CLOSED CARBOX-CHAINS IN THE AROMATIC SERIES. 21 II.-Spthetical Formation of Closed Carbon-chains in the Aronzatic Series. Part 11. By F. STANLEY KIPPING Ph.D. D.Sc. AMONG the vast number of aromatic hydrocarbons there are many which have been proved to contain two or more benzene nuclei con-densed together that is to say they are compounds in which certain carbon-atoms are common to both rings as for example naphthalene, anthracene and phenanthrene. Other substances are known in which one or more atoms of nitrogen displace one or more atoms of carbon in the closed chain without producing any alteration in the fundamental constitution of the com-pound ; for instance quinoline acridine and pseudoanthroline. Quinoline. Acridine. Pseudoanthroline. Now in examining the formul~ which according to the present theoretical views have been proved both by synthetical and analytical methods to represent the constitution of the compounds mentioned, i t is at once evident that the carbon-atoms which are common to any two rings are situated in the ortho- or 1 .2-position ; up to the present time no compound has been obtained in which two carbon-atoms, occupying the meta- or para-position i n one closed chain enter into the formation of the adjoining one ; although in the present state of our knowledge of the formation of closed chains there is no & priori reason why such compounds should not exist and why for instance, isomeric naphthalenes having the formu1~-I.1 11. 1 could not be formed. In formula I two carbon-atoms in the meta-position would be common t o both rings and in formula I1 this would be the case with two carbon-atoms in the para-position.I t seemed an interesting subject for research to attempt to syntlresis 22 KIPPING SYNTHETICAL FORMATION OF either such hydrocarbons themselves or some derivatives of them in order to obtain experimental evidence as to the possibility of the existence of compounds so constituted. In Part I of this paper (this vol, p. l) W. H. Perkin jun. has described the synthesis of naphthalene from ortho-xylene by a series of reactions which forms an additional proof that condensation of the benzene-rings takes place in the ortho-position ; if isomeric naphtha-lenes or derivatives of such compounds can be produced it seemed probable that they would be obtained in an exactly similar manner when meta- or para-xylene was used in place of the ortho-compound.Acting on the kind suggestion of Perkin I undertook to work out this research and proceeded in the following manner meta-xylene was treated with bromine and the meta-xylylene bromide thus obtained was acted upon in ethereal solution with ethylic chloro-malonate and sodic et hylate when ethylic meta-xylylenedichlorodi-malonate was formed the reaction being expressed by the following equation :-By treating this product with reducing agents the chlorine is dis-placed by hydrogen and ethylic meta-xylylenedimalonat8e formed, thus :-c6H4[ CH,*CC1(COOCzH,)z]z + 2Hz = CsH4[ CH2*CH(COOC,H5),] + 2HC1. On adding sodic ethylate to an ethereal solution of this ethylic salt a disodium-derivative is produced thus :-C6Ha[CH,.CH(COOCzH,)z]2 + 2C2H5.0Na = ~ 6 ~ - [ ~ ~ 2 * ~ ~ a ( ~ ~ ~ ~ 2 ~ > l + 2C2H5*OH.By a,ppropriate treatment vith iodine or bromine it was thought that the sodium would be e'liminated and from this compound a tetrahydrometanaphthalene-derivative obtained according to the fol-lowing equation :--CH,.C (COOC,H,)2 + 2Na1, CH2. CNa (COOCZHJ 2 from which by hydrolysis and removal of two carboxyl-groups tetra-hydroxnetanaphthalenedicarboxylic acid would be obtained CLOSED CARBON-CHAINS I N THE AROMATIC SERIES. 23 A dicarboxylic acid was in fact obtained which was at first thought to be the desired compound ; on careful examination however it was found that it contained no meta-ring but was simply metaphenylene-dipropionic acid--CH,*CH,.COOH /\ It will be seen that this acid differs in composition from the hypo-thetical metanaphthalene-derivative only by two atoms of hydrogen, and that therefore analysis alone would be insufficient to distinguish between them with certainty ; it can be proved beyond doubt how-ever that no closed ring has been formed in the above reactions for when ethylic met.a-xylylenedimalonate is directly hydrolysed without previous treatment with sodic ethylate and iodine (or bromine) the dicarboxylic acid which is ultimately obtained is identical with that which was supposed to be a tetrahydrometanapht#halene-derivative.It is diflicult to explain what takes place when the above-mentioned sodium-derivative is treated with iodine (or bromine) but it would seem that the halogen destroys a portion of the substance with formation of hydriodic (or hydrobromic) acid which then reacts on the remainder reproducing the ethereal salt and the respective halofcl sodium compound.In spite of the failure of these experiments t o produce a meta-closed ring the conclusion was not justified that a para-ring could not be formed since according t o the present theories held regarding the constitution of the numerous compounds containing a pyridine-ring and of the paraquinones a union of two atoms in the para-position either directly or through the interposition of other atoms is assumed. Para-xylene was therefore prepared and submitted to a series of reactions exactly in the same way as in the case of the meta-hydro-carbon ; the sodium compound of ethylic para-xylylenedimalonate was obtained and treated with iodine (or bromine) in the hope of forming a tetrahydroparanaphthalene-derivative.But here again no closed ring could be produced ; when iodine is used the reaction which takes place appears t o be similar to that snggested as the most probable in the case of the meta-cornpound, and the dicarboxylic acid which is the ultimate product is not 24 RIPPING SYNTHETICAL FORMATION OF tetrahydroparanaphthalene-derivative but simply paraphenylenedi-propionic acid-,yCH,*CH,*COOH I I \~-CH,.CH,~COOH This was proved in a manner similar to tbat already described. When bromine is used the reaction is quite different ; substitution takes place and ethylic para-xylylenedibromodimalonate is formed, most probably as expressed by the equation-2C6H4[CH2*CNa(COOC,H,),l + 2Br2 = C6H4[CH2*CBr(COOC,H5>?]2 + ZNaBr + C6H4[ CH,.CNa(COOC2H5),],.From these experiments then the conclusion may be drawn that compounds containing rings condensed together in the meta- or para-position are incapable of existence otherwise they would have been formed by the methods which have been described ; now this may be so for one of two reasons-either on account of the relative positions of the two rings or because of the number of carbon-atoms which would form the reduced ring ; in tetrahydrometanaphthalene as will be seen from the figures already given there would be seven in tetra-hydroparanaphthalene there would be eight carbon-atoms in the larger closed chain.To help to a decision on this point experiments were made with the object of obtaining a derivative of a compound having the formula-which like naphthalene itself would contain six atoms in each ring ; a compound of this character should be obtained by treating meta-xylylene bromide with 2 mols. of sodic ethylate and l mol. of ethylic malonate according to the equation-+ CH2(COOC,H5) + ZC,H,.OKa = ,\CH,Br () CH,Br \/- CH2 + 2C2Hs.0H + ZNaBr, in a manner exactly similar to that in which hpdrindonaphthene CLOSED CARBCN-CHAINS I N TBE AROMATIC SERIES. 