G MORRIS ON c~-METHYL-I~YDROXTSUCCINIC ACID ETC. II.-On cc-Met7~21l-hydrozysuccinic A.cicl the p-oclzcct of the Actioqz oj’ Anhydrous Hydrocyanic Acid upon Ethyl Aceto-acetate. By GEORGE H. MORRJS. DEAIAR~AY (Cowpt. rend. 82 1337) has described under the name of oxypyro tartaric acid the acid obtained from ethyl aceto-acetate by the action of anhydrous hydrocpanic acid and decomposition of the resulting cyanide by boiling with acids. He states that the acid so obtained is an unstable uncrystallisable syrup and that the barinm salt when boiled with excess of water is decomposed carbonic anhy-dride being evolved and barium carbonate precipitated whilst the barium salt of a different acid viz. hydroxyisobutyric acid remains in solution. H e gives the following equation for this decomposition :-OH H3C*C< C 0 .0 ‘ L l >Ba + OH2= [(CH,),C(OH).CO,],Ba + BaC03+ CO,. ir c- c 0. o From the mode of forma,tion of this acid as shown by the following H,C.CO H,C.C(OH)(CN) equations-+ HCN = I HZC. C 00 CZHG I HZC. COOC,H, an BiORRIS ON a-METHYL-HYDROXYSUCCINIC ACID ETC. 7 it is the next higher homologue of malic acid the salts of which are well-defined and crystalline thus :-HC (OH) (COOH) H,C.C(OH) (COOH) H:,C.COOH I H,C . CO OH I RIalic acid. New acid. It would therefore appear probable that such an acid would be a stable compound yielding well-defined salts. On this account a t the suggestion of Prof. Wislicenus I undertook to go over the work of Demarpy in order to obtain more conclusive evidence of the nature of this acid.Preparation of the Acid. I prepared the acid according to Demarqay's directions by heat-ing a mixture of 2 parts of pure ethyl aceto-acetate and 1 part of anhydrous hydrocyanic acid in sealed tubes for three days at 100". The tubes were then opened and the light brown contents heated on a water-bath for some time to expel the excess of hydrocyanic acid. The cyanide so obtained was then mixed with about twice its weight of rather diluce hydrochloric acid and boiled for some time in a flask connected with an inverted condenser. When the decomposition was complete the liquid was evaporated to drive off the excess of hydrochloric acid and also the alcohol formed in the reaction. The mixture of ammonium chloride and acid remaining after evapora-tion was shaken with dry ether in order to extract the acid and the ethereal solution filtered from ammonium chloride ; this was repeated until the whole of the acid was extracted; and the ethereal solution was then evaporated on a water-bath when the acid' was left as a thick brown syrup.To purify this impure product it was dissolved in water neutralised with ammonium hydrate and basic lead acetate added to precipitate the lead salt. This was collected on a filter tho-roughly washed then suspended in water and decomposed by passing a stream of sulphuretted hydrogen through the liquid. The solution of purified acid was filtered from the lead sulphide formed and con-centrated to a small bulk on a water-bath. This concentrated solu-tion was then shaken with ether several times.The syrupy acid, obtained by evaporating these ethereal solutions was still slightly colcured ; it was allowed to stand for a few days in a vacuum over sulphuric acid when it set to a crystalline mass. The completely dried mass was again dissolved in dry ether filtered from a little insoluble matter and evaporated as before. The pure crystalline acid so ob-tained was dried in a vacuum until it ceased to lose weight and was then analysed. It gave the following numbers : 8 XORRIS ON a-METHYL-HYDROXY SUCOINIC ACID ETC. I. ,1448 gram acid gave on combustion 02160 gram C02 and 11. *1990 gram gave -2975 gram CO, and -1000 gram OH2. ,0745 gram OH,. Found. Calculated for C5Hs05. I. 11. C 60 40.54 40.67 40.77 H 8 . . . . . . 8 5.M 5.72 5.58 0 6 80 54.05 I -- -148 99.99 The acid i3 very deliquescent easily soluble in water alcohol and ether andcrystallises on standing over sulphuric acid or in a vacuum, from its solution in these solvents in sta,r-like groups of needles.It melts at 108" and is decomposed at a higher temperature. Xaltsof the Acid C5EI,05. The fkee acid gives no precipitate with neutral or acid lead acetate solution but is completely precipitated by basic acetate. A solution of a neutral salt gives a white precipitate with basic acetate of lead, and also with nitrate of silver solution but with neutral lead acetate or with barium hydrate it gives no precipitate. Barium Salt.