68 SPENCER AND STOKES: THE DIRECT INTERACTION OFVII.---The Direct Interaction of Aryl Halides andMagnesium.By JAMES FREDERICK SPENCER AND ELEANOR MARGUERITE STOKES.IN the course of a series of reactions involving the use of the Grignardreagent, we obtained evidence which led us to doubt the necessity ofthe presence of ether or any catalyst in the preparation of magnesiumnryl halogen compounds. Preliminary experiments showed that aryliodides and magnesium react, on heating, with the formation ofmagnesium aryl iodides, which in view of the subsequent action ofwater must be constituted according to the type R*Mg*I.The reaction took place with the evolution of a large amount ofheat, and was generally complete in two to tbree minutes; on theaddition of water, after cooling, the parent hydrocarbon was re-generated with the evolution of heat :RMgI + H,O = R*H + Mg(0H)I.The ease with which the combination occurred indicated that thiswas possibly a reaction suitable for removing halogens from cycliccompounds, and so capable of being used as a means of orientation i nthe case of substituted compounds.With the object of testing the suitability of the reaction for suchdeterminations, we have studied the action of magnesium with anumber of aromatic halogen substitution products, and have foundthat i t is possible to remove iodine and bromine from such compoundsalmost quantitatively, and to obtain a large yield of the parentsubstance 8s product of the reactionARYL HALIDES AND MAGNESIUM.69Thus m-broinoaniline gave a yield of 90 per cent.of the theoreticalquantity of aniline, and p - bromophenol gave 40-50 per cent. of thetheoretical yield of phenol,Halogen acids and nitro-compounds containing halogens did notyield the corresponding acids and nitro-hydrocarbons. I n the case ofthe acids, carbon dioxide was evolved, and with the nitro-compounds,nitrogen peroxide was evolved, which immediately reacted with themagnesium, giving rise to so much heat that the compound wascompletely charred, and, indeed, in one experiment the test-tubeme1 ted.Halogen derivatives of naphthalene react in the same way withmagnesium, a-bromonaphthalene yielding 70-80 per cent. of naph-thalene.Similarly, monobromoacenaphthene gave a yield of about 50 percent.of the theoretical quantity of acenaphthene when treated in thesame way.This reaction does not seem to be general for chloro-substitutionproducts ; out of six substances investigated, namely, benzylidenechloride, 0- and p-chlorophenol, a-chloronaphthalene, p-chlorotoluene,and o-chloroaniline, a reaction was found to take place only in thecase of o-chloroaniline; in this instance, a large yield of aniline wasobtained. Iodobenzene arid bromobenzene require special note, forwith these compounds it was found t h t the initial reaction proceededin two directions, as indicated by the equations :(1) C,H,I: + Mg = C,H,*Mg*I.(2 j 2C,H,T + Mg = C,H,*C,H, + MgI,.The products, benzene and diphenyl, were present in quantitieswhich indicated that the reaction represented by (1) had taken placewith about 45 per cent.of the iodobenzene, and the reaction repre-sented by equation (2) with about 55 per cent. of the iodobenzene.The formation of diphenyl was observed by Tissier and Grignard(Compt. rend., 1901, 132, 32) when carrying out the Grignardreaction under ordinary conditions.We have done little up to the present with aliphatic compounds, b u tpreliminary experiments have shown that methyl iodide, methyleneiodide, trimethylene iodide, and isopropyl iodide do not react at allwith magnesium when the two substances are heated together. Mono-bromosuccinic acid, however, does react, and the action commenceswithout initial heating after the substances have been mixed forabout two minutes.The product on treatment with water yieldssuccinic acid.On treating the magnesium aryl compounds with water, weobtained derivatives which may bo used for deciding the position o70 SPENCER AND STOKES: THE DIRECT INTERACTION OFsubstituting groups ; they can, however, also be used for purposes ofsynthesis. For example, magnesium phenyl iodide, prepared by themethod indicated, was ground ir^ a mortar with a little absoluteether and an excess of solid carbon dioxide for about five minutes,the product of this treatment yielded about half the theoreticalquantity of benzoic acid on the addition of dilute hydrochloric acid.I n the absence of ether, the yield of benzoic acid was much reduced.This reaction, effected without the use of a catalyst, indicates thatether is not absolutely necessary for the reaction, and the Grignaidcompounds are not necessarily formed through oxonium compounds ofthe type:but rather, the view put forward by Tschelinzeff (Bey., 1905, 38,3664) is the more correct one, namely, that the addition occurs firstbetween the iodide and the magnesium, and this then forms an additioncompound with ether :It1 + Mg = R*Mg'I.R * M g * I + 3 2 > 0 = C H C2H5>C)<I Mg*R .2 5 2 5EXPERIMENTAL.Interaction of Idobenzene and itfagnasizcm.