摘要:

1.70 FTNDLAY AND THOMAS TNFLUENCE OF COLLOIDS XIX.--lnJluence of Colloids on the Rate of Reactions lnvolving Gases. Part I. Decomposition of Hydroxylamine in the Presence of Colloidal Platinum. By ALEXANDER FINDLAY and WILLIAM THOMAS. IT has been shown by Findlay and King (T. 1913 103 1170; 1914 105 1297) that the rate of escape of gas from a super-saturated solution is influenced in some cases greatly by the presence of colloids. The influence is a specific one and is greatly affected by changes in the nature of the colloid. It was thought that it would be of interest to study the general problem along a different line and to ascertain the influence of colloids on the velocity of a reaction in which a gas is evolved. By such means i t was hoped information might be obtained which would throw light on the way in whioh the colloid acts in retarding or accelerating the evolution of gas from a supersaturated solution.So far as we are aware no investigation along this line bas hitherto been carried out although Reformatsky (Zeitsch. p 1 ~ y ~ i k . d . ChenL., 1891 7 34) studied the velocity of the acid catalysis of methyl acetate in the presence of agar-agar and found it as was t o be expected unaffected by the presence of the colloid. It was our intention to study a series of reactions in which gas is evolved both from hoinogeneous systems (decomposition of diazonium saltls for example) and from heterogeneous systems in the presence of a oatalyst such as colloidal platinum enzymes etc. Unfortunately, the carrying out of our plan was interfered with by the war an ON THE RATF OF REACTTONS JNVOTIVTNC (IASRS.PART I. 171 we have been able to make only a preliminary study of one reac-tion belonging to the second group of reactions mentioned above. As i t may be some time before we are in a position t o resume and complete the proposed investigations it is thought advisable t o make now a short report on the work so far accomplished. The reaction first chosen for investigation was the decomposition of hydroxylamine hydrochloride in alkaline solution in the presence of colloidal platinum a reaction which had been studied to some extent by Tanatar (Zeitsch. physikal. Chem. 1902 40 475). I n alkaline solution hydroxylamine decomposes for the most part, according to the equation 3NH2*OH = NH + N + 3H,O but about five volumes per cent.of the gas evolved is nitrous oxide which is formed in accordance with the equation 4NH2*OH = 2NH3 + N20 + 3H20. In the presence of platinum-black as catalyst the evolved gas consists mainly of nitrous oxide but a certain amount of nitrogen is also present. The colloids or semi-colloids the influence of which on the rate of decomposition of hydroxylamine was studied were dextrin, starch gelatin peptone and ferric hydroxide (see Findlay and King Zoc. c i t . ) . The colloidal platinum sol was prepared by Bredig’s method and the sol concentrated by evaporation in a platinum vessel until i t had a concentration of 15-16 milligrams per 100 C.C. The same concentration of platinum sol was employed throughout a given series of experiments.The temperature a t which the reaction was carried out was 30*6O and the reaction mixture was agitated by a stirrer driven a t a uniform rate and passing through a mercury seal in the stopper of the reaction tube. The gas evolved was collected in a water-jacketed burette filled with a saturated solution of sodium chloride in which nitrous oxide is soluble only to a negligible extent. As a result of preliminary experiments the reaction mixture was chosen as follows 100 C.C. of platinum sol 1 gram of hydroxyl-amine hydrochloride and 10 C.C. ’ of sodium hydroxide solution containing 60 grams of the solid in 100 C.C. of water. The experi-mental conditions were alike in all cases. The hydroxylamine hydrochloride and the added colloid if any were dissolved in the platinum sol and the reaction tube containing this solution was placed in the thermostat.The stirrer was then put in motion, and when a mnstant rate of stirring was attained the alkali solu-tion which had been maintained at the temperature of the bath, was added. After the addition of the alkali two minutes were allowed to elapse while the necessary connexions were being made, and readings of the volume of gas evolved were then made a intervals of one minute. To test the apparatus and method of working a number of experiments were carried out with the above reaction mixture without added colloid. The results obtained were in practically perfect concordance as is shown in table I. (To economise space we have given only the volumes of gas evolved a t intervals of five minutes in two of the experiments carried out.) In tables 11-VI are given in similar form the results obtained when t o the above reaction mixture the various colloids mentioned above were added.I n order that the influence of the addition of colloids may be more readily grasped some of the experimental results are reprkented graphically in the accompanying diagram (p. 174). TABLE I. Ti me (mins. ) . 