-This salt was prepared by neutralising a strong solu-tion of the acid with barium carbonate evaporating the solution to a small bulk on a water-bath and finally allowing the concentrated solution to stand over sulphuric acid until i t was quite dry.So ob-tained it is a transparent non-crystalline glassy deliquescent mass, and contailis 2 molecules of water which it loses at 150". The analysis of the perfectly dried salt gave the following results :--2128 gram salt lost on drying at 150" C. -0239 gram = 11.23 per cent. Calculated for C5H6Ba0,20H2. Found. OH 11.28 per cent. 11.23 I. *2901 gram of the 150" dried salt gave on combustion -1798 gram 11. *32M gram gave -215.3 gram COz -0640 gram 0.H2 and *2219 111. -2156 gram salt gave ~1466 gram BaC03. CO, -0547 gram OH2 and 1987 gram BaC03. gram BaC03. IT. -5093 77 ,3488 7 MORRIS ON a-METHYL-HYDROXY SUCCISIC ACID ETC.Found. Calculated for C6HBBa05. I. 11. 111. IV. 21.07 21.42 - - C5 60 21-20 H . 6 2.12 2.09 2.09 -B a . . 137 48.41 47.63 47.57 47.73 47-6 0 5 80 28.27 -- - -- -283 100.00 9 In order to test Demarqay's statement that the barium salt was decomposed on boiling with water a solution of the free acid was neutralised with barium hydrate and boiled for several hours. No decomposition however took place the solution remained perfectly clear and no gas was evolved. The solution after boiling was evapo-rated to drpness the salt dried a t 150" until it ceased to lose weight, and analysed. I. -4670 gram salt gave on combustion -2937 gram GOz -0540 gram OHz and *3268 gram BaCOy. 11. -3450 gram gave -2107 gram GOz *0722 gram OHz and .2381 gram BaC0,.111. -5480 gram salt gave a4545 gram BaS04. It gave the following numbers :-Found. Calculated for C5H6Ba0,. I. 11. 111. c 21.20 21-44 20.86 H . . . . . . 2.12 2-23 2.32 -Ba 48-41 48.63 48.00 48.Z 0 . . . . . . 28.27 I - - -The formula for the barium salt of hydroxyisobutyric acid gires the following percentage :-C . . . . . . . . . . . . . . 96 2 7.99 HI4 . . . . . . . . . . . . 14 4-08 Ba . . . . . . . . . . . . . . 137 39.95 0 s 96 27.99 343 100.01 - -The acid obtained by DemarCay probably contained a little unde-composed ethyl aceto-acetate which I found would prevent the acid from crystdllising and would also if the barium hydrate was in excess, cause the precipitation of barium carbonate on boiling the solution of barium salt.The acetate boiled with barium hydrate would saponify, as shown by Wislicenus (Liebig's Awnaleiz 186 166) according to the following equation :-CH,.C0.CIIz.COOC2H6 + Ba(OH) = RaC03 + C,H,OH + C H,. C 0. C H3 10 MORRIS ON a-METHYL-HYDROXTSUCCINIC ACID ETC. CaZcium XnZt.-This is best prepared by neutralising a solution of pure acid with calcium carbonate. It is extremely soluble in water, and is obtained as a deliquescent crystalline mass by long standing of its concentrated solution over sulphuric acid. It was only obtained crystalline by allowing a strong solution to stand for several nioiiths over sulphuric acid. It was prepared by neutralising a solution of the free acid with pure potassium hydrate. Xilver SuZt.- When silver nitrate solution is added t o a solution of the potassium salt a white bulky precipitate is formed which is soluble in hot water.On allowing tlhe hot solution to cool the salt crystallises out in plates which blacken on exposure to light. These crystals when thoroughly dried gave the following numbers on analysis :-I. -1930 gram salt gave on combustion -1145 gram GO2 -0330 gram OH2 and -1130 gram Ag. 11. -1440 gram salt gave -0850 gram C02 -0250 gram OH2 and -0845 gram Ag. PotnssiunL Salt.-This salt is also extremely deliquescent. Found. Calculated for C,I164g20 + +OH, I. I I. C5 60 16.17 H,. . 7 1.88 A g 216 58.22 05+ 88 23.72 -371 99-99 As the salt is decomposed when 16.10 16.12 1.92 1.90 58.68 58.55 heated below loo" the wat,er of crystallisation could not be directly determined.Lead SaZt.