--Dry iodobenzene(40 grams) was mixed with dry magnesium powder (9 grams) in asmall, hard, round-bottomed flask fitted with an air condenser.Themixture was carefully warmed over a free flame to the boiling point ofthe iodobenzene; after boiling for about a minute, the reactioncommenced, and proceeded without any additional heating.Theproduct was a light grey, homogeneous mass, which was slowlydecomposed by the moisture of tho air, forming benzene. When themass had cooled, cold water was slowly added to it ; this brought abouta decomposition which was accompanied by the evolution of heat. Assoon as the decomposition was complete, the products were distilledwith steam, when benzene, diphenyl, and unchanged iodobenzene werefound in the distillate. The yield of benzene was 44 per cent,, and ofdiphenyl 54 per cent., of that required by theory.The melting point of the diphenyI(7OO) was unchanged after mixturewith an equal weight of pure diphenyl.Interaction of Bromobenzeite and Mccgnesium.--In this case, thereaction did not occur a t all readily, it being necessary to boil themixture of bromobenzene and magnesium for about fifteen minutesbefore combination took place.The organic products and yields werethe same as in the case of iodobenzeneARYL HALlDEY AND MAGNESIUM. 71Interaction of p - Iodotoluene and Magne&m,- p - Iodotoluene(15 grams) and magnesium powder (2 grams) were mixed and gentlywarmed in a hard glass flask fitted with an air condenser. As soon asthe boiling point of the iodotoluene mas reached, the reaction alsocommenced, and proceeded without further heating ; it was, however,not as violent as with iodobenzene. The product was a light greymass, which was treated with water and distilled in steam.Thedistillate was extracted, dried, and fractionated, and shown to consistof toluene. The yield was 87 per cent,, and a small quantity ofunchanged p-iodotoluene was also recovered.Interaction OJ o-Byornotoluene und Magnesium-The interaction ofthese two substances took place in exactly the same way as in the caseof p-iodotoluene. The product was the same, and the yield was equallygood.Intemction of m- Bronzoanilins and Mccgnesium.-Dry m-bromo-aniline (15 grams) was mixed with magnesium powder (4 grams) andheated in the same way as the foregoing mixtures; after two minutes,a most violent reaction took place. A yellow, solid mass wasobtained, which reacted so violently with water that the liquid boiled.The whole mass was then distilled in steam, and the oil which passedover was extracled and fractionated.A yield of 90 per cent. ofaniline was obtained.Interaction of o-Cltlo?*oaniline and Hagnesiuna.-This reaction tookplace extremely readily when heat was employed, and a good yield ofaniline was produced.Interaction of p-Bromaphenol and Magnesium. -p-Bromophenol(8 grams) was mixed with magnesium powder (2 grams) and heated,the reaction commencing suddenly after about two to three minutes’heating. The product, a light grey, solid mass, was treated with waterand distilled in steam, when a yield of 40 to 50 per cent. of phenolwas obtained,When tribromophenol was substituted for phenol, the reaction tookplace with extreme violence and evolution of heat, causing the tube tosoften.The products contained less bromine than the originaltribromophenol, for on the addition of bromine water to the solutiona yellowish-white preoipitate oE tribromophenol was formed, butfurther identification was impossible, nor could the vigour of thereaction be lessened,Interaction of a-Brornonaphthalene and Magne&urn.-a-Bromo-naphthalene (20 grams), mixed with magnesium powder (4 grams), washeated to boiling point, when a vigorous reaction commenced, whichcompleted itself without any further heating, The product was awhite and apparently crystalline mass. When cold, the addition ofwater was attended by great evolution of heat, and the presence o72 FORSTER AND FIERZ: THE THIAZO-GROUP. PART I.naphthalene was at once evident from its odour. The mass was theudistilled in steam, when a 72 per cent. yield of naphthalene wagobtained. The yield mas improved by the use of excess of magnesium ;thus, whilst one atomic proportion of magnesium furnished a yield of44 per cent, of the theoretical, from three times the quantity a yieldof 72 per cent was obtained. The change in the yield is due to thesmaller amount of the bromo-compound escaping reaction.Interaction of Brornoncenaphthene and Mngnesiurn.-About 1 gramof bromoacenaphthene was mixed with excess of magnesium powderand heated over the free flame. The reaction did not commence forabout five minutes, and then proceeded quietly. Water WAS thenadded to the product, but action took place only on warming. Thiswas in all probability due to the fact that the magnesium acenaphthylbromide was protected from its action by being coated with unchangedbromoacenaphthene. The products after treatment with water wereextracted with alcohol, and the acenaphthene formed crystallised out ingood yield.The investigation of this reaction is being continued.CHEMICAL LABORATORY,BEDFORD COLLEGE,BAKEI: STKEET, W