THE REACTION OF ETHYLENE OXIDE WITH VARIOUS GRIGNARD REAGENTS by Albert Henry Agett A THESIS Submitted to the Graduate School of Michigan State College of Agriculture and Applied Science in partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry. 1940 ProQuest Number: 10008482 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest ProQuest 10008482 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 ACKNOVfLEDGEMEM? To Dr. S. C. Huston I wish to express my gratitude for the many suggestions and the helpful understanding that have made possible this work. A. H. Agett comas Introduction Historical Experimental I. Preparation of "bromides and materials used. IX. Preparation of Grignard reagents. III.Beactions of Grignard reagents with ethylene oxide. A. One mole ethylene oxide. B. !Two moles ethylene oxide. C. Dialkyl magnesium compounds. 17. Preparation, analysis, and identification of intermediate compounds formed. 7, A. One mole ethylene oxide with one mole of Grignard reagent. B. One mole Grignard reagent with one mole ethylene hronohydrin. C. One mole n-arnyl magnesium "bromide with one mole ethylene oxide, D. One mole magnesium "bromide with one mole ethylene oxide. E. One mole tertiary "butyl magnesium "bromide with one mole ethylene oxide. P. One mole ethyl Grignard with one mole of ethylene chlorohydrin. One mole magnesium "bromide with ethylene oxide; the product reacted with ethyl Grignard reagent. VI Reaction of one mole of ethylene oxide with one mole 17 of propyl magnesium bromide VII Hydrolysis of the reaction mixture from the one mole of magnesium bromide with two moles ofethylene oxide 18 VIII Reaction of one mole of tertiary butyl magnesium bromide with two moles ofethylene oxide 18 IX Preparation of 3,5 dinitrobenzoate of the alcohols 20 X Preparation of the phenyl end alpha-naphthy1 urathanes of some of the alcohols 20 XI Proof of structure for 3,4 dimethyl pentanol-1 XII Proof of structure for 3,3 dimethyl butanol-1 21 23 Theoretical 26 Discussion 30 Summary 34 Tables 37 Bibliography 39 INTRODUCTION The author noted when tertiary butyl magnesium bromide reacted with ethylene oxide that the expected alcohol was not obtained. The main product of the reaction was ethylene bromohydrin. The results were the same whether the mixture was refluxed in benzene or hydrolyzed immediately. This anomalous Grignard reaction promoted the investigation. HISTORICAL In the year 1902 Blaisse (l) reacted ethylene oxide with ethyl magnesium "bromide. He obtained as the principle product ethylene "bromohydrin. The following scheme was for­ mulated by him to interpret the results: BrCHg-CHg-O-lfeBr (I) The product (I) would give ethylene bromohydrin upon hydro­ lysis. But according to this the Grignard reagent is broken between the magnesium and the bromine. All previous cases were explained by the separation of the alkyl group from the magnesium halide. In 1903 Grignard (2) published an article in which he expressed his views on the manner of the reaction. He explained the formation of the bromohydrin obtained by Blaisse through a secondary reaction in which ethylene oxide formed an addition compound with the following formula: (II) This addition compound was called an oxonium salt by Grig- nard. It would give upon hydrolysis the following: CHo C2H5 ] ^0 GH2 CH2 -v- 2H0H— 2 MgBr | >) + 2 C 2H 6 + MgBrg + Mg(0H)2 CH2 The ethylene oxide in the presence of magnesium "bromide and water would give ethylene brcmohydrin according to the known reaction of Wurtz (3). GH2 2j ^0 2H 0 H % B r 2 3- 2BrCH2CH20H -+- %(0H)2 ch 2 Consequently, the formation of the ethylene Bromohydrin was a secondary reaction. Therefore, after the ethylene oxide was added to the ethyl magnesium "bromide, Grignard distilled off the ether on a water "bath in order to heat the mixture. After a large amount of the ether had been removed a second reaction took place. to bubble and swell. The mixture began If it was not controlled very care­ fully an explosion took place with the evolution" of a great deal of heat. phase by Grignard. This reaction was called the second Upon hydrolysis the mixture then yielded 82$ of normal butyl alcohol. This second phase of the reaction could also be produced by distilling off some of the ether and replacing it