-A solution of the free acid or of the neutral salt gives no precipitate with neutral acetate of lead ; but when lead acetate is mixed with a neutral salt and then a few drops of ammonia added or better, by tlhe direct addition of basic acetate solution a dense white preci-pitate is thrown down which is not dissolved when the liquid is boiled, b u t becomes granular. The salt SO obtained consists of a basic salt, and appears to have the formula C5H6Pb05Yb0. I found it impos-sible after repeated experiments to get a pure neutral salt fit for analysis. The experiments always resulted in a more or less basic salt. Copper SaZt.-This was prepared by boiling a solution of the acid with excess of copper carbonate and filtering the solution.This salt is also very deliquescent and on leaving its solution over sulphuric acid it is obtniiied as a transparent blue glassy mass. It was also found impossible to obtain this salt neutral it being always to a greater or less extent basic MORRIS ON a-NETHTL-HYDROXYSUCCINIC ACID ETC. 11 Beduction of Acid with fuming Hydriodic Acid. About 6 grams of the acid were heated with six times its weight of fuming hydriodic acid in a sealed tube to 130-140" for six hours. When cold the tube was opened-there was a moderate pressure in the tube and smell of butyric acid-and the contents shaken with mercury to remove the free iodine which was present in large quan-tity. The nearly colourless liquid was then evaporated to a small bulk it being thought that the product of the reduction would be pyrotartaric acid which is not volatile ; the small quantity of wateiy solution was shaken with pure ether and the ethereal solution separated and evaporated on a water-bath.After evaporation only the merest traces of organic matter remained. The watery liquid from which the ether had been separated was then examined but nothing organic was found in it ; so that the conclusi'on arrived at was that the product had been dissipated in evaporation. The reaction is proba,bly ex-pressed by the following equation ; the pyrotartaric acid first formed, splitting up under the influence of hydriodic acid into either butyric or isobutyric acid and carboQic anhydride :-Either C H3 CH3 I ; CH c< -.-t . . CH CH, I I iH!I ' I .(JiH . . . . 1 iCO()iH + iHj1 = . I + OH2 + co* + 1,. COOH COOH 01' I CH . C 00 .H As the evaporations were performed in a draft cupboard there was no opportunity of hoticing the butyric acid as it was driven off. Dyy Distillation of the Acid. A portion of the acid was placed in a sniall distillation-flask con-nected with a suitable condenser and distilled €rom an oil-bath. The acid mas decomposed a t about 140" ; carbonic anhydride mixed with a little carbonic oxide was given off in abundance and an acid water 12 MORRIS ON a-METHYL-HYDROXY SUCCINIC ACID ETC. liquid distilled over. The oil-bath was kept at 200' until all evolution of gas had ceased when the flask was removed from the bath and heated over a small flame. Nothing came over below 210" but be-tween this and 215" a yellow oil distilled over.When this was all over there remained in the flask nothing but a black coaly residue. I. The Watery DistiZZate.-This was neutralised with potassium car-bonate and re-distilled. It commenced boiling below 70" and the temperature gradually rose until it reached loo" where it remained stationary. When about two-thirds of the liquid had passed over the distillation was stopped and the solution of potassium salt remaining in the flask mas poured into a dish and evaporated t'o dryness. The quantit'y of salt obtained was very small and from its reaction with sulphuric acid and alcohol appeared to be potassium acetate. The distillate was saturated with potassium carbonate which caused a small layer of liquid smelling of methylic alcohol to separate.This was separated dried and distilled ; it began to come over at 68' and fractions were collected between this and 75" 75-80" 90-99". There was however not enough of either fraction to examine. They all smelt more or less alcoholic and were all inflammable. 11. The Yellow 02.