25 derivatives were prepared by Perkin (see Part I this vol. p. ’7). But although the experiment was repeated several times under varied conditions no such compound could be obtained ; the reaction was evidently of a very complicated nature and yielded a resinous mass from which no definite product could be isolated.It seems, therefore that the non-formation of R meta- or para-ring in any of these cases is due not entirely to the number G f atoms which would go t’o form the ring but also and perhaps wholly to its relative position. It is a general law in the chemistry of the aromat,ic compounds that no substance can be obtained in which a ring of any sort is joined to the benzene nucleus in other than the ortho-position e.g., the coumarins carbostyrile-derivatives quinoxaline &c. ; the simplest bibasic acid-phthalic-is readily converted into the anhydride, whilst the isomeric meta- and para-anhydrides have not yet been prepared.The two isomeric meta- and para-phenylenedipropionic acids obtained in this research by reactions already described were of considerable theoretical interest owing to the fact that if anhydrides could be obtained from them they would contain a closed ring in the meta- and para-position respectively as shown by the formuh Attempts were therefore made in this direction but without success, and it was thought that perhaps this was owing to the large number of atoms which would have to take part in the formation of the closed ring in the fatty series as is well known many dicarboxylic acids can be converted into anhydrides but whether this is possible or not depends ou the relative positions of the carboxyl-groups that is on how many carbon-atoms go to form the anhydride-ring.To throw some light on this question meta- and para-xylylene cyanides were prepared and from them by hydrolysis the corresponding phenylene-diacetic acids ; it seemed possible that these compounds containing as they do one atom of carbon less in eslch of the side chains would be more capable of forming anhydrides; these would have the € o r m u l 26 KIPPING SYNTHETICAL FORMATION O F No anhydride however could be obtained from either of these acids, and the conclusion drawn from all the experiments briefly described above confirms the general law that no exterior ring in the ineta- or para-position can be formed. CH2Br (1) Meta-xylylene Bromide C6H4 <CK,Br (3). Meta-xylylene bromide was first obtained by Colson (Compt.rend., 99 40) by brominating boiling meta-xylene ; in preparing large quantities of this substance I have found that the following method gives the best results :-50 gmms of pure meta-xyleiie placed in a retort connected with a long reflux condenser are heated in an oil-bath to 125-130" 160 grams of bromine are then added by means oE a dropping funnel inserted into the tubulus of the retort; at first the bromine may be added tolerably quickly but later on when the meta-xylene is partialIy converted and red fumes escape up the condenser, the halogen must be added slowly otherwise loss is incurred. The hydrogen bromide which is eTolved in large quantities during the reaction is absorbed by conducting it into a concentrated solution of caustic soda'. When all the bromine has been added the contents of the retort, which ought to be only slightly coloured are poured out and left 24 hours in a cool place to crystallise ; the mother-liquor is drained off and the crystals after lying for some time on a porous plate are purified by recrystallisation from light petroleum ; meta-xylylene bromide is thus obtained in the pure state as a colourless solid melting a t 77" the melting point given by Colson (Zoc.cit.). From 50 grams of meta-xylene the quantity of pure bromide obtained is not very large the mother-liquor can however be worked up for a less pure product. In consequence of the painful effect of the vapour 011 the eyes it is best when using meta-xylglene to work as much as possible in the open air. Ethyldc Meta-xylyleqisdichlorodinaalonate, CsHa[CHz*CCl( COOC,H,),],.This compound is obtained by t,he action of ethylic chloromalonate and sodic ethylate on meta-xylylene bromide ; 4.4 grams of sodium, dissolved in as small a quantity of alcohol as possible are mixed with about 10 volumes of pure ether and a solution of 37.8 grams of ethylic chloromalonate in 500 C.C. pure ether slowly added ; after well cooling the mixture 25.5 grams of finely -powdered meta-xylylene bromide are thrown in and the whole well shaken. The white pre-cipitate which is a t first formed is the sodium-derivative of ethyli CLOSED CARBON-CHAINS I N THE A4RO?rlATIC SERIES. 27 chloromalonate but as the meta-xylylene bromide dissolves a lively reaction commences and the ether begins to boil the sodium-derivative being decomposed with formation of et hylic meta-xylylene dichlorodimalonate and sodium bromide.The mixture is then heated on a water-bath for a couple of hours great care being taken to shake occasionally otherwise there is considerable danger of the ether boiling over from the bumping which always occurs ; water is subsequently added the ethereal solution washed and separated dried over calcium chloride and the ether distilled off. Ethylic meta-xylylene dichlorodimalonate remains as a thick, yellowish oil which even after long st'anding over sulyhuric acid in a vacuum shows no signs of crystallising ; on analysis the following results were obtained :-I. 0.1615 gram substance gave 01015 gram AgC1. 11. 0.2805 , I 0.1660 , ,, Found. Calculated for -7 C23H2,0&1:.I. 11. C1 14.4 p. c. 15-5 24.6 p. c . The second analysis was made from a different sample the ethereal salt used in (I) having been found to contain a trace of unchanged meta-xylylene bromide. When the calculated quantities of ethylic chloromalonate and nieta-xylylene bromide are carefully weighed the resulting ethylic rneta-xylylene dichlorodimalonate is almost pure and the yield quanti-tative. Eth yZic Meta-xy Zy Zenedimdonufe C6H4[ CH,*CH(COOC,H,),],. When the chlorine-compound described above is dissolved in about 15 volumes of glacial acetic acid and a small quantity of zinc-dust added reduction at once commences the mixture becoming quite warm. The process is best carried out by adding the zinc-dust in very small quantities at a time for about an hour the whole being constantly shaken in a long-necked fiask ; at the end of this time the solution becomes very thick owing to the formation of zinc acetate and reduction proceeds only very slowly ; water is therefore added the mixture warmed on the water-bath and the ethereal salt extracted with ether not only from the solution but also from the undissolved zinc-dust which retains a considerable quantity of the substance; the ethereal solution after washing first with water and then with sodium carbonate solution to get rid of acetic acid is dried over anhydrou 28 KIPPIKG SYNTHETICAL FORMATION OF potassium carbonate and the ether distilled off.The oil which remains still contains a considerable proportion of the unchanged chlorinated derivative the reduction is therefore repeated in exactly the same manner ; it was found however that even after three such treatments a trace of chlorine is still present; the best results are obtained when only small quantities a t a time are operated upon.In order to prepare this substance pure for analysis use is made of the sodium-derivative ; a-hen the calculated quantity of sodium, dissolved in absolute alcohol and mixed with about 10 volumes of ether is added t o a weak ethereal solution of ethylic meta-xylylene dimalonate a white precipitate of the disodium-derivative is formed. This is washed with ether on the filter-pump as quicltly as possible, and then thrown into water or dilute sulphuric acid when it is a t once decomposed yielding ethereal salt and sodium hydrate (or sulphate).The oil which separates is extracted with ether and after drying the ethereal solution with calcium chloride it is treated again, exactly as before with the calculated quantity of sodic ethylate ; the precipitate is washed as above decomposed with dilute sulphuric acid, the solution extracted with ether and the product isolated in the usual manner. Ethylic meta-xylylenedimalonate is thus obtained in the pure state as a t>hick colourless oil which however could not be obtained in the crystalline form ; it is readily soluble in ether alcohol acetic acid, &c. but insoluble in water. The following results were obtained on analysis :-0.1550 gram substance gave 0,3522 gram CO and 0.1016 gram H,O. Calculated for C22H30QY Found.C . . 