; 7, J‘ ’ 1} «9,1,5 t ‘ , 11 :7 if .1 7.“. ”‘1‘ v}: ’ ”11.7 , .1 w s 7’ "i x ’ ‘ 1. 1". g VP)»; 4 .‘ “t .‘ g 37" \x 7"; ‘N’Lfis .‘ . THE PP; ARATION OF ETHYL 9-BROMONONANOATE AND 10,19-DIOXO-l’28-OCTACOSANDIOIC ACID FROM SL‘BAC IC AC ID By Kelly Farhat AN ABSHUET Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIEPCE Department of Chemistry Year 1952 w I I l " ”— Approved / Tyeie Fans \kg Kelly-Farhat THESIS ABSTRACT This thesis describes the preparation of ethyl 9-bromononanoate and 10,19-dioxo-l,28-octacosandioic acid. To prepare ethyl 9-bromononanoate, sebacic acid was directly half- esterified with ethyl alcohol. The ethyl hydrogen sebacate which resulted was converted into its potassium salt and then ethyl silver sebacate was precipitated by treatment with silver nitrate. Bromination of ethyl silver sebacate under very dry conditions yielded ethyl 9-bromononanoate. 9-Bromononanoic acid, 9-bromononoyl amide, an S-alkylisothioureapicrate derivatives of ethyl 9-bromononanoate, and the S-alkylisothioureapicrate of 9-bromononanoic acid were prepared as derivatives. To prepare lO,l9-dioxoel,28-octocosandioic acid silver sebacate was prepared from sebacic acid. The dry silver salt was then brominated to produce 1,8-dibromooctane. Reacting 1,8 dibromooctane and magnesium in dry ether yielded a Grignard reagent, octamethylene dimagnesium bromide. Treatment of this Grignard reagent with a dry ether solution of zinc chloride produced the corresponding zinc Grignard, octamethylene dizinc chloride. -Carbethoxynonanoyl chloride which was to be condensed with the zinc Grignard reagent was prepared from sebacic acid by half esterifica- tion of the acid followed by treatment of the half-ester with thionyl chloride. «\r\¢1(3(r—~ « e.) -1- Kelly Farhat Condensation of octamethylene dizinc chloride with -carbethoxy- nonanoyl chloride then yielded diethyl lO,l9-dioxo-l,28-octacosandioate. Saponification of this compound with alcoholic potassium hydroxide followed by acidification produced the corresponding acid, 10,19-dioxo-l,28- octacosandioic acid. A 2,b-dnintrOphenylhydrazone derivative was prepared from diethyl- lO,l9-dioxo-l,28-octacosandioate. -2- TEE PKLPARATION OF ETHYL 9-BROMONONANOATE AND lO,l9-DIOXO-l,ZB-OCTACOSANDIOIC ACID FROM SEBACIC ACID Kelly Farhat A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1952 ACKNONLEDGMLNT The author wishes to express his appreci- ation to Professor Bruce E. Hartsuch whose able assistance and guidance made this thesis possible. ****%%K%** *eeeeeae eeeree *eee a% p .K TABLE OF CONTENTS PAGE INTRODUCTION..... .................................................. l DISCUSSICN ......................................................... 3 EXPE QIMLNTAL ....................................................... 12 The Preparation of Diethyl Sebacate. ......... .. ........ ....... 12 The Preparation of Ethyl Hydrogen Sebacate. ................. .. 13 The Preparation of Ethyl Silver Sebacate ..................... . 15 The Preparation of ethyl 9-Bromononanoate... .................. 16 The Preparation of Silver Sebacate ................ . ........... 17 The Preparation ofl ,S-Dibromooctane ............... ... ........ 18 The Preparation of 9-Carbethoxynonanoyl Chloride .............. 19 The Preparation of tle Grignard Peagent of l ,8—Dibromooctane.. 20 The Preparation of 1,8-Oetamethylene Dizinc Chloride .......... 21 The Preparation of Diethyl lO,l9-Dioxo-l,25-Octacosandioate... 21 The Preparation of lO,l9-Dioxo-l,ZS—Octacosandioic Acid ....... 22 Preparation of Derivatives... ................................. 23 9-Brcmononanoic Acid . .................................... 2h The Amide of 9-Bromononanoic Acid. ....................... 2h The SeAlkylisothioureapicrate Derivative of Ethyl 9—Br0mononanoate ......................... . . . . . . . . 25 The S—Alkylisothioureapicrate Derivative of 9-Bromo- nonanoic Acid ........................................ 25 The 2,b-DinitrOphenylhydrazone Derivative of Diethyl lO,19-Dioxo-l,28-Octacosandioate ...... ... ......... .. 26 SUMMARY ............................................................ 27 RJFEM ENCES . . . . ..................................................... 29 INTRODUCTION INTRODUCTION The work reported in this thesis was proposed and carried out as part of a general program leading to the preparation of some long chain cellulose ethers. lihen cellulose is treated with fairly concentrated sodium hydroxide (8-20%), a sodium cellulosate is formed, in which a sodium atom has re- placed an alcohol hydrogen. lihen this sodium cellulosate is digested with alkyl halides or with substituted alkyl halides, a cellulose ether is formed, in which the organic part of the halogen compound is attached through an oxygen atom to the cellulose chain. Using this general method many cellulose ethers have been prepared having normal, secondary and tertiary side chains containing from one to seven carbon atoms. The general plan for the study of cellulose ethers in this laboratory involves the preparation of ethers with unbranched side chains containing from nine to