-This oil boiling between 210-215O was re-distilled ; it came over between the same limits and from its boiling point and appearance was apparently citraconic anhydride. In order t o prove this it was dissolved in water and converted into salts. At the same time some citraconic acid which Professor Wislicenus kindly gave me mas taken and also converted into salts so that the two might be compared. The following salts were prepared and analysed :-Silver Salt.-Prepared from both by addition of silver nitrate to neutral solution of acid.I n both cases a white precipitate was formed, which dissolved on boiling and crystallised from the mother-liquor in star-like clusters of needles. When dried and analysed the following results were obtained :-I. Salt from acid from distillation:-- 1 l i 3 gram salt gave -0745 gram COZ -0170 gram OH and ~0713Ag. 11. Salt from citraconic acid :--1312 gram gave *0832 CO, -0190 gram OH, and ,0797 gram Ag. Calculated f o r Found. (C5H4Agj04)2 -t OH?. I. 11. C, 120 Hi . . . . . . 10 Ag4 . . . . . . 432 0 144 706 -Lead Salt.-Prepared by 16-99 17.32 17-29 1.41 1.61 1.6'1 61.19 60.78 60.75 20.39 - -99.98 -precipitating a neutral salt with neutra MORRIS ON ~-METHPL-HYDROXk’SUCCINIC ACID ETC.13 lead acetate. with another liquor. results :-The precipitate obtained becomes granular on boiling When collected and dried it gave the following I. Salt from acid from distillation--2262 gram salt gave on combustion -1455 gram COz -0290 gram 11. Salt from citraconic acid-~2380 gram gave *1513 CO, -0288 grhm OH2 and ,1645 gram PbO. Calculated for C5H,Pb04. I. 11. C5 60 17.91 17.54 17.33 H . . . . . . 4 1-13 1.42 1.34 P b . . . . . . 207 61.79 61.84 61.40 OH, and -1507 PbO. Found. - - 0 4 64 19.10 335 99.99 - -The identity of these salts coupled with the boiling point proves the The principal reaction yellow oil to have been citraconic anhydride. wm therefore undoubtedly the following :-CH3 CH, .co’ whilst a second reaction went on which resulted in the formation of carbonic anhydride carbonic oxide acetic acid and some of the lower alcohols probably methylic ethylic and isopropylic. &emarks o n the Constitution of some of the Isomerides gf a-MetlLyl-hy droxys uccinic A c i d . Of the constitutional formulae of the five known acids of the general formula C6Hg05 very little is known. I think however that know-ing the constitution of the foregoing acid we may fix with tolerable certainty the constitution of a t least two of the previously known . - acids. Citramalic acid prepared by Curius (AnnaZen 129 153) by the action of hypochlorous acid upon cit’raconic acid and reduction of the resulting monochlorocitramalic acid with zinc and hydrochloric acid may have either of the two following formule: ON #-METHYL-HYDROXYSUCCINIC ACID ETC.CH3 CH3 I I C.OH.COOH or (11.1 dH.cooH i CH.OH.COOH CH,. COOH It is evident that the first of these formule is identical with the formula of the acid described in the preceding pages and therefore the two acids should exhibit the same properties ; but on the contrary, they are totally different citramslic acid being an uncrystallisable syrup which may be obtained by long standing in a vacuum as an amor-phons transparent mass whilst a-methylhydroxysuccinic acid is easily obtained in well-defined crystals. The salts of the two acids are also entirely different. This would seem to point t o the second of the above formulae as being the correct one f o r citramalic acid.Itamalic acid which Swarts (ZeiCsch. Chem. 1867 646) prepared by the action of hydrobromic acid upon itaconic acid and decomposition of the resulting itamonobromopyrot'artaric acid by boiling with water or alkalis may also have either of two following formule :-CH3 CH2OH 1 I I I (I) C.OH.COOH or (11) CH.COOH C H,. C 0 OH CH,.COOH But the first of these is identical both with the first of citramalic acid and also with that o f ' a-methylhydroxysuccinic acid whilst the acid itself differs from both in its propert,ies forming long deliquescent iieedles which melt a t 60-65" and giving salts different from those of the other two acids. This would again point to the second as being the correct formula. The formulm for these three acids would then be :-CH3 CH,.OH CH3 I CH.COOH I CH.OH 1 COOH I C.OH. C OOH CH2 I COOH 6-methylhydroxysuccinic acid. acid. acid. I CH. GO O H 1 CH, I and I COOH Citramalic acid or Itamalic a-Methyl111 droxysuccini