with benzene or toluene. The new mixture was refluxed for several hours and then hydro­ 3- lyzed as before. According to Grignard the mechanism of the reaction was the rupture of the oxonium salt by heat as shown "below: CHg GgHg I ^0^ CH2^ MgBr Heat--- G2H5 -GH2-GH2QMgBr (III) The product (III) upon hydrolysis would yield normal "butyl alcohol, Hie mechanism of the above reaction has "been generalized by Grignard, (2) L. Henry (4), Delby (5), and others using ethylene oxide and different Grignard reagents. Mieisenheimer (6) contributed analytical data in confirmation of Grignard's ideas. He analyzed the pro­ duct which was precipitated when ethylene oxide was re­ acted at -21° with ethyl magnesium bromide. The results corresponded to the molecular addition between the Grignard reagent and ethylene oxide. Meisenheimer assigned the fol­ lowing formula to this product: h 2c H 6 - O E CHj ch2 gh 2 The primary product of reaction between a Grignard reagent and eyclohexene oxide was the halohydrin derivative, which gave the smaller ring when heated. Ihen eyclohexene oxide was heated with magnesium bromide the cyclopentane aldehyde was obtained (13). Consequently when the dimethyl magnesium was reacted *10- with the eyclohexene oxide the product was 2-methyl eye lohexanol, the expected compound. It has been observed in this laboratory that when tertiary butyl magnesium bromide reacts with ethylene oxide that the expected alcohol was not obtained. The main pro­ duct of this reaction was ethylene bromohydrin. This had been observed by Whitmore (14) when he reacted tertiary butyl magnesium chloride with ethylene oxide and obtained ethylene chlorohydrin as the main product. EXPERIMENTAL I. The following bromides were prepared from the correspond­ ing alcohols (Eastman Technical) by the action of su.lfti.ric acid and hydrobromic acid (15): Ethyl bromide N-Propyl bromide iso-propyl bromide N-butyl bromide iso-butyl bromide sec* "butyl bromide N-amyl bromide iso-amyl bromide 1 bromo 2 methyl butane 2 bromo pentane N-hexyl bromide 2 bromo hexane 2 bromo 4 methyl pentane Tert* butyl bromide Tert. Aoayl branide 2 bromo 3 methyl butane was prepared from the corresponding carbinol (16) by R. L. Guile. Bromocyclohexane, phenyl bromide, ethylene chlorohydrin, and benzyl chloride were Eastman Technical. They were frac- tdonated and dried over calcium chloride. The dioxane was Eastman Technical and this was fractionated and dried over sodium. Ethylene oxide m s obtained from Dow Chemical Company and was fractionated and dried by passing it over soda lime. The ethylene dibromide was obtained from the action of bromine with ethene (17). It was fractionated and dried over calcium chloride. Ethylene bromohydrin was prepared by the action of mag­ nesium bromide with ethylene oxide in an anhydrous ether sol­ ution, It was purified by fractionation. Magnesium turnings, prepared for Grignard reagents, were used after drying over calcium chloride. The anhydrous ether was dried over sodium for one week before it was used in the reactions. Grignard reagents prepared and studied were: Ethyl magnesium bromide Propyl magnesium bromide Iso-propyl magnesium bromide N-butyl magnesium bromide Iso-butyl magnesium bromide Sec. butyl magnesium bromide Tert. butyl magnesium bromide. N-amyl magnesium bromide Iso amyl magnesium bromide 2 methyl 1 bromo-magnesium butane 2 "bromo magnesium pentane 3 methyl 2 branomagnesium butane N-hexyl magnesium bromide 4 methyl 2 bromomagne sium pentane 2 bromo-magnesium hexane Phenyl magnesium bromide Bromo-magnesium cyclohexane Dibromo-magnesium ethylene II. The Grignard reagents were prepared as follows: In a dry1 two liter three neck flask, fitted with a glycerine sealed stirrer, efficient reflux condenser and dropping funnel, were placed twenty-four grams (one mole) of magnesium turnings and a crystal of iodine. The flask was heated with a small flame until the iodine vaporized. It was then allowed to cool while one mole of the halide was weighed out. Two mis. of the halide was added to the re­ action flask and the remainder-was mixed with two hundreds mis. of anhydrous ether. After the reaction mixture stood for a few minutes a few mis. of anhydrous ether were added and stirring was started. The reaction began almost