1 5 10 15 20 26 30 35 40 Volume of gas evolved in C.C. /-I. 11. 0.30 0.30 2.25 2.25 6.50 6-53 12.25 12.22 19.00 19.00 25.25 25.27 30.75 30.73 35.55 35.55 40.33 40.35 TABLE TI. Colloid A d&d Dertrin. Kahlbaum’s purest ‘dextrin was employed. the dextrin was dissolved in the platinum sol.An exact weight of Volume of gas evolved in C.C. Time (mins.). 1 5 10 15 20 25 30 35 40 1 per. cent. solution. 0.35 2.30 5.70 10.10 14.78 19.50 24.30 28.50 31-76 2 per cent. solution. 0.30 2.00 4.20 6-50 9-50 12.50 15.80 18.‘70 21.00 3 per cent. soluti 0x1. 0.30 1.50 3.00 4.75 6-50 8.75 11-50 13-76 16-2 ON THE R~ATE OF REACTIONS INVOLVING GASES. PART I. 173 TABLE 111. Colloid Added Starch. Kahlbaum’s pure soluble starch was employed. Time (mins.). 1 5 10 15 20 25 30 35 40 Volume of gas evolved in C.C. A / \ 1 per cent 2 per cent. 3 per cent. solution. solution. solution. 0.30 0.25 0.20 2.25 2.00 1.75 6-00 5.20 4.25 10.80 9-25 7.70 16.50 14.00 11.70 22.25 19.30 18.20 27.75 2450 21.00 32.20 28.80 25.00 37.60 32-50 28.00 TABLE LV.Colloid Added Gelatin. French gelatin free from salts was employed. Volume of gas evolved in C.C. Time (mins.). 1 5 10 1.5 20 25 30 35 40 1 p& cent. solution. 0.80 4.00 7.30 10.00 12.25 13-80 15.00 16.70 16.25 3 per cent. solution. 0.50 3.00 6.75 8.20 10.20 11.70 12.30 12.70 12.85 - 5 per cent. solution. 0.80 2-50 5-00 6.80 8.20 8-80 9.50 9.78 9.92 TABLE V. Colloid Added Peptone. Solution A.-The peptone was dissolved in the platinum sol as Solution R.-The peptone was dissolved in the alkali and the in the case of the other colloids. platinum sol added. Volume of gaa evolved in C.C./ A \ A . B. Time 0-25 per cent. 0.1 per oent. 0.26 per’cent. (mins. ). solution. solution. solution. 1 0.20 0.10 0.10 5 1.50 0.76 0.60 10 3-20 2.20 1.60 16 4.00 3-80 3.00 20 4-26 6.60 4.20 26 4.60 7-00 6-60 30 Platinum coagulated 8-30 6.110 35 andevolution of gaa 9.00 7.00 40 practioally ceased. 9-80 7.3 174 FINDLAY ANT) THOMAS TXPLTTR’NCF OF COTJ,OTT)S CoUoid Added Ferric Hydroxide. The ferric hydroxide sol was prepared by the method of A. A. Concentration 3-05 Noyes ( J . Amer. Chem. SOC. 1905 37 94). per cent. Time Volume of gas (mins.). in C.C. 1 0.40 5 3.25 10 7-50 15 1350 20 19.70 Tinie Volume of gas (mins. ). in C.C. 25 25*@0 30 no.00 35 33-50 40 36.43 Time in minuti’s. 1. No colloid added.2. 3.05 per cent. ferric hydroxide. 3. 3 per cent starch. 4. 3 per cent. gelatin. 5. 0.25 per cent. peptone (series R ) 6. 0.25 per cent. peptone (series A ) . Although it is not possible a t the present stage of the work adequately to discuss the effeot of colloids on the rate of reactions involving the evolution of gases there are one or two points t o which attention may be ‘directed. On examining the results obtained it will be observed that apar ON THE RATE OF RRACTTONS TNVOLVTNC GASES. PART ‘C. 175 from the initial stages of the reaction the effect of added colloids is in all cases to retard the rate of reaction as measured by the volume of gas evolved in a given time. This influence of colloids may find an explanation in two different directions.I n the first place one will be inclined to assign the behaviour to an interfer-ence by the added colloid with the catalytic aotivity of the colloidal plat,inum. An extreme case of this is found in the experiment with peptone series A where owing t o the inadequate ‘‘ protec-tion” of the platinum by the peptone coagulation of the catalyst occurred and the platinum became practically inactive. From this fact one might conclude that the varying influenoe of different colloids in retarding the reaction investigated is due to a difference in the “protective” power of the added colloids on the platinum sol and to a consequent variation of the extent t o which the particles of the platinum catalyst increase in size or aggregate together. Although no actual coagulation