62.56 per cent. 6197 per cent. H . . . . . . 7.11 , 7-28 ,, 0 . . 30.33 , 30.75 ,, The sodium-derivative of this ethereal salt is a white solid which, however on account of its extremely hygroscopic nature soon decom-poses on exposure to the air. To prepare a sample for analysis a few grams of ethylic meta-xylylenedimalonate are dissolved in a large volume of dry ether and the calculated quantity of sodic ethylate, mixed with about 10 volumes of ether added to the solution. The flask is tightly corked and the mixture after being well shaken is allowed to stand in order that the precipitate may settle; when it has completely subsided i t is washed twice by decantat,ion with pure ether (the air being excluded as much as possible) and after drainin CLOSED CARBON-CHAINS I N THE AROXATIC SERIES.29 off the ether as far as possible quickly transferred to a tared s t,o ppered weighing- bo t tle. After standing in a vacuum over sulphuric acid a sodium determi-nation was made from which the following result was obtained :-0.4055 gram substance gave 0.1215 gram sodic sulphate. Calculated for CzzHzF4NazQ3. Found. Na . . . . . . . . 9.9 per cent. 9.7 per cent. Experiments were now made to obtain a tetrahydrometanaphtha-lene-derivative from the disodium malonate. The following is a description of the methods employed :-The calculated quantity of sodium dissolved in alcohol and mixed with about 10 volumes of ether is added to a weak ethereal solution of ethylic meta-xylylenedimalonate and after allowing the sodium-deriva-tive to separate out which takes but a very few moments a very slight excess of the calculated qiiantity of iodine also dissolved in ether is added and the mixture vigorously shaken.The dark-brown colour of the iodine rapidly disappears and in a few minutes the reaction is completed the slightly brown colour of the solution being due to the small excess of the halogen used; after adding water t o dissolve the sodic iodide and removing the trace of iodine by means of dilute sulphurous acid the colourless ethereal solution is washed with water dried and the ether dislilled off when the ethereal salt of a tetracarboxylic acid remains behind as a thick brown oil. This salt is hydrolysed by boiling it for two hours on the water-bath with a solution of potash in methyl alcohol ; the alcohol is then evaporated and the potassium-compound dissolved in water ; on acidifying this solution with dilute sulphuric acid a small quantity of impurity is pre-cipitated this is filtered off and the filtrate extracted repeatedly with ether.The acid obtained on distilling off the ether is a thick brown oil which as it could not be obtained crystalline was without purify-ing converted into the dicarboxylic acid ; this is effected by heatirig it in a flask in a metal-bath at 180-200" a considerable amount of decomposition taking place with a copious evolution of carbonic anhydride. When no more gas is given off the process is at an end, and on cooling the contents of the flask become solid. To purify the product it is now converted into the ethylic salt ; for t'his purpose it is dissolved in dilute ammonia the excess of the latter removed by standing in a vacuum over sulphuric acid and the solution filtered to get rid of a small quantity of oil which is present; silver nitrat'e is then added to the filtrate and the silver salt thus formed is washed with water ou the filter-pump.After careful drping it is suspeiide 30 KIPPINQ SYNTHETICAL FORJIATION O F in ether treated with an excess of ethyl iodide and boiled on the water-bath with reflux condenser for about two hours whereby the silver salt is completely decomposed. The silver iodide is filtered off and well washed with warm ether the washings added to the filtrate and the ethereal salt obtained by distilling off the ether ; it is however still impure and is therefore submitted to fractional distillation under diminished pressure.A trace of ethyl iodide first comes over the t’hermometer then rapidly rises and the whole distils between 200” and 250” (60 mm. pressure) ; the high boiling portion is again fractioned when almost the whole distils over without decomposition constantly a t 247-250” as a colourless mobile oil. An analysis was now made which however agreed equally well for either of the two formula+-CII,CH(COOC2H,) CH2*C H2 (C 0 0 C,H,) \CH,CH (cooc,~,) “CH2*CH,( COOC,H,j C6H4’ I or CH’ As the acid itself would probably crystallise and conId therefore be more easily obtained in a high degree of purity furtber attempts to purify the ethereal salt were considered unnecessary and it was converted into the acid by boiling with a solution of pure potash in methyl alcohol ; when hydrolysis was complete the solution was evaporated to dryness the potassium salt dissolved in water dilute sulphuric acid added in excess and the organic acid extracted with ether.After isolating the product in the usiial way and recrystal-lising it from boiling water it was obtained in beautiful glistening plates ; an analysis was made which again did not decide whether the ring formation had taken place or not and other means had therefore to be employed to settle this point beyond doubt. Metaphenylenedi-propionic acid was prepared according to the methods described later on and as this substance which must hare the formuIa CsH,(CH2*CH,.COOH)2, is identical with the acid obtained above i t is quite clear that no met,a-ring had been formed by the action of iodine on the sodium-derivative of ethylic meta-xylylenedimnlonate.Another attempt to obtain a tetrahydrometanaphthalene-derivative was therefore made ; in this case bromine was used instead of iodine, and since there was a considerable loss from decomposition on heating the tetracarboxylic acid in the metal-bath another method for eliminat-ing the two carboxyl-groups was adopted. The following is an account of the manner in which the experiment was carried out to a solution of 1.4 grams of sodium dissolved in 15 grams of absolute alcohol and mixed with 50 C.C. of pure ether 12 grams of ethylic meta-xylylene CLOSED CARBON-CICAIKS IN THE AROMATIC SERIES.31 dimalonate in ethereal solution were added and the mixture allowed to stand in a freezing mixture until the sodium compound had com-pletely subsided ; 4.6 grams of bromine were then gradually dropped in the whole being kept well cooled. Each drop of bromine was immediately absorbed the few last oiily causing the solution to assume a slightly yellow colour. The product of this reaction