im­ mediately and after it had reacted for a few minutes, or if the reaction became too vigorous, fifty mis of anhydrous ether were placed in the flask. The remaining halide ether ^The flask was dried in an oven for twenty-four hours and then rinsed twice with anhydrous ether. The reaction must be protected from carbon dioxide, water vapor and oxygen of the air. -4— solution was then placed, in the dropping funnel. It was added at such a rate that gentle reflux took place. After all the halide ether solution had been added the mixture was stirred for two hours and allowed to stand overnight. Each Grignard reagent was treated in three different ways. First, the Grignard reagent was reacted with one mole of ethy­ lene oxide and then refluxed with benzene. Second, the Grig­ nard was reacted with two moles of ethylene oxide with no heating. (Third, the reagent was treated with dioxane (18) precipitating the halides. (The ether soluble dialkyl mag­ nesium compound was then reacted with two moles of ethylene oxide. A typical run for each of the three reactions will be described, III A. Beaction of one mole of a Grignard reagent with one mole of ethylene oxide with heating. The Grignard reagent that has been previously prepared was cooled in an ice salt bath. One hundred mis. of anhydrous ether that had been cooled to zero degrees was placed in a dropping funnel fitted with a tight cork. One mole (44 grams) of ethylene oxide was measured out in a flask previously cal­ ibrated and cooled to zero degrees. into the dropping funnel. tents mixed. It was poured immediately The funnel was corked and the con­ Addition of the ethylene oxide ether solution to the Grignard reagent was started cautiously. fast, a very vigorous reaction takes place.) (If added to The ice salt -5- bath was removed as soon as all of the ethylene oxide had been added and the contents were allowed to come to room tem­ perature with vigorous stirring. The reflux condenser was now set for d i s t i l l a t i o n . A water bath kept at fifty degrees^ \a/(JLs used for the removal of about two hundred and fifty mis. of ether, until the contentns become pasty. Now the con­ denser was set for reflux and two hundred and fifty mis. of anhydrous benzene were added. The water bath was heated until the benzene solution refluxed gently. Heating was continued with stirring for six hours and the mixture allowed to stand overnight. of water. It was hydrolyzed with one hundred mis. The magnesium hydroxide precipitated as a thick paste and the benzene solution was decanted. The paste of magnesium hydroxide and water was extracted three times with small portions of ether. hydrochloric acid and ice. It was then neutralized with dilute This solution was extracted three times with small amounts of ether. The ether extracts were combined, dried over anhydrous sodium sulfate and fractionally distilled. Ill B. Beaction of two moles of ethylene oxide with one mole of Grignard reagent. The Grignard reagent that was prepared as in II was ^-All precautions for ether distillation are used. 2If the temperature of the contents is raised above fifty degrees a very vigorous reaction takes place with decompo­ sition. cooled in an ice salt bath. One hundred and fifty mis. of anhydrous ether cooled to zero degrees was poured in the dropping funnel. mixed. Addition The funnel was corked and the liquids of the ethylene oxide solution to the Grignard was q,uite slow due to the vigorous reaction. The ice salt bath was removed as soon as all of the ethylene oxide was added and the contents were allowed to come to room temperature with stirring. After standing overnight the material was hydrolyzed as described in (A). The com­ bined ether extracts were dried over anhydrous sodium sul­ fate and fractionated. The alcohols may be freed of any ethylene bromohydrin by extraction then twice with water. This was avoided cohol. if possible due to the large loss of al­ However,where the boiling point of the alcohol and the ethylene bromohydrin were close together it was essential. Ill C. Beaction of two moles of ethylene oxide with one mole of dialkyl magnesium. Dioxane was added through a dropping funnel to the stir­ red Grignard reagent at such a rate that gentle reflux took place (18). The addition of