of the platinum was observed except in the oase already mentioned there is of course, considerable room for variation in the size of particles within the limits of the colloidal state.If we take the protective power of colloids towards platinum sols as being represented qualitatively at least by the “gold numbers” of the colloids we should expect that of the three colloids gelatin dextrin and starch gelatin (being the most efficient as a protective colloid) should cause a smaller retardation of the reaction than dextrin and dextrin a smaller retardation than starch. Apart however from the initial stages of the reaction we find the order reversed. The behaviour of peptone it is true would agree with the view that the varying influence of oolloids is due to variation in their protective power, because the peptone employed consisting as it did largely of albumoses has a very feebly protective power and indeed as Zunz (Arch.int. Physiol. 1907 5 111 245) has shown secondary albumose has slight coagulating power. It must however be borne in mind that the protective action of colloids is a specific one in reference to the material protected and we may be wrong in assum-ing the order of protective action to be the same for platinum as for gold but in default of fuller knowledge on this point we must conclude that variation in the size of the platinum partides is probably not the main cause of the retarding influence of colloids on the reaction under investigation. A second view which might be taken is that the retarding influence of colloids is due to a contamination of the surface of the platinum particles by adsorption of the added colloid and to the molecules of the hydroxylamine being thereby prevented from getting acoess to the platinum catalyst.The literature of ‘‘con 176 FINDLAY AND THOMAS INFLUENCE OF COLLOIDS ETC. tact catalysis ” contains many examples of “ poisoning ” which may be explained in this way. Thus Bodenstein (Zeitsch. physikal. Chem. 1903 46 730) found that traces of grease diminished the rate a t which platinum accelerates the union of hydrogen and oxygen to form water (see also Bancroft J . Physical Chem. 1917, 21 734). As the protective power of a colloid is probably closely related to the extent to which it is adsorbed the second explan-ation leads us practioally only to the same point as we reached with the help of the former explanation.I n either case however, the behaviour of ferric hydroxide would seem to be anomalous. The electric charge on colloidal ferric hydroxisde is opposite in sign to that on colloidal platinum and complete adsorption of the platinum by the ferric hydroxide no doubt takes place. Yet ferric hydroxide has the least influence of all the colloids studied on the rate of evolution of gas in the reaction under consideration. It may be that ferrio hydroxide is itself a weak positive catalyst for the reaction but this point has so far not been examined. Whatever may be the importance of the adsorption of colloid by the platinum catalyst as a factor in retarding the catalytic decom-position of hydroxylaniine as measured by the gas evolved we cannot ignore the fact already established by Findlay and King (Zoc.c i t . ) that colloids exercise a very marked andl individual action on the velocity with which gases are evolved from solution under conditions where no explanation on the lines of the sugges-tions already made oan be offered. I n the present case therefore, we believe that this specific action of colloids on reactions involving the evolution of gases must also be a factor the importance of which however it is not yet possible to decide. Although a com-parison of our results with those obtained by Findlay and King seems t o indicate the existence of certain analogies one has to hear in mind that the processes investigated are not quite comparable. The influence of a colloid on the escape of gas from supersaturated solution is doubtless related in some way t o the influence of the colloid on the rate of evolution of a gas from a reaction mixture in which the gas is formed but the connexioii between the two cases is not very clear. We must therefore wait until we are able to investigate the influence of colloids on a reaction which takes place in a homogeneous system and in the absence of a solid catalyst. This investigation will be begun as soon as possible. CHEMISTRY DEPARTMENT, UNIVERSITY OF ABERDEEN. [Received Decembev 28th 1920.