twenty carbon atoms. The purpose of the work done in the present problem was to synthesize some of the non-cellulosic compounds which may be used for the preparation of such ethers. After considering various long chain halogen compounds, it was de- cided to prepare an omega bromo ester rather than an alkyl halide or an omega bromo acid. The reason for this decision was based on the probable properties of the cellulose ether which might be made from these halogen compounds. Two such prOperties were considered. First, if a cellulose ether containing a long side chain with an oxygen containing end-group were to be dispersed and then extruded, there would be the possibility of hydrOgen bonding between the side chain end-groups and the main cellulose chain. Second, an ester end-group seemed to be preferred, rather than a carboxyl end-group (the omega bromo acids), because the ester end-group would be more stable to the action of cold alkali. Because the method chosen for preparing omega bromo esters involved the use of dicarboxylic acids as starting materials, a method of preparing such long chain dicarboxylic acids was also investigated. DISCUSSION DISCUSSION The selection of all preparative methods reported in this thesis was based on the applicability of the reaction for preparing all the members of the gomologous series with nine or more carbon atoms, the availability of the starting materials, the number of necessary steps and the yield at each step, the cost of materials and the ease of carry- ing out the reaction. The preparation of a long chain omega bromo aliphatic acid by treat- ing the corresponding alcohol with hydrogen bromide has been carried out by Chuit (1). Only a few omega hydroxyaliphatic acids having nine or more carbons are readily available (1,2,3); therefore, this method was not investigated as a possible general preparative means for obtaining the long chain omega bromo esters. Gaubert, Linstead and Rydon (h) have investigated the preparation of omega bromo acids and their esters by the addition of hydrOgen bromide to unsaturated aliphatic acids or esters having double bonds between the last two carbon atoms under various conditions. The presence of a peroxide catalyst, e.g. benzoyl peroxide, caused the hydrogen bromide to add in reverse to the Markonikow rule which would normally place the bromine atom on the carbon atom with the least number of hydrogen atoms attached. For example, the addition of hydrogen bromide to 5 nonenoic acid would normally yield 8-bromononanoic acid, but in the presence of peroxides in the reaction mixture 9-bromononanoic acid is obtained. Several workers (5,5) have treated undecylenic acid with hydrOgen bromide to prepare ll-bromoundecanoic acid. General use of this method for pre- paring the omega bromo esters or acids with more than eight carbon atoms in chain length is hindered because the starting materials are not readily available, except in the specific case of undecylenic acid which is com- mercially available. lfhen Borodine (7) observed that bromination of dry silver acetate yielded a heavy bromine containing gas (methyl bromide), he opened the way to a general method of preparing organic bromides. Bochmueller and Hoffman (8) extended the reaction between dry silver salts and bromine to a series of the lower aliphatic silver salts such as silver butyrate and valerate which yielded propyl bromide and butyl bromide, respectively. Patents issued to Hunsdiecker, Hunsdiecker and Vogt (9) describe the preparation of omega bromo esters from such compounds as ethyl silver sebacate to yield ethyl 9-bromononanoate. Hunsdiecker and Hunsdiecker (10) have reported the preparation of omega bromo methyl esters having five to seventeen carbon atoms from the silver salts of the half methyl esters of dicarboxylic acids. A review by Kleinberg (ll) covers all the literature up to 19h? on halogenation of dry silver salts of carboxylic acids. The method of Hunsdiecker and Hunsdiecker (10) was used in the present investigation to prepare ethyl 9-bromononanoate. This method starts with sebacic acid, which is commercially available, and converts it to its monoester. The monoester is changed to the silver salt, and this silver salt is then brominated to yield ethyl 9—bromononanoate. The monoesters of dicarboxylic acids have been prepared by three methods. Fourneau and Sabetay (12) heated equimolar quantities of the dicarboxylic acid and its diethyl ester for several hours. The length of time necessary to carry out the reaction and the generally lower yields have kept this method out of general use. Half saponification of the diethyl ester of a dicarboxylic acid was used by‘falker (13) to Get the monoethyl esters. This is a two step method requiring the prior 0 1 preparation of the diethyl ester, and it has been shown that the yields vary considerably; A variation of the half saponification procedure introduced by Hunsdiecker and Hunsdiecker (10) uses barium hydroxide to half saponify the diethyl ester. An insoluble barium salt which precipi- tates is treated with a strong acid to free the