dioxane was continued until no more halides were precipitated. This could easily be seen by stopping the stirrer and letting the white precipitate settle. ether solution one drop of dioxane was added. To the clear If a white pre­ cipitate was formed it was necessary to add more dioxane. iThe yield of the dialkyl magnesium compound was diminished to a large extent if a great excess of dioxane is added. The mixture was then q.uickly and carefully poured into dry centrifuge tubes and corked. The white precipitate of the halide was thrown down so that the clear ether solution of the dialkyl magnesium compound could be decanted into a flask fitted as previously described for a Grignard reaction. The total volume of the ether solution was measured. A two ml. aliquoit was removed and hydrolyzed with ten mis. of water. The magnesium hydroxide was titrated with standard .1 N HC1 solution using phenolpthalein as an indicator. The best re­ sults were obtained if a measured excess of the standard . I N acid was added and then back titrated with standard .1 N NaOH. Erom the data obtained one can quickly compute the yield of the dialkyl magnesium compound. It was now possible to add the theoretical amount of ethylene oxide (two moles), which is weighed and added as previously described in III B. The solution was stirred for three hours and allowed to stand overnight. It was then hydrolyzed with enough water so that the magnesium hydroxide becomes a thick paste. The ether so­ lution was decanted and the paste of magnesium hydroxide was extracted three times with small portions of ether. The com­ bined ether extracts were dried over anhydrous sodium sulfate and fractionated. Hie following alcohols were prepared by the three methods described: normal butyl alcohol normal amyl alcohol iso amyl alcohol 3,3 dimethyl butanol 1 3 methyl pentanol 1 iso hexyl alcohol normal hexyl alcohol normal heptyl alcohol 3 methyl hexanol 1 iso heptyl alcohol 4 methyl hexanol 1 normal octyl alcohol 3,4 dimethyl pentanol 1 3 methyl heptanol 1 benzyl carbinol 2 cyclohexyl ethanol 1 3 phenyl propanol 1 IV. Preparation, analysis and identification of the inter­ mediate compound that was formed in the reaction of ethy­ lene oxide with Grignard reagents A. Beaction of one mole of ethylene oxide with one mole of ethyl magnesium bromide. The preparation of the Grignard reagent and the addition of the ethylene oxide was accomplished as previously described in III A. The mixture was stirred for two hours at room tem­ perature and was allowed to stand for one hour. The white pre­ cipitate settled out and the ether solution was quickly decanted. The precipitate was washed four times with one hundred ml. por­ tions of anhydrous ether. The washed solid was quickly trans- -9- ferred to a weighing "bottle which was immediately placed in a vacuum desiccator.^ The desiccator was first evacuated with a water pump for three hours and then with an oil vacuum pump at two mm. pressure for twelve hours. removed. All of the ether was The following aaalysiswfcre made: Total "bromine "by Parr bomb (19). In a Parr bomb fhsion cup were placed fourteen grams (14 g.) Of sodium peroxide, seventy-five hundreths grams (.75 g) of powdered potassium nitrate, four-tenths grams (.4 g) of finely powder cane sugar, and two-tenths grams (.2g) of the sample. The Parr bomb was assembled ami the contents were carefully mixed by shaking. The sample was ignited by heating two to three minutes with a bunsen burner. The bomb was allowed to cool, opened and the fusion cup placed in a six hundred ml. beaker. The top was thoroughly rinsed off in the beaker with distilled water. A cover glass was then placed on the beaker and two hundred mis. of water were added. As soon as all of the material had dissolved from the fusion cup it was removed and rinsed with distilled water. The solution was boiled until all of the hydrogen peroxide was expelled and all the sodium peroxide decomposed. A measured excess of standard .1 N silver nitrate was added to the hot solution which is then heated for fifteen minutes to insure complete precipitation of the silver bromide. The solution was cooled and carefully acidified with iThe white solid hydrolyzed immediately if exposed to the moisture of the air. -10- concentrated