ISSN:0368-1645    

DOI:10.1039/CT9211900170

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

MORGAN AND VINING DIHYDROXYNAPHTHALDEHYDES. 177 XX.-D ih y dr oxy nap h thalde h y des. By GILBERT T. MORGAN and DUDLEY CLOETE VINING. THE ten dihydroxynaphthalenes have been submitted to the hydrogen cyanide synthesis for aldehydes with the result that each of these isomerides has been found to condense with one molecular proportion of the reagent. I n this condensation the dihydroxy-naphthalenes display greater reaotivity than the three dihydroxy-benzenes only one of which namely resorcinol undergoes the Gattermann condensation. 3 4-Dihydroxy-1-naphthaldehyde was obtained from 1 2-dihydr-oxynaphthalene hydrogen cyanide and hydrogen chloride in the presence of anhydrous zinc chloride. There was distinct evidence of steric hindrance which was not overcome by using aluminiuni chloride as catalyst.I n either case a considerable proportion of unohanged dihydroxynaphthalene was recovered. It is noteworthy that in the case of the other o-dihydroxy-derivative namely, 2 3-dihydroxynaphthalene there was no sign of this inhibition, the condensation proceeding to completion with the production of 2 3-dihydroxy-l -naphthaldehyde. 1 3-Dihydroxynaphthalene (naphtharesorcinol) also condenses readily yielding 2 4-dihyd~oxy-l-naphthaldehyde whereas in the case of 1 4-'dihydroxynaphthalene steric hindrance is again mani-fested a large proportion of the dihydroxy-derivative remaining unchanged even after prolonged treatment. Three of the heteronucleal dihydroxynaphthalenes had been condensed in this way by Gattermann (Annalen 1907 357 341), but his observations appear t o be of a'preliminary character and we have been able to confirm the results only in one of these three cases.I n acoordance with the earlier experiments we find that 2 6-dihydroxy-1-naphthaldehyde from 2 6-dihydroxynaphthalene has properties agreeing with those described for the substance by Gattermann (loc. cit.) but on the contrary 2 7-dihydroxy-1-naphthaldehyde from purified 2 7-dihydroxynaphthalene is readily obtained anhydrous and melts a t 159*5-160*5° whereas the product obtained by Gattermann was stated to contain water (1H,O) and to decompose a t 21?-215O. 4 8 - Dihydroxy - I - naphthaldehyde 1 5 - dihydroxy-naphthalene formerly stated to melt a t 195-215O (Zoc. cit.), blackens only a t 280° and does not melt below 300° provided that it is obtained free from 4-hydroxy-1 -naphthaldehyde.This fro I78 MORGAN AND VINING DlHYDItOXYNAPHTHALDEHYl~ES. impurity arises from the a-naphthol often present in commercial 1 5-dihydroxynaphthalene. 4 5-Dihydroxy-l-naphthalde?~yde is the main product of the hydrogen cyanide condensation with 1 8-dihydroxynaphthalene, but there is indication of a small amount of aldehydic by-product, which may possibly contain an ortho-aldehydic group. The two heteronucleal a8-dihydroxynaphthalenes give rise to two pairs of isomeric dihydroxynaphthaldehydes. 1 6-Dihydroxy-naphthalene condenses smoothly t o 4 7-dihydroxy-1-naphth-aldehyde as main product (yield 60-70 per cent.) and to 2:5-di-hydroxy-1-naphthaldehyde ( 1 2-20 per cent.).1 7-Dihydroxy-naphthalene reacts completely giving 4 6-dilzydroxp-1-naphth-aldehyde and 2 8-dihydroqi-1 -naphthaldehyde the former isomeride being produced in somewhat larger proportion. The aldehydes having their aldehydic groups in ortho-positions to hydroxyl are easily obtained crystalline and condense with aniline yielding yellow to orange dihydroxyiiaphthylideneanilines (Schiff bases) excepting the Schiff base from 1 4-dihydroxy-2-iiaphthaldehyde which is red. The p-hydroxyaldehydes of this series are less orystallisable and give rise to red Schiff bases. Both o- and p-aldehydes furnish characteristic p-bromophenylhydr-azones. E X P E R I M E N T A L . General Methods of Preparation. ( I ) l~iJ~ydrosynaphthaldehydes.