organic half-ester. Although this method was reported to give good yields, it was used by Hunsdiecker and Hunsdiecker only when a more simple method developed by Swann, Oehler and Buswell (1h) failed with the higher homolcges of the dicarboxilic acid series (fifteen carbon atoms or more). The method of preparing half—esters developed by Swann, Oehler and Buswell directly esterifies one of the carboxyl groups of a dicarboxilic acid. Hunsdiecker and Hunsdiecker (10) used the method as found in "Organic Synthesis" (1h) and Jones (6) used a modification of this method for isolating the product. The modification of this method reported by Jones is of general applic— ability for the preparation of some of the half-esters of dicarboxilic acids containing fifteen or more carbon atoms which can not be easily separated by the distillation required by the original method. In this method ethyl alcohol, the dicarboxilic acid, the full ester, a little n-butyl ether and a concentrated hydrochloric acid catalyst are refluxed together. The full ester is introduced to prevent the formation of more diethyl ester from the dicarboxilic acid during the reaction. Since the method of direct esterification has come into general use and gives good and constant yields, it was used to prepare the ethyl hydrogen sebacate necessary for this research. .The method of Swann as given in "Organic Syntheses,‘ the modification of this method which Jones used and another further modification were used. This last modification is described in detail together with the results obtained from the other two variations on page 1h. The preparation of ethyl hydrogen sebacate by the direct esterifica- tion method necessitated the preparation of diethyl sebacate. A method of esterification reported by Micovic (15) which uses one and one-half times the theoretical quantity of ethyl alcohol, a catalyst of concen- trated sulfuric acid and toluene to form an azeotrOpic mixture of alcohol, toluene and water was used to make the diethyl sebacate. The experimental details are on page 13. Two methods which are frequently used for the preparation of the silver salts of carboxilic acids are, first, the conversion of an alkali metal salt of the carboxylic acid to a silver salt by treating the alkali salt with freshly precipitated silver oxide, and, second, the precipita— tion of the silver salt from a solution of the alkali salt with a solu- tion of silver nitrate. Because the second method required no digestion and because the precipitate may be purer, it was used to make ethyl silver sebacate. The experimental details are on page 16. The ethyl silver sebacate was brominated according to the method of Hunsdiecker and Hunsdiecker (10) except that three minor variations were used. First, about twice the quantity of the reaction medium (carbon tetrachloride) was necessary to permit efficient stirring; second, the bromine was dried by shaking it with concentrated sulfuric acid, a method found in "Organic Syntheses" (16); third, the bromine was in carbon tetrachloride solution when added to the suspended silver salt. The experimental details for this reaction are found on page 17. The complete list of reactions necessary to make ethyl 9-bromononanoate is shown below: 202HgOH + Hooc(cwz)scoow acaeboodcwgacooc Zea + 2H20 CQHEOH + Hooc(cwz)ecoow aczxsoocmuzncoow + H20 CerigOOC(CH2)BCOOH + KOH ——-> Canbcocmwgacoox + H20 czwsoocmwpacoo‘K“ + AgW0‘—> C2H500C(C}12)8C00Ag + K+Nog czasoocwwzncomg + Bra—e 02H5000(0H2)53r + AgBr + co2 To obtain some derivatives of the carbethoxy end of ethyl 9-bromo- nonanoate the ester was converted to 9-bromononanoic acid by a ester exchange reaction used by Hunsdiecker and Hunsdiecker (10) to get the acid from methyl 9-bromononanoate. This involves the use of excess glacial acetic acid containing three per cent hydrogen bromide. The reaction follows: 02H50m(CH2)BBr + CH3COCH —-er HOOC(CH2)e,Br + ceacoocgh6 A derivative (17) of 9-bromononanoic acid was obtained when the acid was heated with thionyl chloride to make 9—bromononanoyl chloride. The acid chloride was cooled in an ice bath and mixed with cold twenty eight per cent ammonium hydroxide solution to yield 9—bromononanamide. Br(CH2)8COOH + 30012 ’9 Br(CH2)8C03L + so2 + HCl Br(CH2)aCOCl + zlma —_> Br(CH;3 Boom;2 + NH401 Derivatives of the alkyl bromide end of the ethyl 9-bromononanoate and 9-bromononanoic acid molecules were sought. Finally, the S-alkyl- isothiourea picrates of these compounds were made according to the general method found in Cheronis and Entrikin (18). The experimental details for the preparation of 9-bromononanoic acid and 9-bromononanamide are on page 2b and those for the S-alkyl—iso- thioureapicrates of 9-bromononanoic acid and ethyl 9—bromononanoate are on page 25. A second objective in this study was to prepare a long chain a, - dicarboxilic acid. The older and more conventional methods of syntheses (5,6,19,20) will increase the length of a carbon chain one, two or four carbon atoms at a time. A method without this limitation was sought. Although the Kolbe-Brown (21) electrosynthesis