was separated as already described in the experiment with iodine and the ethereal salt hydrolysed by boiling it for two hours with alcoholic potash ; 9 grams of the tetra-carboxylic acid were thus obtained in the form of a brown oil which could not be got to crystallise; it was therefore mixed with about three times its weight of water and heated in a sealed tube first a t l l O o then after opening the capillary to get rid of the carbonic anhydride a t 150° and finally again a t 180" until the evolution of gas entirely ceased.The contents of the tube when cool consisted of two layers; the undermost after standing for some time solidified almost completely, whilst the upper one was filled with colourless crystals. After purifi-cation these crystals were found to be identical with the acid obtained before-viz. metaphenylenedipropionic acid. The solid mass which formed the lower layer was dissolved in boiling water and after filtering allowed to crj-stallise; again the same acid was proved to have been formed and the quantity obtained was from l$-2 grams. About 3 grams of a black resinous mass from which no crystalline product could be isolated remained on the filter ; the mother-liquor from the acid yielded nothing but a further quantity of impure meta-phenylenedipropionic acid which took the form of a slightly yellow oil.The result of these two experiments showed therefore that by acting on the sodium compound of ethyl meta-xylylenedimalonate with bromine 01- iodine no meta-ring is produced and the same compound is ultimately obtained after hydrolysis and elimination of two carb-oxyl-groups whether this reaction is carried out or not ; the possibility that such a ring can exist is thereby rendered improbable. ~ ~ t a - x y ZyZenedimaZonic Acid C6H,[ CH,-CH( COOH)2]2. When the crude ethylic salt of this acid (see p. 27) is boiled with excess of potash in methylic alcohol until no oil is precipitated on addition of water hydrolysis is complete ; this is the case after about two hours.The alcohol is distilled off and the potassium salt which is obtained on evaporation to dryness is dissolved in water; the addition of sulphuric acid in excess produces a slight turbidity the solution i 32 KIPPING SYNTHETICAL FORMATION OF therefore filtered and the product extracted from the filtrate hy shaking out about 10 times with pure ether. When the ehher is distilled off meta-xylylenedimalonic acid remains behind as a thick, slightly brown oil which is readily soluble in water alcohol and ether Unfortunately it could not be obtained in the crystalline form so no analysis of it was made; that its constitution is that represented by the above formula is however proved by its method of formation and by its decompositions also by a comparison with the isomeric para-acid described later on to which it is strictly analogous.iMet apheny Zenedipropio~zic Acid CsH,( C H,* CH,*CO OH) ,. On heating the tetracarboxylic acid just described it is readily convertfed into metaphenylenedipropionic acid with elimination of 2 mols. of carbonic anhydride. This conversion may be effected in two different ways:-First, meta-xylylenedimalonic acid is heated in a small flask in a metal-bath ; as soon as the temperature rises above 100" carbonic anhydride begins to be evolved the evolution becoming more rapid as the temperature rises; after keeping the bath a t about 180" for some time no more carbonic anhydride is given off and the change is com-plete.On cooling the dark-brown oil solidifies to an almost solid cake which is dissolved in hot water boiled with animal charcoal, and the solution filtered ; the new acid crystallises from the filtrate in beautiful colourless plates and is obtained pure by three recrystallisa-tionfi from boiling water. The second method for the preparation of this substance takes up far more time but it is preferable as the yield is far better mek-xylylenedimalonic acid is dissolved in about three times its volume of water and heated in a sealed tube for one hour a t 100-120"; the tube is then allowed to cool and the capillary opened to permit of the escape of the carbonic anhydride ; after re-sealing it is heated again a t 150" for an hour the gas formed allowed to escape and the contents of the tube raised to a temperature of about 180" for Some considerable time.The tetrabasic acid is now completely decomposed, and the liquid which was a t first homogeneous is found after cooling, to have separated into two layers; the bottom one solidifies corn-pletely whilst the top one is filled with leaf-like crystals. To purify the product the whole is collected and the residue after spreading on ;t porous plate to remove traces of oil is dissolvcd in boiling mrater, and the solution filtered. On cooling pure metaphenylenedipropionic acid separates in magnificent lustrous colourless plates. It was drie CLOSED CARBON-CHAINS IN THE AROMATIC SERIES. 33 over sulphuric acid then a t loo" and analysed with the following result :-0.1690 gram substance gave 0.4010 gram CO and 0.0985 gram H,O.Calculated for Cl2H140.1. Found. C 64.86 per cent. 64.71 per cent. H . . . . . . . . 6.31 , 6.47 ,, 0 . . . . . . . . 28-85 , 28.82 ,, Metaphenjlenedipropionic acid when pure melts a t 146-147" ; it is nioderately soluble in hot but almost insoluble in cold water and dissolves easily in ether and alcohol. On precipitating a neutral solution of the ammonium salt with silver nitrate the silver salt falls clown as a white amorphous precipitate; after washing well and drying over sulphuric acid in a vacuum it was analysed with the following result :-0.3318 gram substance gave 0.1638 gram Ag. Calculated for CnH,:Ag@4. Found. Ag 49-54 per cent. 49.37 per cent.This salt is very stable and does not darken when exposed to diffuacd light. In an aqueous neutral solution of the ammonium salt lead acetate gives a white amorphous precipitate copper sulphate a light bluish-green precipitate and zinc sulphate a white crystalline precipitate. Barium chloride gives no reaction. An experiment was made to try and obtain the anhydride of this acid by heating a small quantity for half an hour in a metal-bath a t a temperature of 250". The acid darkened in colonr traces of water were evolved and a slight amount of decomposition took place ; on heating more strongly over the naked flame a brown oil distilled and almost immediately solidified. After washing the product with ether on a porous plate an almost colourless substance was obtained which showed the same melting point as the original acid; no anhydride was therefore formed.ill eth y 1 ic Metapheny lenedipropionate. This ethereal salt is obtained in the usual manner by treating argentic metaphenylenedipropionate with methyl iodide. The pure dry silver salt is mixed with at1 excess of methyl iodide dissolved in 10 volumes of ether a i d the mixture heated on the VOL. LIII. 34 RlPPING SYNTHETICAL FORMATION O F water-bath with reflux condenser for about five hours ; a t the end of this time the silver salt will be entireIy decomposed. To isolate the product the silver iodide is filtered off wnslied with ether and the washings added to the filtrate ; on distilling off tlie ether a thick colourless oil remains behind which in a short time solidifies to a hard cake of crystals ; it is purified by spreading on a porous plate and recrpstdlising from dilute methyl alcohol.Mc4hylic meta-phenylenedipropionate is thus obtained pure in tlie form of colonrless plates and on analysis the following results