nitric acid. Hydrazine sulfateufcs added until all reaction ceases and no more nitrogen is given off. The solutionis cooled and the bromine determined by the Volhard method for bromine (21). Total magnesium was determined by decomposing a weighed sample with a Meeker burner. The results compared with the method of converting the magnesium to magnesium sulfate , dried and weighed as such (3). The former m e t h o d m u c h faster; ignition capsules were ignited to a constant weight. A weighed sample was placed in each and carefully ignited with a small flame (care must be taken not to heat so fast that the sample catches fire). The temperature was gradually raised until the full heat of the Meeker burner was used. was ignited at this temperature for one hour. The sample They were then allowed to cool to almost roan temperature and placed in a des­ iccator for fifteen minutes. The crucible was weighed and ig­ nited for ten minutes at the highest Meeker burner temperature. It was again allowed to cool as above and weighed. This pro­ cess was repeated until the crucible attained constant weight. Generally, two ignitions were all thaliwere necessary. From the data it is possible to calculate the percent magnesium in the sample. By this formula the amount of magnesium was obtained: wt. MgO x .6032 x 100 * wt. of sample ° C HgC ^XMgBr c2 H 5 Heat — ^CgHg-CHgCHg-O-MgBr MgBr CgHg- CHgCHg- 0~MgBr +• HOH — ^ CgHg-CHgCHgUHgOH -+- MgBr OH The mechanixm was given analytical support hy Messenheimer (6 ). He assigned the following formula to the molecular addition compound: H2C I HgC C2 H5 \ / C2Hg ' ' I \ Cgife N0 ------M g ----------- 0 Br This idea remained unchanged until Ribas and Tapia (3), {?), (8 ), (11) published their series of articles. According to them the main portion of the reaction goes as follows: Reaction of two moles of ethylene oxide with one mnla-nf etJwfl magnesium bromide. (2CgH5]%Br ^^(CgHgJgMjgrHgBrg) + 4 ■+■ (CgHgCHgCHgOgUg (Br„CHg-CHg-0)2% {Br-CHg-CHg-OlgHfe 2 -t (CgHgCHgCHgOjglfe -+■ 4 HOH — 2 CgKgCHgCHgOH ~t H C 2 |)o 2Br-CHgCH^OH ■+• 2 Mg(QH)g jteaction of one mole of ethylene oxide with one mole of ethyl magnesium “bromide. (2CgH5 ]S%Br ^^(CgHgJgMg "I” MgBrg) -+- 2 V v )^>0 — »■ (BrGH^CHgOjgMg (CgHg)gMg (BrCHgCHgO)glg i.fCgHgJglg + Heat — -5 * (CgHgCHgCHg-OjgMg -+■ MgBr2 (CgH^CHgCHgO)gMg •+MgBrg+ 2 HOH — ■* 2 CgHgOHgCHgOH -* MgBr2 +Mg(OH)g (I) was said to he a mixture of BrCHaCHx-0-MgBr and (BrCH^CH,-0-)^lg, although all of their analytical data confirms the formula B r C H O - M g B r with fractional parts of ether ab­ sorbed. The following series of reactions are the result of work accomplished in this laboratory. Where R is used it refers to any alkyl or aryl grouping other than a tertiary alkyl1 or ben­ zyl group ^Tertiary butyl group did react under certain conditions as seen in the experimental part to give the expected alcohol. Two moles of ethylene oxide with one mole of G-rignard. (2BMg|^£=^B2Hg + I%Br2 ) + 4 HgC |^0 -- => (BrCH2CH2-0)2Mg h 2g (RCHg GHg-0)gig (BrGH2CH2-0)2Mg ~t~ (HCH2CH2-0)2Mg + 4 HOH + 2SCH2CH20H + 2BrCHgCHgQH 2 Mg(OH)2 One mole of ethylene oxide with one mole of magnesium bromide h 2g 2I%Br2 ---- =» (BrCH2CH20)2% 2 +. I%Br2 h2 c ^ (BrCH20H20)2Mg 4- MgBrg + 2 HOH -- =» 2 Br GHg GHgOH + % B r 2 -f % ( o h ) 2 Two moles of ethylene oxide with one mole of magnesium bromide. 2 H2 G^ | H CT (BrCH2 CH2- 0 ) 2% + MgBr2 4- ----^ (BrGH2GH2-0)2Mg 2 HOH — =^2BrGH2 GH2 OH M g (o jl)2 Pae mole of ethylene ~bromoh.yd.rin with one mole of GrIgnard HS%35k+ BrCH2 CH2 0H — » BH + Br CHgCHgOMgBr BrGHgGI^OMgBr + 1 HOH =*> Br-CH2CH2-0H + MgBrOH - + BH One mole of ethylene chlorohydrin with one mole of Grignard. C1CH2 CH2 0H -- » BH ClCH2 CH2 0l%Br + ClGH2 CH2 0 % B r ■+ 1 HOH — -ClCHgCHgOH + %BrOH -j- BH (Pro moles of ethylene oxide with, one mole of dialkyl magnesium H2C r 2% + 2 } \ o --------- (bch2 ch2 o ) 2i % h 2g (RGH2GH20)2% + 2 HOH ---=*2 RGHgGHgOH + Mg(0H)g 30 DISCUSSION The intermediate compound that is formed in the reaction of a Grignard reagent with one mole of ethylene oxide is (BrCHgCHgO and not the oxonium salt, H2 G| , V n0 — H2 ° t%,C 2 % 1 0 Br described by Grignard and Meisenheimer (2 ), (6 ), compound was prepared by two methods. This intermediate First, by the