is possible, an original method was preferred in order to investigate other means of synthesizing these compounds, and to add to the literature. It was for this reason that the method used by Blaise and Koehler (22) for making keto acids from an alkyl zinc iodide and omega carbethoxyalkoyl chlorides was extended to a -polymethylene dizinc chlorides. The reaction of Blaise and 3 Koehler is :7an + Cloch‘hg)acoocgw6 «a. P£O(CH2)BCOOC2H5 + Zn(Cl)(l) By hydrolysing the keto esters they obtained the keto acids. Others who have used this reaction for making keto acids are Ruzicka and Stoll (23), Spath and Darling (2h), Spielman (25), and Schnieder and Spielman (2o). Schuette, Roth and Christensen (27) made 10—ketotricontanoic acid from docosyl zinc chloride and 9-carbethoxynonanoyl chloride. Jones (6) prepared a series of long chain keto acids from octadecyl zinc chloride and various -carbethoxyalkoyl chlorides. By extending this reaction to a polymethylene dizinc chloride it was possible to prepare diethyl lO,19-dioxo-l,28-octacosandioate from 1,8-octamethylene dizinc chloride and -carbethoxynonoyl chloride. 20 2H500C(CH2) 80001 + ClZn(CH2)BZnCl a 02H5000(CH2) 8come 2)eco(c,Hg)scooc 2H5 + 27.le Saponification with a strong base yielded the corresponding 10,19-dioxo- 1,28-octacosandioic acid. The Conditions which were used for preparing diethyl lO,l9-dioxo—l,28— octacosandioate were essentially those of Spielman (25); however, the method of isolation differed. The experimental details for preparing this compound and the corresponding 10,l9—dioxo-l,28—octacosandioic acid can be found on pages 21 and 22. In order to extend the reaction of Blaise and Koehler to diketo di- carboxylic acids a source of a polymethylene dizine chloride was needed. 10 This was found in a series of reactions starting with sebacic acid. Sebacic acid was converted into its potassium salt and disilver sebacate was precipitated by the addition of a solution of silver nitrate, The disilver sebacate was brominated in dry carbon tetrachloride in a manner quite similar to that used by Lutteringhaus and Schade (28) for making 1,8-dibromooctane. In order to get 1,8-octamethylene dizinc chloride, 1,8-dibromo- octane must be converted to a dibasic Grignard reagent and this in turn reacted with anhydrous zinc chloride dissolved in anhydrous ether. Di- basic Grignard reagents, similar to that produced from l,8-dibromooctane and magnesium, have been previously reported by Grignard and Vignon (29), Braun and fipbecki (30), Chuit (l2), and Lespieau (31), but that of l,8-dibromooctane has not been previously reported. Coupling has been shown to occur in the preparation of dibasic Grignard reagents (29,30,31). However, Chuit (l9), and Lespieau (31), have used dibasic Grignard re— agents for preparing bifunctional compounds. In the laboratory the preparation of the Grignard reagent from 1,8-dibromooctane was carried out in good yield by adding an ether solu- tion of 1,8-dibromooctane to an equivalent of magnesium under anhydrous conditions. This Grignard reagent was treated with a dry ether solution of zinc chloride, thus converting it to 1,8-octamethylene dizinc chloride. Omega carbethoxynonanoyl chloride has been prepared from ethyl hydro- gen sebacate and thionyl chloride under various conditions, all requiring the use of heat (23,27). In the present investigation this compound was 11 prepared without the use of heat by mixing the reactants and allowing them to stand overnight before distillation of the product. The experimental details for the preparation of silver sebacate, 1,8-dibromooctane, 9—carbethoxynonanoyl chloride, the Grignard reagent from 1,8-dibromooctane, and 1,8—octamethylene dizinc chloride are found on pages 17, 18, 19, 20 and 21 respectively. A complete summary of the reactions for making lO,19-dioxo-l,28— octacosandioic acid is shown below: 1. HOOC(CHz)BCOOH + 2KOH ~41 Kocc(CHZ)8c00K + 21120 2. KOOC(CHg) 8000K + 2AgN03 ~‘)Ag{)OC(CH2) 8000Ag + 2Kl‘u103 3, P.gO(X3(CH3)8COOAg + 213:;2 ~—}Br(CHz)eBr + 2002 + 2;:gar h. Br(CHz)e,Br + ZMg '-5}3n~1g(CH2) aMgBr 5. Bng(CH2)81‘v’EgBr + 22n012 -—) Cizn(CH2)82re1 + MgBrg + M3012 6-. cszoocmHgacooe + 30012 ~> 02H5000(CH2) 80001 + HCl + so2 7. 2C2H5000(CH2)BCOCI + ClZn(CH2)BZnCl ——§ 02H5000(CH2) 800(CH 23) 800(CH 2) Epooc 2H5 (1) 2 NaOI—I (2) H01 8. 02H5000(CH2)BCO(CH2)8CO(CHB)800002H5 ~> HOOC (CH 2) 880(CH 2) 8cc (CH 3) 8coon A 2,h-dinitrOphenylhydrazone derivative of diethyl-10,19—dioxo- 1,28-octacosandioate was made following.a general method given by Cheronis and Entrikin (32) (See page 26). l2 EXPERIMENTAL The details of the laboratory procedures used in the synthesis of all compounds prepared in this investigation are found on the following pages. The Preparation of Diethyl Sebacate 303 gm (1.5 mole) sebacic acid ShO m1 (h.5 mole) absolute alcohol 270 m1 toluene 1.3 m1 concentrated sulfuric acid 225 gm anhydrous potassium carbonate The sebacic acid, absolute alcohol, toluene and sulfuric acid were placed in a 2 liter flask which was connected to a downward condenser leading into a receiver that contained