were obtained :-0.1570 gram substance gave 0,3865 gram CO and 0.1065 gram H,O. Calculated for C14HI,O,. Pound. C 67-2 per cent. 67.14 per cent. H 7.2 , 7.53 7, 0 25.6 , 25-33 ,, This substance is very easily soluble in ether alcohol and benzene, but only sparingly in cold methyl alcohol ; it melts at 51" and when heated in small quantities at a time distils a t a high temperature almost without decomposition. E t lay 1 i c Met ap he 11 y 1 e n e d $3 r opio n at e .This is obtained in a manner exactly similar to that used in the preparation of the methyl salt just described; it is a colourless oil, which boils at 247-250" (60 mm. pressure) without decomposition. The analysis gave the following results :-0,1243 gram substance gave 0.3171 gram CO and 0.0866 gram H,O. Calculated for C 6 ~ ~ 2 2 0 4 * Found. C 69.07 per cent. 69.55 per cent. El 7.91 , 7.74 ,, 0 23.02 , 22.71 ,, Pal-a-zy 7y lene Bromide C6H4 ( C H,Br) 2. This is prepared by brominating para-xylglene in the same way a s described ullder the meta-compound. When the reaction is finished, the product is allowed t o stand for a short time ; it then solidifies and is purified by simply spreading on a porous plate. The yield obtained ranges from 85-90' of the theoretical.The action of the vaponr of this substance on the eyes is extremely painful it is adrisable there-fore to work as much as possible iii the open air CLOSED CARBOY-CHAIXS I N THE RRONATIC SERIES. 35 Ethy 1 ic Para-x y 1 y Ze ned icl~lorod imn Zonat e C5H4 [ CH,C C 1( C 0 0 C2H5) 2 1 2 . 36.8 grams of ethylic chloromaIonate are dissolved in about 500 C.C. of ether the calculated quantity of sodium dissolved in absolute alcohol and mixed with 10 volumes of ether added and then 25 gmrris para-xyljlene bromide thrown in and the mixture vigorously shaken. The sodium-derivative of ethyl chIoromalonate first forms, but on continued shaking the para-xylylene bromide dissqlves and the reaction commences ; no appreciable warming of the mixture how-ever is noticed.After heating on the water-bath with reflnx condenser for about! four hoiirs care being taken on account of the great tendency of the niixture t o bump water is added to dissolve the sodium bromide, and the ethereal solution is then separated ant1 dried. When the ether is distilled off ethyl para-xylylenedichlorodimalonate remains as a thick brownish oil the amount obtained being in accordance with theory. On standing for some houm this oil is converted into a crystalline solid mass ; the adhering mother-liquor is removed by spreading on a porous plate and the product puizified by recrystallisation from alcohol. In this way the ethereal salt is obtained in large colourless six-sided plates ; it melts a t 86-87' is readily soluble in alcohol ether light petroleum acetic acid &c.but is insoluble in water. A chlorine determination gave the following result :-0.3715 gram substance gave 0.2200 gram AgCl. Calculated for W%*CI,O,. Found. (31 . . . . . . 14.4 per cent. 14.6 per cent. Rt lag Zic Para-xy 1 y ZenadirnaZonnte C,H [ CH,*CH ( C 0 0 C,H j 2] 2. To prepare this substance the chlorine-compound just described is reduced with zinc-dust and acetic acid in exactly the same way as the corresponding meta-ethereal salt. About 15 grams are dissolved in 15 C.C. of glacial acetic acid and a small quantity of zinc-dust added ; as the reduction commences the mixture becomes warm After constant shaking for about an hour with frequent addittion of small quantities of zinc-dust wafer is added and the mixture warmed on the water-bath in order to dissolve the zinc acetate ; the whole is then shaken out with ether several times the ethereal solution washed with water and sodium carbonate to get rid of the acetic acid dried over anhydrous potassium carbonate and the ether distilled off.Ethyl para-xylylenedirnalonate remains behind as in the case of the u . 3 6 RIPPING SYNTHETICAL FORJIATIOX OF meta-derivative however a very considerable quantity of the chlorine eompoiind still remains unreduced ; it is necessary therefore to repeat the treatment with zinc-dust and acetic acid and it was found best to work with only small quantities of the substance and submit it to a t least three reductions. It is thus obtained as a slightly yellow oil, which if allowed to remain over sulphuric acid in a vacuum in a few days deposits a quantity of colourless crystals ; these are freed from oil by spreading them on a porous plate and washing with a trace of alcohol, when the substance is obtained quite pure ; after drying in a vacuum, an analysis was made with the following results :-0.2285 gram substance gave 0.5215 gram C02 and 0.1495 gram H20.Calculated for C22HROO,S. Found. C 62.56 per cent. 62.24 per cent. H . . 7.11 , 7-27 ,, 0 . . 30-33 , 30.49 ,, E thylic para-xylylenedimalonate is a fine white crystalline sub-stance which melts at 51" ; it is very readily soluble in ether far less so in alcohol and insoluble in water. This ethereal salt like the corresponding meta-derivative forms a compound with sodium which is obtained as a white precipitate when the calculated quantity of sodic ethylate aiid ethyl para-xylylenedimalonate are mixed together in ethereal solution.The precipitate is washed with pure dry ether, by decantation the same precautions being taken to exclude the air as mere described in the preparation of the meta-compound thrown on to a filter and transferred as quickly as possible to a tared stoppered weighing bottle ; after standing in vacuum over sulphuric acid a sodium estimation was made of which the following was the result :-0.5010 gram substance gave 0.1520 gram Na,SOp. Calculated for C,H,,N 5t208. Found. Nn 9.9 per cent. 9.8 per cent. This sodium-derivative is white and extremely hygroscopic in con-sequence of which it decomposes quickly on exposure to t h e air and immediately on the addition of water or acids the ethereal salt being regenerated.It was thought that if this compound were treated with iodine or bromine a tetrahy dropam-naphthalene-derivative would be formed, as shown by the following equation : CLOSED CARBON-CHAINS 1;N THE AROMATIC SERIES. 37 CH2.CXa (COOC,H,) __ C H2* C ( C 0 0 C2H5) 2 + Bra = (‘1 1 + 2NaBi \L C H2. C (COOC,II,) 2 \L CH2.CNa (COOC,H,), The experiment was carried out thus 12 grams of ethyl para-xylylenedimalonate dissolved in about 50 C.C. of ether were mixed with an ethereal solution of sodic ethylate containing 1.4 grams of sodium and then 5 grams of bromine slowly added. The colour of the bromine disappeared instantly on coming into contact with the sodium-compound the last few drops however causing a yellow coloration.Water was then added to dissolve the sodium bromide, and after separating and drying the ethereal solution the ether WRH distilled OB ; 16 grams of a heavy oil remained which on examina-tion was found to contain bromine. On standing for 24 hours in a cool place ci-jxtals were deposited ; these were partially separated from the mother-liquor by speading on a porous plate and while still on the plate purified by washing with a trace of methyl alcohol in which the oil is readily soluble whilst the crystals are hardly dissolved at all. After crystallising from hot methyl alcohol, 4.5 grams of a fine colourless substance were obtained and oil analysis numbers were found agreeing with the formula-C,H4[CHzCBr(COOCzH,)z]z.I. 0.2370 gram substance gave 0.3990 gram COz and 0.1090 gram H20. 11. 0.29U~ gram substance gave 0.1820 gram bromide of silver. Found. r-A- -7 Calculated. I. 11. H . . . . . . 