reaction of one mole of magnesium bromide with one or two moles of ethylene oxide. Second, by the reaction of one mole of various Grignard reagents with one mole of ethylene oxide. the molecular formula C^I^OgBrglWg. Analysis agree with When two moles of ethylene oxide react with one mole of magnesium bromide the resulting yield of ethylenebromohydrin is sixty percent based on the two moles of ethylene oxide used. Whereas, if the oxonium compound or BrCHj,CHL-QMgBr were formed, fifty percent would be the maxi­ mum yield possible. Furthermore when the Grignard reagents are crossed as in the experimental part VI the products could not have been formed from the oxonium salt. If the intermediate compound were the oxonium salt the resulting alcohols in each instance would have been the opposite. This also proves that the dialkyl magnesium compound must have been contained in the ether solution that was crossed with the precipitates. This is 31 in agreement with the equations that are proposed "by us in the theoretical part. Bihas and Tapia (3)submit the following experimental evidence as a basis for their proof that the oxonium salt is not formed. /i/hen the intermediate compound is hydrolysed in 1 1:5 nitric acid approximately one half of the total bromine is given up in the fona of the bromide ion. The oxonium salt, according to the formula assigned to it, would have all of the bromine given up. They therefore claimed that the intermediate compound was BrCHgCHg-O-MgBr. It can be seen that this com­ pound would yield one half of the total bromine as the bromide ion on hydrolysis with 1:5 nitric acid. The intermediate com­ pound when hydrolyzed in 1:5 nitric acid gives 27% - 30% bromide ion. If the compound is BrCHgCHg-O-MgBr, one half of the total bromine should be freed as bromide ion. should be 35%. Theoretically this This value has never been attained in any of the experimental results. The percent bromide ion that is ob­ tained, 27% - 30%, is in agreement for the theoretical of (BrCHgCHg-O ) % M g if one half of the total bromine is freed upon hydrolysis (29.3%). We therefore conclude, with the support of the analytical data, that one molecule of ethylene oxide is attached differently, so that one half of the bromine is liber­ ated as the bromide ion upon hydrolysis. In a later paper Kibas and Tapia (11) stated that they "believe that the intermediate compound was a mixture of BrCHgCHg-O-MgBr and (BrCHgCHg-0 )gI%. However, all of their analytical data was in support of the BrCHgCHgO-lUgBr It is a known fact that the reaction of one mole of a "bromo Grignard with one mole of ethylene oxide will yield only ethylene bromohydrin unless heat treated or allowed to stand for a long period, (l), (2). It has been shown in this paper that ethylene oxide will react in the cold, without standing, with the dialkyl magnesium compounds. Also, the reaction of two moles of ethylene oxide with one mole of a Grignard reagent yields the expected alcohol without heating. If the intermediate compound is a mixture of the two compounds, BrCHgCHg-O-MgBr and (BrCHgCHg-OjgMg then the following reactions would take place when one mole of ethylene oxide reacts with one mole of a Grig­ nard reagent: H2C (2BMgX. S^HglJg + -+- 2 + (BrCHgCHgO)gMg HgC and HgC (2 HMgX ♦SRgiidg 2 0 HgC ------ ErCHgCHgGWfeBr -33- The dialkyl magnesium ccmpound and the ethylene oxide that are found as products in the second series of reactions would react as follows: ^2 % +2 H2C |^> H^C HCHgCHgOH 0 --- -> (HCHgCHgO)gldg and (HCHgCH2 0)2% - v 2H0H— ** ■+ I'%(OH)g Therefore one would find measurable amounts of the alcohol. Absence of the alcohol makes it seem very improbable that there could be a very large percent of the single compound present. If the bromine ion produced by hydrolysis (27$ - 30$) is due to the single compound then the mixture, as proposed by Hibas and Tapia, would be about eighty percent BrCHgCHgOMgBr. The unused ethylene oxide would react with the dialkyl magnesium compound to produce the alcohol, contrary to results. j The reaction of benzyl magnesium chloride withethylene i oxide is very interesting. When one mole of benzyl magnesium chloride reacts with one mole of ethylene oxide the alcohol is obtained without heating or standing. chlorohydrin