the anhydrous potassium carbonate. The flask was heated and at about 760 C. the azeotrOpe of alcohol, water and toluene began to distill. The distillation was allowed to proceed slowly and finally stepped at 79° c. The distillate was shaken with the anhydrous potassium carbonate, and after allowing the salt to settle, was decanted and returned to the reaction flask. Distillation was then resumed and again discontinued at 790 C. The residue was then transferred to a 1-liter flask. The reaction flask was rinsed with alcohol and the rinsings added to the contents of the l-liter flask. The solvents were distilled from the above reaction mixture at . . o atmospheric pressure to approximately 130 C. The removal of solvents was then continued at reduced pressure in order to prevent overheating the main product. Diethyl sebacate was collected at léh-ldéo C./3 mm. A small afterrun was discarded. The yield was 330 gm or 92.1% of the theoretical yield of 358 gm based on the sebacic acid used. The Preparation of Ethyl Hydrogen Sebacate 101 gm (0.5 mole) sebacic acid 75 gm (0.29 mole) diethyl sebacate 25 ml n-butyl ether 15 gm (12.5 ml) concentrated hydrochloric acid to ml 95% ethyl alcohol ° The sebacic acid, diethyl sebacate, n—butyl ether and concentrated hydrochloric acid were put into a round bottom 500 m1 flask. The mixture was boiled one—half hour under a reflux condenser. Thirty milliliters of 95% ethyl alcohol was then added through the condenser. After this mixture was refluxed two hours, 10 ml of 95% ethyl alcohol was added and the refluxing continued for an additional two hours. The reaction mixture was then transferred to a 2-1iter separatory funnel. Four hundred milliliters of ethyl ether was added, well mixed and the water layer separated and discarded. The ether solution was extracted once with 1 liter and once with 0.5 liter of O.hM sodium carbonate. This converted the ethyl hydrogen sebacate and sebacic acid into their soluble sodium salts, leaving the diethyl sebacate and n-butyl ether in the ether layer. This layer was washed with water and the wash- ings added to the main water layer. The ether layer was distilled at atmospheric pressure removing the ethyl ether and n-butyl ether, and then the diethyl sebacate was distilled off in vacuo at 161-1650 c. at 6 mm. 1h The combined aqueous extracts containing the sodium salts of sebacic acid and ethyl hydrogen sebacate were acidified with concentrated hydro- chloric acid until acid to congo red paper. This precipitated the acidic compounds in a pasty, semi—solid mass. Four hundred milliliters of ether was added, the mixture shaken and the two layers were separated. The aqueous layer was washed twice with 200—ml portions of ether and the washings added to the main ether layer. The ether solution was then evaporated, leaving a residue which contained crude sebacic acid and ethyl hydrogen sebacate. In order to separate these two compounds the mixture was extracted with 1.5 liters of petroleum ether, or Skelly "B", «1.. by letting the petroleum ether stand on the residue for one hour at room temperature with occasional stirring. The petroleum ether dissolved the ethyl hydrogen sebacate leaving the sebacic acid undissolved. The sebacic acid was filtered with suction and washed with a little petroleum ether. The petroleum ether solution and washings were combined and distilled at atmospheric pressure over a steam bath. This left ethyl hydrogen sebacate as a residue. The crude half-ester was distilled in vacuo using a l25-ml modified Claison flask with a to cm vigereau column. The main portion came over at 170—17ho C/l mm. It was colorless. The half-ester was re- crystallized from Skelly “B" using approximately twice the volume of Skelly "B" as half-ester. The yield of ethyl hydrogen sebacate melting at 35.5-37.0O C was 62.2 gm (5h.l% based on sebacic acid). The yields from three other runs were h3.h%, 5h.7$ and 52.6%. One r? run following the method of Jones (6) yielded 55.2”. Another run followed the method given in "Organic Synthesis" (16) and yielded 132 gm (57.L%) from one mole of sebacic acid. Etlyl hydrogen sebacate was also prepared in a manner similar to the above, except that when the crude half-ester was obtained it was not distilled, but simply cooled and recrystallized; yield 56.h% The preparation of ethyl hydrogen sebacate was carried out in still as )1), another way. In this method the diethyl sebacate was separate described above, leaving a mixture of sebacic acid and the half-ester. This mixture was then distilled in vacuo, but it was found difficult to obtain a good separation of these two compounds, and a much longer time was necessary in this procedure. The Preparation of Ethyl Silver Sebacate One liter of distilled water, in which 16.5 gm of potassium hydroxide was dissolved, was added with stirring to 57.5 gm (0.25 mole) of ethyl hydrogen sebacate (m.p. 3h-350 C.). If the ethyl hydrogen sebacate was not completely dissolved more potassium hydroxide was added. Dilute nitric acid was then added until a permanent cloudiness was obtained. This cloudiness