4.8 , 5.1 -c 45% p. c. 45.9 -Br . . . . . 27.5 , - 26.5 per cent. This substance is therefore ethylic para-xylylenedibromodi-malouate ; it is only sparingly soiuble in cold but readily in hot methyl alcohol from which i t crystallises on cooling in magnificent glittering plates which melt a t 1Oi-108’. In order to prove that it really had the constitution assigned to it above it was treated with zinc-dust and acetic acid when it was reduced to ethyl para-xylylenedi-malonate and far more readily than the corresponding chlorine-compound. On extracting with ether a quantity of a crystalline product was obtained which melted a t 51” and possessed all the properties of ethylic para-xyljlenedimalonate there can therefore be no doubt that this substance is para-xylylenedibromodimalonate 38 RIPPING SYNTHETICAL FOR,lIhTION OF Since bromine acted on the sodium-compound of ethylic para-xylylenediiiialonate in this manner an experinlent was made in exactly the same way using the calculated quantity of iocliiie however, instead of bromine ; the product of the reaction was a thick brown oil which owing to the impurities present could not be got to crystal-lise; on heating a few drops in a test-tube violet vapours were evolved and traces of iodine sublimed.The analysis of the impure substance showed that it contained only from 2 to 2 i per cent.of iodine vhich was present in the combined state so that ethylic para-xylyleaediiododimalonate was either not formed a t all or oiily in very small quantities as this compound would contain 38 per cent. of iodine. The oil was hherefore hydrolysed and converted into the dicarboxylic acid by heating i t i n a sealed tube a t 180" until there was no further evolution of carbonic anhydride ; the product was isolated recrystallised and when pure its melting point was found to bc 223-224" (the melting point of para-phenylenedipropionic acid). In this case also i t is clear that no r i n g formation had taken place. Para-xy Zy Zmedima Zonic Acid C6H* [ C Hz* C H (C 0 OH) J2. When the ethereal salt of this acid which has been already described is boiled for about four hours on a water-bath with excess of potash in methyl alcohol hydrolysis is complete.The solution is theu evaporated to dryness to drive off the alcohol the residual potassium salt is dissolved in water and tho solution after acidifying with snlphuric acid is extracted repeatedly with ether. When the ethereal solution has been dried over calcium chloride and the ether distilled off para-xyljlenedimalonic acid is obtained as a thick syrup which on standing solidifies to a hard cake. This is spread on a porous plate in order to get rid of traces of oil and is then warmed with water and a small quantity of animal charcoal ; the filtered solution when allowed to evaporate over sulphuric acid in a vacuum slowly deposits a finely divided crystalline powder; this is collected and dried first on a porous plate and then in a vacuum over sulphuric acid.0.1982 gram substance gave 0.3905 gram GO and 0.0885 gram HzO. C14H1408. Found. The analysis gave the following results :--Calculated for C 54.2 per cent. 53.7 per cent. H 4.5 , 4.9 7 0 41.3 , 41.4 ,, On heating para-xylylenedimalonic acid melts a t about 195" wit CLOSED CARBON-CHAINS IN THE AROMATIC SERIES. 39 slight> decomposition and evolution of carbonic anhydride ; the whole mass then suddenly becomes solid and on continuing to heat melts again a t 223" (the nieltiug point of para-phenylenedipropionic acid see below). The silver salt was prepared by precipitating a neutral solution of tlie ammonium salt with silver nitrate;.it is a white amorphous substance and on analysis gave the following result :-0.5843 gram substance gave 0.3393 gram silver. Calculated for C14H10h~408- Found. Ag . 58.47 per cent. 58.07 per cent. The neutral solution of the ammonium salt gives with lead acetate, barium chloride and calcium chloride white precipitates of the respective salts. Pwapheizy lenedipropionic Acid C6H ( CH,*CH2*C 0 OH),. Para-xylylenedimalonic acid when heated is easily decomposed yield-ing paraphenylenedipropionic acid and 2 mols. of carbonic anhydride. This decomposition is best effected by h e a h g a solution of the acid in about 3 parts of water at 180" in a sealed tube until the evolution of gas ceases exactly as is described in more detail under the corre-sponding meta-compound.The reaction in this case goes beautifully, and after cooling the tube is found to be filled with hard white nodular masses of almost pure paraphenylenedipropionic acid. To purify the product it is collected and spread on a porous plate to remove traces of oil and is then dissolved in boiling methyl alcohol ; this solution on standing f o r 24 hours deposits the pure acid in curious nodular aggregates which after washing with a little alcohol and drying a t loo" were analysed and the following results obtained :-0.1770 gram substance gave 0.4200 gram CO and 0.1010 gram H20. Calculated for C12H1404. Found. C 64-86 per cent. 64.73 per cent. H . . 6.31 , 6-34! ,, 0 28.83 , 28.93 ,, Paraphenylenedipropionic acid melts at 223-224" ; it is only sparingly soluble in cold methyl or ethyl alcohol and almost insoluble in water.The silver salt is a white apparently amorphous precipitate which does not blacken on exposure to diffused light ; it was obtained in th 40 KlPPING * SYNTHETICAL FORMATION OF usual way by precipitatinq the neutral solution of the ammoniuni s.tlt and after drying a t loo" analysed. 0.4300 gram substance gave on ignition 0.2125 gram silver. Calculated for C12H12dg204. Found. Ag . . . . . . . 49.54 per cent. 49.42 per cent. The following reactions are shown by a neutral solution of tlic ammoiiinm salt :-Copper eulphate gives a faint greenish-white amorphous precipitate sparingly soluble in cold more readily in hot water ; lead acetate throws down the lead salt as a white compound insoluble in water; zinc sulphate and mercuric cliloride also give white precipitates whilst barium chloride and calcium chloride give no precipitate.The following experiment was then made in the hope of obtaioing an anhydride of this acid:-U,3 gram of the pure substance was heated in a test-tube in a metal-bath for half an hour a t 300-320', the acid darkened slightly but no appreciable quantity of water was given off; on raising the temperature considerably and as quickly as possible the whole distilled over in the form of a thick oil which, however a t once solidified on the colder parts of the tube while a very small quantity of carbonaceous matter remained behind. The solid distillate was scraped out of the tube and washed on a porous plate with a small quantity of ether when an almost colourless substance was obtained which melted at 221-224'; it is there-fore clear that no anhydride was formed but that on heating, paraphenylenedipropionic acid (m.p. 223-224") distils uiichanged. Metlt y lic Parapheny lenedipropionate. By boiling the silver salt of the acid suspended in ether with excess of methyl iodide for about four hours on a water-bath methylic paraphenjlenedipropionate is formed. The solution is filtered the residual iodide of silver washed well with warm ether the washings added to the filtrate and the whole distilled when the ethereal salt remains behind as a white crystallirie mass. It was purified by spreading on a porous plate and then recrystnllising from hot methyl alcohol.0.16225 gram substance gave 0.3995 gram C 0 2 and 0.1050 gram H,O. The analysis gave the following results :-Calculated for Cl-IHI*O'i. Found. C . . . . . . . . 