-Anhydrous zinc chloride (10 grams) freshly prepared by acting on molten zinc with chlorine was added to pure dry ether (50 c.c.) in a closed vessel and left until dissolved the dihydroxyiiaphthalene (10 grams) and anhydrous hydrocyanic acid (8 c.c.) were then added and a rapid stream of 'dry hydrogen chloride was passed into the ethereal solution con-tained in a vessel fitted with inlet tube mechanical stirrer and reflux condenser.A t first the reagents were cooled to Oo but sub-sequently the vessel was allowed to acquire the ordinary tempera-ture of the laboratory. The solution gradually assumed a yellow colour and the intermediate product an aldimine hydrochloride, separated usually as a yellow or brown oil. After removing the supernatant ether and washing the product with this solvent the aldimine hydrochloride was dissolved in cold water (200 c.c.) and the solution warmed when the sparingly soluble dihydroxynaphth-aldehyde separated.(2) p-Bromophenylhydrazones of the Dikydroxynaph t haldehyd es, C,H4Br*NH -N :CH*C,,H,( OH) .-The dihydroxynaphthaldehyde (0-5 gram) dissolved in the minimum amount of alcohol was adde MORGAN AND VlNINB DlIiYDROXYNAPHTHALDEHYDES. 179 slowly with stirring to a solution of p-bromophenylhydrazine (0.5 gram) in 10 per cent. acetic acid (50 c.c.) when the yellow p-bromo-phenylhydrazone was forthwith preoipitated. (3) Bihydroxynaph t hylideneanilines C6H,*N CH C,,H,( OH)2.-Dihydroxynaphthaldehyde (0.5 gram) and aniline (0.25 gram) were warmed on the water-bath in alcoholic solution for fifteen minutes, when the condensation product separated either immediately from the hot solution or on cooling.OH The aldimine hydrochloride separating after three hours as a yellow oil was dissolved in water the solution filtered from a dark brown viscid oil and the filtrate heated to the boiling point when it became cloudy and much paler in colour. 3:4-Dihtydroxy-l-mzphthaldehyde was precipitated as a yellow amorphous solid and was freed from a dark brown flocculent impurity (m. p. 190-ZOOo) by washing with benzene. A further amount was obtained by boiling the foregoing oil with the aqueous filtrate. It crystallised with difficulty but was obtained by slow evaporation from ether-petrolem solution in clusters of minute pale yellow needles, larger crystals having a brown colour.It melted a t 178-180° (Found C = 70.09 ; H = 4.54 per cent.). The benzene washings from the crude naphthaldehyde contained unaltered 1 Z-dihydr-oxynaphthalene ; the yield of 3 4-dihydroxy-1-naphthaldehyde was only 21 per cent. and this was further diminished during crystal-lisation by the formation of a dark resinous product. The p-bromophenylhydrazone crystallising in yellow needles from alcohol with difficulty owing to the formation of brown resinous matter melted a t 137-138O (Found Br =22*72 per cent.), 3 4-Bihydroxynaphthylideneaniline separated in deep red needles with a green reflex; i t melted a t 200-202" (Found: N=5.44 per cent.). Calculated percentages required in the comparison of the fore-going and the following analytical data: Dihydroxynaphthaldehydes CllH8O3 requires C = 70.19 ; H = 4-29 per cent.p Bromophenylhydrazones C,7H,,0,N,Br requires Br = 22.38 per cent. Dihydroxynaphthylideneanilines CI7Hl30,N requires N = 5 -32 per cent 180 MORGAN AND VININC .blHYDROXYNAPHTHALDEHY DXS. OH /\/\ 2 4-Bihydwzy-1 -naphtJmldehyde (,.!,,)OH * CHO The condensation reached completion within two hours the separation of the brown oily aldimine hydrochloride occurring after twenty minutes. This intermediate product dissolved in 200 parts of water to a brown solution from the filtrate of which the crude aldehyde