is due to a small amount of ethyl hydrogen sebacate being precipitated. At this point the solution was neutral to litmus. A solu- tion of h2.5 gm (0.25 mole) of silver nitrate in 930 m1 of water was added slowly, with vigorous stirring, to the solution of ethyl potassium sebacate. The ethyl silver se‘acate which precipitated in small curds was filtered with suction in a 6-inch Buchner funnel and washed with one liter of water and twice with 300-ml portions of acetone. The ethyl silver sebacate was dried in vacuo over phosphorous pentoxide and ground A in a mortar. A second 0.25 mole batch of ethyl silver sebacate was pre- ‘ O pared in a similar manner. A third and fourth bate} was prepared with a ..2 reduced rate of stirring as the only variation from the method described above. The precipitates from the latter two batches were coarser and more easily filtered than those from the first two. The yield from one mole of ethyl hydrogen sebacate was 251 gm (0.7h8 mole or 7h.€3). The Preparation of Ethyl 9-Bromononanoate A 2—1iter three-neck round bottom flask was fitted with a therm- ometer well, reflux condenser, drOpping funnel and a stirrer. A calcium chloride tube was attached to the top of the condenser, and another to the dropping funnel. Moisture was removed from the apparatus by rinsing it with carbon tetrachloride which had been dried with phosphorous pentoxide and distilled. Five hundred milliliters of dry carbon tetra- chloride and 251 gm (0.7h8 mole) of dry finely powdered ethyl silver sebacate were added to the 2-liter flask. One hundred nineteen grams or 38.1 ml (0.7h8 mole) of bromine (dried over phosphorous pentoxide) was dissolved in 300 ml of dry carbon tetrachloride and poured into the drooping funnel. Stirring was found to be difficult, and :00 ml more dry 1. carbon tetrachloride was added to the reaction flask to facilitate the stirring. with constant stirring. After all the bromine had been added, the re- action mixture was refluxed one and one-half hours. Bromine was not in The bromine solution was added over a forty-five minute period ‘1’."a-I 1? excess at this point. Additional bromine was added at room temperature until an excess was present (reddish brown color). The reaction mixture was then refluxed forty—five minutes with stirring. After cooling, the precipitate of silver bromide was filtered off and washed h—é times with carbon tetrachloride the washings being added to the filtrate. This combined carbon tetrachloride solution was washed with 5% sodium bisulfite solution; then with two 250-m1 portions of 2% sodium carbonate and twice with water. Anhydrous sodium sulfate was used to dry the carbon tetrachloride solution. After drying one day, the carbon tetrachloride solution was decanted from the sodium sulfate and distilled. The carbon tetrachloride was re- moved at atmospheric pressure and the residue was then distilled through a 12-inch Podbielnick column yielding 1,8-dibromooctane (b.p. l36-lh0O C/8 mm., and n23 l.h987) and ethyl 9-bromonanoate (b.p. 151-151;O 0/8 mm n2g 1.h620). Elemental analysis of ethyl 9-bromononanoate Theoretical C h9.l9% h9.81 H 8.00 7.98 Lr 30.3h 30.1u A boiling point of lhd-lSZO 0/8 mm has been reported for ethyl 9-bromononanoate. The Preparation of Silver Sebacate Fifty and one quarter grams (0.25 mole) of sebacic acid (Eastman white label) was dissolved in 1 liter of potassium hydroxide solution containing LO gm (0.5 mole) of 85% potassium hydroxide pellets (Merck's fine—4 18 reagent grade). The solution of potassium sebacate was acidified with 10% nitric acid solution until a little sebacic acid precipitated and remained out of solution when stirred. Eighty five grams (0.5 mole) of silver nitrate (Baker's C.P. grade) in one and one-half liters of dis- tilled water was added with mechanical stirring to the potassium sebacate through a dropping funnel. The precipitated silver sebacate was fil- tered with suction, washed well with water and then with acetone. The product was placed in a vacuum desiccator over phOSphorous pentoxide to dry. It was then ground and redried. In this and several similar experiments the yields of silver seba- cate ran from 90 to 96%. The Preparation of 1,8-Dibromooctane Two hundred seven grams (0.689 mole) of silver sebacate was sus- pended in 750 ml of dry carbon tetrachloride in a dry three-neck flask fitted with a mechanical glycerine-sealed stirrer, a reflux condenser, thermometer well and a dropping funnel. Calcium chloride tubes were attached to the tOp of the reflux condenser and the dropping funnel. Two hundred twenty-one grams, or 70.8 ml (1.38 moles) of bromine, which had been dried by shaking with two 80-ml portions of concentrated sul- furic acid, was dissolved in 200 ml of dry carbon tetrachloride. This solution was added by means of the dropping funnel to the suspended silver sebacate over a forty-five minute period with efficient stirring. A bath of cold water kept the reaction mixture near room temperature “1““! 