67.20 per cent. 67.14 per cent. H . . . . . . i . 2 0 , 7-19 ,) 0 . . 25.60 ) 25-67 ; CLOSED CARBOS-CHAINS IN THE AROMATIC SERIES. 4 1 Methylic paraphenylenedipropionate is only spa]-ingly soluble in cold methyl alcohol and separates out from the hot solution 0x1 cooling in beautiful glittering plates which melt at 115". Paraphenylefiediacr ylic Acid C,H,( CH CH*COOH j2. Ethylic ortho-xylylenedichlorodimalonate is readily decomposed on heating it with alcoholic potash into hydrochloric acid carbonic anhydride and orthophenylenediscrylic acid. It was thought interesting to try a similar reaction with an analogous para-derivative.2 grams of ethylic para-xylylenedibromo-dimnlonate were mixed with a solution of 3 grams of caustic potash in a little alcohol and the mixture boiled on a water-bath for four hours. After addition of water the solution was evaporated to dryness in order to expel the alcohol and the residual potassium salts dissolved in water; on acidifying with dilute sulphuric acid a yellowish flocculent substance probably consisting of impure para-phenylenediacrylic acid was precipitated ; this was not filtered off but the whole extracted with ether the ethereal solution dried and the ether distilled off. The semi-solid mass which remained was mixed with about 10 C.C. of water and heated in a sealed tube for eight hours a t 180° in order to deconipose any tetrabasic acid which might be present.The slightly yellow flocculent snbstance thus obtained was purified by dissolving it in ammonia boiling with animal charcoal and rcprecipitating ; lastly the substance was well washed with water dried a t 100" and analysed when the following results were obtained :-0.2010 gram substance gave 0.4840 gram CO and 0.0892 gram H,O. Calculated for cI,I-rlOo+ Found. C 66.05 per cent. 65.67 per cent. H 4.59 , 4.96 ,, 0 29-36 , 29.37 ,, Paraphenylenediacrylic acid has been already obtained by Loew (Annalen 231 361-384) by the action of sodic acetate and acetic mhydride on terephthalic aldehyde ; the product obtained from ethylic para-xylylenedibromodimalonste is identical with Loew's compound its properties agreeing with those already given by him.JIeta-xylylene Cyanide C6H4( C K2-CN)2. The method used in the preparation of this substance is as follows :-13 grams of meta-xylglene bromide are dissolved in alcoho 42 KIPPINU SYNTHETlCAL FORMATION O F and a slight excess of the calculated quantity of potassium cyanide, in aqueous solution added. No reaction takes place a t first but on warming potassium bromide begins to crystallise out on the sides of the flask the reaction bcing as follows :-C,H,(CH,Br) + 2KCN = CsH4(CH2-CN) + 2KBr. By boiling 011 a water-bath with reflux condenser for a con-siderable time the change is completed the alcohol is then distilled off and the residue treated with water whereby the crude meta-xylylene cyanide is precipitated 8s a brown oil ; after separating it from the aqueous solution by means of a separating funnel it is dried over calcium chloride and fractioned in a vacuum.Thus purified it is obtained as a colourless oil which on cooling almost immediately solidifies to a white crystalline mass ; the analyses gave the following results :-I. 0.2410 gram substance gave 0.6775 gram GO and 0.1120 gram 11. 0.2195 gram substance gave 0.6155 gram CO and 0.1095 gram 111. 0.22425 gram substance ga~7e 34.75 C.C. Nitrogen. Bar. 718 mm. HZO. H,O. T. 17". Found. Calculated for r-h- - -7 ClO&N2. I. 11. 111. C 76-92 p. c. 76.48 76-67 -H . . 5.13 , 5.54 5.16 N . . 17.95 ,, -17.75 p. c. - -Meta-xylylene cyanide is a colonrless crystalline substance melt-ing at 28-29"; i t boils under 300 nim.pressure a t 305-310" a certain amount of decomposition taking place. Ether alcohol and chloroform dissolve it readily ; it is insoluble however in water and light petroleum. n/letaphenyZenediccelic acid G,H,(CH,*COOH),. To obtain this acid the crude oil formed by the action of potassium cyanide on meta-xylylene bromide is without further purification, hydrolysed by boiling it on a water-bath with an excess of alcoholic potash solution whereby it is converted into the potassium salt of rnetaphenylenediacetic acid. Hydrolysis does not take place a t all readily and it is necessary to boil for about six hours before the cessation of the evolution of am-monia shows that the reaction is a t an end. When this is the case, the alcohol is distilled off and the residue dissolved in water; on acidifying with dilute sulphuric acid a small quantity of a brow CLOSED CARBOS-CIIAINS IN THE AROJlhTIC SERIES.43 resinous substance is precipitated and is filtered off; the clear solution is then extracted about 20 times with ether and after drying the ethereal solution over calcium chloride and distiiling off the ether the acid is left in slightly yellowish crystalline crusts. The product is spread on a porous plate and while still on the plate the traces of oil are washed away with a few drops of ether ; after two crystallisations from water it is obtained piire in the form of colourless needles. It was dried a t loo" and analysed with the following results :-0.1697 gram substance gave 0.3830 gram CO and 0.0800 gram H,O.Calculated for ClOHIOO4. Found. C 61.86 per cent. 61.54 per cent. H 5.15 , 5.24 ,, 0 . . 32.99 , 3 - 2 2 ,, Metaphenylenediacetic acid melt,s a t 170" ; it is easily soluble in water alcohol and ether but almost insoluble in light petroleum or chloroform it crystallises from water in beautiful clusters of con-centric needles. The yield obtained by hydrolysing the cyanide with alcoholic potash is almost theoretical. On adding silver nitrate to a neutral aqueous solution of the am-monium salt the silver salt is obtained as a white amorphous pre-cipitate ; after washing well with water and drying first on a porous plate and then at loo" analysis gave the following result :-0.1305 gram substance gave 0,0685 gram Ag.Calculated for G o H*&O,. Found. Ag. 52.8 per cent. 52.5 per cent. I n an aqueous solution of the arnmonium salt lead acetate gives a white amorphous precipitate but with sulphate of zinc a crystalline substance is produced ; barium and calcium chlorides do not give any react ion. No anhydride of the acid was obtained although an experiment was made as follows :-0.2 gram of the pure acid was heated in a metal-bath for 40 minutes a t 300-320" and theu distilled by quickly raising the temperature a small amount of carbonaceous residue was left but almost the whole of the acid passed over as a slightly yellowish oily distillate which condensed on the cooler parts of the tube and immediately solidified; it was purified by washing with a small quantity of ether and a melting point taken, which was found t o be 170-172" ; metaphenylenediacetic acid melt 44 KIPPING SYNTHETICAL FORJIATIOX O F at 170° and i t is therefore clear t,hal no anhydride had been farmed.Para-xylytene Cyanide C,H,(CH2.CN),. 