separated on warming. Crystallisation from water or dilute alcohol yielded 2 4-dihydroxy-l-napJ~thuldehyde in pinkish-brown needles ; from alcohol it separated in purplish-brown acicular prisms and from benzene-petroleum in almost colourless needles.The various fractions had the same streak on powdering and melted a t 214O (Found C=69*66; H=4.20 per cent .). It was found difficult to free 2 4-dihydroxy-1-naphth-aldehyde from a brown impurity which was however less soluble than the aldehyde in water. This aldehyde is somewhat more soluble than its isomerides in water aloohol or benzene; the yield of purified compound was 42 per cent. of the theoretical. The p- bromophenylhydrazone obtained with difficulty in crystal-line form separated from dilute alcohol in yellow needles melting a t 165*5-166*5O (Found Br=22*06 per cent.). 2 4-Dihydroxy-1-naphthylideneaniline formed lemon-yellow needles changing to red1 a t 245O and melting a t 252O (Found: N = 5-58 per cent.).OH 1 4-Dihydroxy-2-naphthaldehyde I / \ / \ j C R O \/\/ OH After thirty minutes the ethereal solutioii assumed a pale yellow tint and an almost black viscid oil slowly separated during two hours. Water (200 c.c.) was added to the oil when a brown solution was obtained together with a brown insoluble residue consisting of impure 1 4-dihydroxynaphthalene. From the filtrate heated to boiling a brownish-yellow crystalline aldehyde separated, which when recrystallised from hot dilute alcohol or boiling water, was obtained in silky greenish-yellow felted needles melting a t 188-190O (darkening a t 160-170°) (Found C=69.42; H =4.55 per cent .) . The yield of 1 4-dihydroxy-Z-naphthaldehyde which was only 13.0 per cent.of the calculated amount was not improved by pro MORGAN AND VINING DIHYDROXYNAPHTHALDEHYDES. 181 longing the reaction for five hours this alteration leading to an increase in the quantity of coloured impurity. The p-bromophenylhydrazone a greenish-yellow product crystal-lising from aqueous alcohol or dilute acetic acid in dark green scales giving a greenish-yellow streak decomposed without melting at 214O' (Found Br=22.77 per cent.). This hydrazone differed markedly from its isomerides especially in giving rise to intense bluish-red solutions which became bluer on dihtion. 1 4-Bihydroxy-2-naphthylidenenrtiline crystallised readily from alcohol in dark red feathery needles melting a t 184-185O (Found : N = 4.88 per cent.). CEO /\ /\OH \/\/ 2 3-Dihydroxy-l-nnphthaldehyde 1 I lo=, The aldimine hydrochloride separated after fifteen minutes as a yellow oil which became semi-solid the reaction being complete in ninety minutes.This product dissolved in water (200 c.c.) to a yellow solution containing only a trace of tarry matter. The aldehyde separated! as a bulky yellow crystalline precipitate on heating the filtrate to boiling and crystallised readily from dilute alcohol or from hot benzene in yellow acicular prisms melting at 133'5-134.5O (Found C=70-29; €1=4*42 per cent.). 2 3-Di-hydroxy-1-naphthaldehyde is with the exception of the 4:8-isomeride the least soluble member of the series. It is pro-duced with far greater readiness than the aldehydes from the 1 2-and 1 4-dihydroxynaphthalenes the yield of purified material being upwards of 62 per cent.The p- bromophenylhydrazon e crystallising readily from dilute alcohol in lustrous yellow scales melted and decomposed at 200° (Found Br =22-87 per cent.). 