19 during the addition. The silver sebacate turned yellow as the reaction proceeded. Excess bromine which was indicated by the color of the solu- tion was present at the end of the addition. At this point the reaction mixture was refluxed for thirty minutes. After cooling, the silver bromide was filtered off and washed with carbon tetrachloride. The filtrate and washings were combined and shaken with hOO-ml of a 5% sodium carbonate solution. The carbon tetrachloride solution was then separated and washed twice with 200 ml of 5% sodium carbonate solution, then once with 100 ml of distilled water. Finally the carbon tetrachloride solu- tion was dried over anhydrous sodium sulfate. When this solution was distilled, 97.6 gt (51.9% yield of 1,8-dibromooctane was obtained based on silver sebacate). This product boiled at 129-1320 0/5 mm and had an index of refraction of l.h987 at 200 C. These values in the literature are b.p. 150—1510 0/20-25 mm and 92-930 c at 0.14.5 mm and n13 1.5011. The Preparation of 9—Carbethoxynonanoyl Chloride Eighty and five-tenths grams (0.35 mole) of ethyl hydrogen sebacate (m.p. 3b,.O-35.5O C.) was put into a 500-ml round bottom flask provided with a reflux condense . Then 83.3 gm or 50.8 ml (0.70 mole) of thionyl chloride was poured slowly through the condenser (under a hood). A tube of calcium chloride was then connected to the tOp of the reflux condenser. Hydrogen chloride and sulfur dioxide were given off. The reaction mixture was allowed to stand twelve hours. The excess thionyl chloride was then distilled off at atmospheric pressure. The 9-carbethoxynonanoyl chloride er. n”..- 20 in the residue was distilled in vacuo (b.p. l37~lhlO C/less than 1 mm) and yielded 83.3 gm (89.73 based on the ethyl hydrogen sebacate). The Preparation of the Grignard Reagent of 1,8-Dibromooctane Twenty-four and eight-tenths grams (1 mole) of magnesium turnings were placed in a dry 3-liter round bottom flask fitted with a reflux condenser, mechanical glycerine-sealed stirrer and a drOpping funnel, and covered with 100 ml of dry ether. The usual precautions for the making of a Grignard reagent were observed except that no inert atmos- phere was used. The stirrer was started and about 25 ml of 1,8-dibromo- octane was added all at once. A small amount of a white cloudy precipi- tate soon appeared in the flask. The remainder of the liquid was added at such a rate as to keep the reaction at reflux. A total of 137 gn (0.501 mole) of l,8-dibromooctane was added. This solution was analyzed by withdrawing two lO-ml samples of the Grignard reagent with a graduated pipette and hydrolysing by running into 500 ml flasks containing approximately 50 ml of distilled water. An excess of approximately 0.3N standard sulfuric acid was added to dis- solve the precipitated basic magnesium bromide. It was necessary to warm the mixture to complete solution of the basic salt. Approximately 0.26N standard potassium hydroxide was used to back titrate the excess sulfuric acid to a phenolphthalein endpoint. The total Quantity of the solution of Grignard reagent was found by filling a flask of egual size and shape as that containing the Grignard reagent with water to the same 21 level as the solution of Grignard reagent in the reaction flask, and then measuring the volume of the water. The yield of Grignard reagent was calculated to be 0.83 mole or 83% based on the amount of magnesium used. The Preparation of 1,8-0ctamethy1ene Dizinc Chloride One hundred fifty grams (1.1 mole) of freshly fused and powdered zinc chloride was dissolved in approximately 800 ml of anhydrous ethyl ether. This zinc chloride solution was added with stirring to C.h mole of Grignard reagent from 1,8-dibromooctane (see previous page) rapidli enough to maintain the ether at reflux. The reaction mixture was heated at reflux for two hours after all of the solution of zinc chloride had been added. The use of an electric mantle for heating caused the mag- nesium bromide which precipitated during the reaction to cake on the sides of the flask. The Preparation of Diethyl 10,19—Dicxo-l,28-Octacosandioate One liter of anhydrous benzene was added to the flask containing the 1,8-octamethylene dizinc chloride (see preceding page) and the mixture was distilled until the temperature reached 780 C. This was done to raise the reaction temperature which increases the reaction rate and to eliminate possible side reactions due to the presence of ether. Three hundred grams of 9-carbethoxynonanoyl chloride in.600 ml of dry benzene was added with stirring over a fifteendminute period. The reaction mixture was then 22 refluxed one hour before the formation of a gummy mass stOpped the stirrer. After cooling g, the reaction mixture was hydrolysed by extraction with 500 m1 of 2N hydrochloric acid. (Enough hydrochloric acid was used to effect complete solution of all inorganic salts. The gummy mass also dissolved at this point.) The benzene solution was then extracted twice with 200-ml portions of 5% sodium hydroxide and then washed once with water. The benzene was evaporated yielding an impure product weighing 225 gm and melting at 55-600 C. After six recrystallizations from methanol, diethyl 