15 grams of para-xglylene bromide are dissolved in alcohol and 8 grams of potassium cyanide dissolved in as little water as possible are added to the solution; no reaction takes place at first but on heating the mixture double decomposition commences a large quantity of a bright-yellow flocculent substance separating ; this is, however not para-xylylene cyanide but a resinous compound formed by some more complicated secondary reaction after boiling on the water-bath with reflux condenser nntil the penetrating odour of the bromide is no longer perceptible the alcohol is distilled off and water added to the residue when the cyanide is precipitated as a dirty-white solid.On shaking with ether it is dissolved while the in-soluble resinous impurities are left ; after drying the ether is evapo-rated and the para-xylylene c-janide is obtained crystallised on the sides of the flask. It may be purified by redissolving i t in ether boiling with a little animal charcoal filtering and allowing the filtrate t o evaporate slowly by this means well-defined crystals are obtained which after recrjstallisation are quite pure. On analysis the following results were obtained :-I. 0.1690 gram substance gave 0.4755 gram CO and 0*0620 gram Bar. 718 mm. H,O. T = 7". 11. 0,12425 gram substance gave 19.7 C.C.Nitrogen. Calculated for r----C,,H$J,. I. 11. Found. - C 76.90 per cent. 76.TS H 5.13 , 5 3 9 -N . . 17.95 , - 18.16 per cent. Para-xylylene cyanide is a colourless substance which crystallises i n long three-sided prisms and melts a t 96" ; in whatever manner the conditions are varied it seems impossible to avoid the formation of a large amount of the yellow resinous substance alIuded to above ; the yield of cyanide is therefore a poor one being only about 50 per cent. of the theoretical. Parap h eny Zenediacetic Acid CsH4 ( C H2* C 0 0 H) ,. When the preceding compound is hydrolysed paraphenylene-cliacetic acid is obtained ; for its preparation 2 grams of the cjanid CLOSED CARBON-CHAINS I N THE AROMATIC SERIES. 4 5 were added to an excess of tbe calculated quantity of potash dissolved in methyl alcohol and boiled on the water-bath with reflux con-denser until no further evolution of ammonia could be observed.After evaporating the alcohol the residue was taken up with water and dilute sulphuric acid until the solution showed a strongly acid reaction ; paraphenylenediacetic acid was thus partially precipitated together with a small quantity of brown impurity; the whole mas shaken out well with ether when the acid dissolved leaving most of the impurities. The ethereal solution was dried and the product obtained as a crystalline crust when the ether was distilled off; the yield was almost theoretical. To purify the acid it was dissolved in alcohol boiled with a little animal charcoal the filtered solution evaporated to dryness on the water-bath and the residue twice re-crystallised from water containing a trace of alcohol.Para-phenylene-diacetic acid is thus obtained in fine colourless needles which are very readily soluble in alcohol far less so in water or ether and melt a t 240-241'. The analysis gave the following results :-0.1730 gram substance gave 0.3920 gram C02 arid 0.08850 gram H,O. Calculated for C,OH,OO,* Found. C 61.86 per cent. 61.80 per cent. H . . 5.15 , 5.46 ,, 0 32-99 , 32.74 ,, The silver salt is precipitated on adding nitrafe of silver to a neutral solution of the ammonium salt as a white amorphous mass ; i t was collected well washed with water and dried first on a porous plate then at loo" and analysed with the following result :-0.2805 gram substance gave 0.1473 gram silver.Calculated for ClOH8Ag20.4. Found. -4g. 52.8 per cent. 52.6 per cent. Paraphenylenediacetic acid is not converted into its anhydride even on heating at 300-320" for half an hour since by quickly raising the temperature a product distils over which after washing with a little ether showed the same melting point as the original acid. NOTE ON THE PREPAR.ATION OF ISOPHTHALIC ACID. Meta-xy Zy ZenediefhyZ E t h e r C6H,(CH,0C2H,),. The preparation of isopht,halic acid by the oxidation of meta-xylene with potassium &chromate a.nd dilute sulphuric acid (Pittig Velguth 46 SYNTHETICAL FORJIATION OF CLOSED CARBOY-CHAISS. Awnden 148 ll) is known to be very tedioiis owing to the extreme stability of this hydrocarbon the greater portion of which remains unchanged 'even after continued boiling for days tozether.The following method was therefore woyked out and found to give excel-lent results mets-xylene is treated with the requisite quantity ( 2 mols.) of bromine a t 125" and the dibromide formed without purification is boiled with alcoholic potash and thus conrerteci into meta-xylrlene diethy1 ether. When this is treated with potassium dichrornate and sulphuric acid a quantitative yieId of isophthalic acid is obtained ; the ether was however first prepared i n the pure state and investigated a little more closely ; for this purpose pure meta-xylylene bromide was boiled for six hours with a solution of alcoholic potash which contained twice the calcuhted quantity of alkali water was then added and the oii which sett,led t o the bottom separated from the aqueous solution dissolved in ether amd allowed to stand over calcium chloride ; after distilling off the ether the product remained as a dark-brown oil which was found still to contain bromine eitker owing to the presence of a trace of nildecomposed nieta-xylylene hromide or of an intermediate compound having the formula C6H,Br*CH2*OC,IC5.It was therefore treated with zinc-dust and acetic acid. and the ether thus freed completely from bromine was isolated in the usual manner and submitted t o fractional distillation, when almost the whole distiiled over between 235" and 255"; this process w a s twice repeated and meta-xylylene diethyl ether obtained in the pure state in the form of a colourless mobile pleasant-smelling oil ; it was analysed with the following results :-0.1392 gram substance gave 0-3805 gram COs and 0.1153 gram H,O.Calculated for C12H1802. Fonnd. C '74.23 per cent. 74.53 per cent. H . . 9.28 , 9.19 7, 0 . . 16-49 , 16.28 ,, This compound boils at 246-247" (712 mm. pressure uncorr.), and does not solidify even when cooled to 0" ; it is isomeric with the ortho-derivative prepared by Leser (Ber. 17 1825) which boils at 247-240" (720 mm. pressure). Of course when this ether is required for the preparation of isophthalic acid it is quite unnecessary to purify it the crude product of the action of alcoholic potash on the xylylene brornide is mixed directly with an excess of potassium dichromate in a large flask and sulphuric acid added ; the reaction is very energetic and great lieat, is evolved so that in working with large quantities i t is advisable to ad2 the acid very slowly or the mixture will be apt to froth over THE ISTERACTION OF ZINC AND SULPHURIC ACID.47 When the oxidation is completed which is the case in a very short time the isophthalic acid seen as a white sandy powder on the bottom of the flask is collected and washed with water ; it is however, very difficult to get rid of the last traces of the green mother-liquor, slid to obtain the product quite pure it is dissolved in sodium car-bonate and the filtered solution acidified with sulphuric acid. After washing the precipitated isophthalic acid and drying it a t loo" an analysis was made with the following result :-0-1477 gram substance gave 0.3128 gram CO and 0.0500 gram H,O. Catciilated for CgHFO.1. Pound. C 57.77 per cent. 57.75 per cent. H . . 3.61 , 3-76 ,, 8 . . . . . . . . 38.62 , 38.49 ,, I n this way an almost theoretical yield of isopht'tialic acid can be obtained from meta-sylene with comparative ease and in a short time, whereas by the old method several days were required to prepare a few grams. This work was carried out in the laboratory of Professor A. v. Raeyer in Munich and before closing I desire to express my gratitude for the kindness which was invariably shown to me and for the interest which he took in this research