produced is small. The amount of ethylene When two moles of ethylene oxide are used and the anount of benzyl magnesium chloride is again one mole, the yield of the alcohol is practically the same as when one mole of ethylene oxide is used and the amount of ethylene chlorohydrin produced is again small. It appears that the benzyl magnesium chloride is the only Grignard reagent studied which reacts faster with ethylene oxide than do the -34- raagnesium chloride or the dialkyl magnesium produced from it. The only product that is obtained when tertiary amyl magnesium bromide, tertiary hexyl magnesium bromide, or the Grignard that is produced when one mole of ethylene dibromide is reacted with two moles of magnesium, is ethylene bromohydrin. Stem this we conclude that the dialkyl magnesium compounds formed from these Grignard reagants do not react readily, if at all, with ethylene oxide or (BrCHgCHgOlgMg* A technique for reacting tertiary butyl magnesium bromide with ethylene oxide was devised and small yields of the alcohol 3,3 dimethyl butanol-1 were obtained. A proof of structure of 3,3 dimethyl butanol-1 and 3,4 dimethyl pentanol- 1 was obtained by oxidizing the alcohols with alkaline potassium permanganate to the corresponding acid. Each acid was then synthesized. Derivatives of the acids were made, analyzed, melting points determined, and mixed melting points of the derivatives of the corresponding acids showed no depression. The equations outlined in the theoretical part give very good evidence for the support of the Schlenk equilibrium of a Grignard reagent: z BMgX MgXg +- HgMg -35- SUMMARY 1 « The interm.ed.iate compound fomed in the reaction of one mole of ethylene oxide with one mole of the Grignard reagent is (BrCHjCH^-Oj^Mg. All Analytical data support the molecular formul^but the structural formula is not definitely established in as much as one half of the bromine is liberated upon hydrolysis. 2 . The reaction of one mole of ethylene oxide with one mole of a Grignard reagent takes place in two steps as: HgC^ MgBrg) + 2 £o -----> (BrCHgCHgOjgMg h 2c -* RgMg (BrCHgCHgO)glvlg -*■ Rgl% -v Heat — 9 (R-CHgCHgO)gMg 4 -MgBrg (R-CHgCHgO}gMg + MgBrg -*-2HQH — » 2RCHgCHgOH +-Mg (OH)g-j- IvlgBrg 3. One mole of a Grignard reagent will react with two moles of ethylene oxide at room temperature according to the fol1 owi ng e quat ions: HgO ( 2HMg& MgBg ) ■+ 4 EoMg ) ^ 0 ------ - (BrGHgGHgO) gMg + H2c (R-CHgGHgO)£?vlg (BrCHgCHgO) 2R-GH2 GH 2 0H £6g ■+ (R-CHgCHgQ) g;ig -v 4HQH ■ At 21% (OH), 2BrCHgCHgOH + -36- 4. Dialkyl magnesium compounds react with two raoles of ethylene oxide at room temperature to form the expected alcohol. 5. Tertiary amyl and tertiary hexyl Grignards and the Grig­ nard formed from the action of ethylene dibromide do not react with ethylene oxide to fonm the expected alcohol. 6 . Tertiary butyl Grignard reagent did react with ethylene oxide to produce the alcohol in small yields. 7. Benzyl magnesium chloride reacts with one mole of ethylene oxide at room temperature to form the alcohol, 8 . Additional evidence is given in support of the equilibrium equation 9. (2HMgX llgXjjf Bglg) as proposed hy Schlenk (32). Proof of structure of two alcohols, 3,3 dimethyl butanol-1 and 3 ,4 dimethyl pentanol- 1 was given and derivatives pre­ pared. 10. A number of primary alcohols were prepared hy the reaction of a Grignard reagent with ethylene oxide. Table I .Alcohol B.P. at 740 mm. n2 5 d 24° Normal butyl (33) 116°-118° 1.3993 .807 Normal amyl (33) 135°-137° 1.4100 .816 Iso amyl 130° 1.4081 .812 153°-155° 1.4131 .818 151&1520 1.4112 .823 (33) 150°-152° 1.4132 .815 dimethyl butanol - 1 1439-143° 1.4160 .814 Normal heptyl (33) 173°-174° 1.4231 .816 3 methyl hexanol-l (38) 161°-162° 1.4213 .817 Iso heptyl (33) 168°-169° 1.4251 .819 168°-169° 1.4233 .821 190°-193° 1.4303 .824 1.4261 .819 1.4293 .821 (34) Normal hexyl (33) methyl pentanol- 1 3 Iso Hexyl 3 , 3 methyl hexanol- 1 4 (34) (38) Normal Octyl (33) 3,4 dimethyl Pentano1-1 (:39) 160°-162° 3 methyl heptanol- 1 (40) 101 ° @ 26 mm 1.5258 1.027 Benzyl carbinol (37) 217°-219° B-cyclohexyl ethyl 88o -90° © 17 mm 1.4693 .9185 gamma phenyl propyl (41) 233°-235° 1,5351 1.006 Derivative a-naphthyl urathane 71° (35) a-naphthyl urathane 65.5° (35) a-naphthyl urathane 