10,19-dioxo-l,28-octacosandioate (m.p. 75° c.) was obtained. The yield was 65.6 gm (30.5% based on the 0.h mole of Grignard reagent). The product was recrystallized a seventh time (m.p. 78.0—78.6O C.) and analysed. Elemental analysis Theoretical C 71.33% 71.33% H 10.82% 10.85% The Preparation of 10,19-Dioxo-l,28-0ctacosandioic Acid Ten grams of diethyl 10,19-dioxo-1,28-octacosandioate was saponified by refluxing it with 1N alcoholic potassium hydroxide solution for three hours. The ethyl alcohol was evaporated off leaving the potassium salt as a residue which was subsequently dissolved in 600 ml of distilled water. The organic acid was precipitated out by the addition of excess concentrated hydrochloric acid, filtered with suction and washed with water. After recrystallizing the organic acid four times from glacial acetic acid, 5.6 gm of 10,l9-dioxo-1,28-octacosa dioic acid (m.p. 139.9- 1M)O C. uncorrected) resulted. 23 To find the equivalent weight of 10,lQ-dioxo-l,28-octacosandioic acid two samples of the purified acid of approximately 0.6 gm were ac- curately weighed into two 300-ml flasks. The samples were dissolved in 100 ml of hot neutralized ethyl alcohol and titrated immediately with a standard methyl alcohol solution of potassium hydroxide which was approxi- mately 0.1 N. After 75-80% of the theoretical amount of potassium hydroxide solution had been added, 30 ml of carbon dioxide free distilled water was added, and the mixture was heated on a steam bath to dissolve the potassium salt which had precipitated during the titration. The ti— tration was carried to a phenolphthalein endpoint. The eguivalent weights found were 2hl.2 and 2h2.h; the theoretical equivalent weight is Zhl.3. Preparation of Derivatives It was thought necessary to prepare derivatives of two of the com- pounds synthesized during this investigation, the nine-carbon bromine compound and the long chain diketone. Four direct and indirect derivatives of ethyl 9-bromononanoate were prepared--the bromo acid, the amide of the acid, and the S-alkyl iso- thiourea picratis of both the ester and the acid. The phenylhydrazone (2,h-dinitro) of the long chain diketone was prepared. The details of these preparations are given on the following pages. "Wu—II 2L; 9—Bromononanoic Acid Sixteen and two-tenths grams of ethyl 9-bromononanoate, 30 ml of glacial acetic acid and 1.8 ml of LEA hydrobromic acid were poured into a 100 ml flask fitted with a reflux condenser. The .ixture was refluxed for at least eight hours. The reaction mixture was transferred to an evaporating dish and the acetic acid evaporated on a steam bath until a brown liquid which had only a slicht odor of acetic acid remained. The crude 9—bromononaoic acid was distilled in vacuo at 155-15000/2 mm and then recrystallized twice by freezing it out of pe roleum ether using O O I i O 1‘ I a dry ice-acetone bath. The solid aCid meltec at 3h.7-36.0 C. Neutrali- zation eguivalent: 236.3; theoretical: 237.1. The Amide of 9-Bromononanoic Acid Two hundred mg of 9-bromononanoic acid and 0.8 ml of thionyl chloride were heated together in an eight-inch test tube, which had a cold finger condenser attached, at 75—650 C. using a water bath. The crude acid chloride which formed was cooled in an ice bath and 5 m1 of ice cold cen- centrated (28%) ammonia solution was added slowly. The mixture was kept in the ice bath for five minutes and shaken occasionally. The solid amide was filtered off and washed. After recrystallizing four times from an , . . . . . O alc0nol-water mixture, the melting pOint of this amide was 85.0-55.3 0. [‘0 MI The S-Alkylisothioureapi Ethyl 9-Bromo- Six drops of ethyl 9-bromononanoate, 300 mg of thiourea and 5 ml of ethyl alcohol were refluxed one and one-half hours in an eight inch test tube fitted with a cold finger condenser. Then 1 ml of a saturated solution of fiCTiC acid in alcohol we added. Long yellow and colorless needles formed as the solution cooled. The yellow crystals melted at 1780 C. and the colorless ones at 1800 C. Since thiourea melts at 1800 C. it is pr bable that these were crystals of thiourea. he mother liquor from these crystals was diluted with water and the resulting pre- cipitate was recrystallized from an alcohol-water mixture. After three 1 recrvstallizations, crystals of the S-alkyl isoiniourea picrate of ethyl . o . 9-bromononanoate melting at ll} C. were obtained. The SeAlkylisothioureapicrate Derivative of 9-Brcmononanoic Acid Two hundred milligrams of 9-bromononanoic acid, 300 mg of thiourea and 5 ml of alcohol were refluxed together for one and one-half hours. Yellow crystals of the S-alkyl isothiourea picrate of 9-bromononanoic acid were obtained by recrystallization from a 50% alcohol—water mixture. 0 a I O s I o The melting pOint was 159.5-191 C. uncorrected after four recrystalliza- tions. The 2,h-DinitrOphenylhydrazone De r Diethyl lO,l9-DiCXO-l,2t -Octa CO _;V so 1 Sc‘flO. Two-tenths of a gram (0.001 eq) of 2,h-dinitrOpherylhydrazine dis- solved in a minimum of hot 95% ethyl alcohol was added to 0.27 gm (0.001 eq) of diethvl 10,19-dioxo-l,28-octacosan n