67° (35) a-naphthyl urathane 59° (36) a-naphthyl urathane 58° (36) a-naphthyl urathane 60° (36) 3,5 dinitrobenzoate 83.5° a-naphthyl urathane 62° (36) a-naphthyl urathane 45°-47° (38) 3 ,5 dinitrobenzoate 54.5° (36) a-naphthyl urathane 50° (38) a-naphthyl urathane 6 6 ° (36) amide of the acid 95.5' anilide ” ,f M 67° a-naphthyl urathane 119° (36) 3,5 dinitrobenzoate 70.5° phenyl urathane 47° (36) Table II Yields of Alcohols '\ H2 % + 2 H2 C-GH2 0 j f i B M g X + 2 R z * 6 ~ Q H 2 no heat % yield of alcohol % bromoyield hydrin alcohol % , H1%X+IH 2 G— GH2 heat O s bromohydrin % yield alcohol % Ethyl bromide 88 45 72 11 79 Propyl bromide 90 43 75 6 76 Iso propyl bromide 85 45 70 7 74 N-butyl bromide 82 41 71 5 70 Sec.butyl bromide 70 51 65 9 65 Iso butyl bromide 80 41 69 3 64 Tert. butyl bromide 9 50 none 60 none EE-amyl bromide 75 40 60 10 69 Sec. amyl bromide 60 44 50 5 53 Iso amyl bromide 70 41 56 4 59 68 35 53 9 58 Tert. amyl bromide none 48 none 42 none Tert. amyl chloride none none none N hexyl bromide 50 47 chlorohydrin 39 4 1 -brcmo 2 methyl butane chlorohydrin 35 30 49 2 bromo 3 methyl Butane 45 41 46 10 40 55 47 40 6 43 2 bromo Pentane 48 42 Tert. Hexyl bromide .42 58 Phenyl bromide 72 50 55 Gyclohexyl bromide 60 42 45 8 50 Ethylene dibromide none 68 none 51 none 5 79 3 73 Benzyl chloride bibliography 1. Blaisee - 0 . R. 134 . 552 (1902 ) 2. Grignard - Bui. 29., 944 (1903) 3. Ribas and. Tapia - Estos Anales. .§&, 947 (1932) 4. L. Henry - C. R. .145. 453 (1907) 5. 6 Delby - C. R. 176. 1153 (1923) . Meisenheimer - A. 442. 180-210 (1925) B. Ribas and. Tapia - Estos Anales. 28, 637 (1930) Blaisse - B. 34, 4117 (1901) 7. 8 . 9. 1655 (1921) Ribas and Tapia - Estos Anales. 29_, 778-797 (1931) Grignard - G. R. 141, 44 (1905) Bui. 33, 918 (1905) Bui. 37, 1451 and 1641 (1925) 10. Meisenheimer - B. 54, 1655 (1921) Z.f. Anorg. und Allg, Ch. 147, 331 (1925) 11. Bibas and Tapia - Estos Anales 30, 944-270 (1932) 1 2 . Godchot and Conquil - Compt. rend. 186, 955 (1928) 13. Bedos - Ibid 189, 255 (1929) Tiffeneau - Bull. soc. chim. J3, 1942 (1936) Bartlett and Perry - J.A.C.S. 56, 2683 (1934) Norton and Hass - Ibid 58, 2683 (1936) Gottie and Rowell - Ibid 58, 2267, (1936) 14. P. C. Whitmore - J.A.C.S. 55, 4209-14 (1933) 15. Organic Synthesis - Coll. V o l . I, p. 23 16. Huston and Guile - Doctor^ Thesis; Michigan State College 17. Anderson and Bacbnan - Laboratory Manual of Organic Chemistry p. 72 18. Noller and White - J.A.C.S. 59, 1354-6 (1937) Cope - Tbid 60, 2215-17 (1938) 19. Fisher - Laboratory Manual of Organic Chemistry 4th ed. p. 386 (1938) 20. Pregel - Quanitative Organic Micro Analysis 3rd ed. p. 15-68 (1937) 21. Tredwell and Hall - Analytical Chemistry vol 11 5th ed p. 707 23. Malone and Seid - J.A.C.S. 51, 3426 (1929) Bryant - Ibid 54, 3758 (1932) Brown and Kramers - J.Am. Phaim. Assoc. 1 1 . 607 (1922) Phillips and Keenan - J.A.C.S. 39, 304 (1917) Lyman and Reid - Ibid 42 615 (1920) 24. Bickel and Erench - J.A.C.S. 48, 747 (1926) French and Wirtel - Ibid 48, 1736 (1926) Neuleery and Kansky - Bio. Chem. Z. 20,,446 (1909) 25. Hickenbottom - Reactions of Organic Compounds; p 88 (1938) 26. Robertson - J. Chem. Soc. 115. 1210 (1919) Kamm - Qualitative Organic Chemistry; sec. ed. p. 177 Shriner and Fusion - Identification of Organic Compounds. pl44 (1935) 27. Hickenbottom - Reactions of Organic Compounds - p. 384 (1938) 28. Shriner and Fusion - Identification of Organic Compounds p 144 (1935) Robertson - J. A. C. S. 115. 1210 (1919) 29. Fisher - Laboratory Manual of Organic Chemistry 4th ed. (1938) 30. Whitmore - United States Patent; 2,004,066 (1935) 31. Pregel - Quanitative Organic Micro Analysis 3rd ed. (1937) 32. Schlenk and Schlenk - Ber. 62, 920, (1929) Noller - J.A.C.S. 53, 635 (1931) 33. Lange - Handbook of Chemistry 2nd ed. (1937) 34. Hodgman - Handbook of Chen, and Fhysies 22nd ed. (1937) 3 5 . Kamm - Qualitative Organic Analysis 2nd ed. (1935) 36. Shriner and Fusion - Identification of Organic Compounds (1935) 37. J.A.C.S. 48, 1089-1093 (1926) 38. DeWael and Weakering - Bull. Soc. Chim Belg. 33, 495-504 (1924) 39. Wojeick and Adkins — J.A.C.S. 55, 4939 (l93o) 40. Levene and Marker - J. Biol Chem. 91 77-103 (1931) 41. Kirner - J. A. C. S. 48, 1111-2 (1926) 42. Heirs and Adams — J.A.C.S. 48., 2385—93 (1926)