2 3-Dihydroxy-l -mphthylideneaniline crystallised readily from alcohol in lustrous lemon-yellow needles melting at 1 99-200° (Found N = 5-57 per cent.). OH 4 8-Dihydroxy-l-nnphthaldehyde I I 1 /\/\ \/\/ OH CHO The product obtained by Gattermann (loc. c i t . ) from 1 :5-di-hydroxynaphthalene was stated! to melt indefinitely between 195 182 MORGAN AND VTNTNQ T ) I R Y ~ R O X Y N A P H T T € A L ~ ~ T € ~ ~ ~ ~ S , and 215O. We consider that this product was aontaminated with 4-hydroxy-1-naphthaldehyde derived from the a-naphthol present in crude 1 5-dihydroxynaphthalene for in a preparation of aldehyde melting partly between 190" and 200° the impurity was detected in the form of its Schiff base 4-hydroxy-1-naphthylidene-aniline (Gattermann and Horlacher Ber.1899 32 284). 4 8-Dihydroxy-l-naphthaldehyde when obtained in a state of purity by successively washing the crude product with warm 50 per cent. alcohol and crystallising the residue from boiling water, separated in minute yellow needles blackening a t 280° but having no melting point below 300° (Found C=69*94; H=4.20 per cent.). It dissolves readily in ether but in other solvents it is the least soluble member of the series. The p-bromophenylhydrazone crystallising readily from dilute alcohol in lustrous golden-yellow scales melted and decomposed a t 206O (Sound Br = 22.45 per oent.).4 8-Dihydroxy-1-naphthylideneaniline was obtained in dark red needles melting a t 200° whereas Gattermann gave 194-195O as the melting point (Found N=4-98 per cent.). Aldehydes from 1 g-Di~~ydroxynaphthalene, HO OH /\/\ I l l \/\/ CHO Main product (a). HO OH By-product ( b ) . The brown oily intermediate product collected after two hours was treated with water (200 c.c.) when a yellow solution was obtained with a considerable amount of black tarry matter similar to that formed in the alkali fusion of a-naphthol-8-sulphonic acid. The filtrate after heating to boiling slowly deposited the greenish-yellow crude aldehydic material the main bulk (a) of which remained undissolved when extracted in a Soxhlet apparatus with light petroleum ; the yellow extract yielded on evaporation about 0.8 per cent.of a yellow crystalline aldehyde ( b ) . 4 5-Dihydroxy-l-naphthaldehyde the main product ( a ) , separated from dilute aloohol in lemon-yellow nodular crystals darkening a t 150-160° and melting and decomposing a t 164-166O (Found C = 70.70 ; H = 4.59 per cent.). This isomeride, although insoluble in light petroleum dissolved more readily in boiling water benzene ether or alcohol; the yield of purified pro-duct was about 24 per cent. of the theoretical. Prolonged boiling with water or benzene led to a darkening of the product and to MORCAN ANT VTNTNG DTFIYDROXYNAPHTHALTETTY~ES. 183 rise in melting point, probably owing to a condensation between the peri-hydroxyl groups and the aldehydic complex of another molecule giving rise by loss of water to the compound, The presence of a small amount of this condensation product in the specimen analysed would account for the somewhat high result in oarbon.The p-bromopT~enylh?/drazone separating from dilute alcohol in brownish-yellow felted needles melted and decomposed a t 180° (Found Br = 22.66 per cent.). 4 5-Dihydroxy-l-napi~thylideneaniline was obtained in dark red needles with slight green reflex darkening at 200° but not melt-ing below 280° (Found N=4.68 per cent.). The by-product ( b ) sparingly soluble in light petroleum dis-solved more readily in hot water alcohol ether or benzene and melted a t 134-135O. The p- broinophenylhydrazone a lemon-yellow crystalline powder darkening a t 170° and decomposing at 18l0 depressed the melting point of the p-broniophenylhydrazone of aldehyde ( a ) by 1 5 O .The dihydroxynaphthylideneaniline from aldehyde (6) formed reddish golden-yellow needles melting a t 229-230°. Owing to the small amounts obtained i t was not possible t o subject aldehyde ( b ) and its derivatives t o detailed examination but the estimation of bromine and nitrogen in the hydrazone and Schiff base respectively! gave numbers indicating a m on o a1 d e h y d e . 2 6- and 2 7-Dihydroxy-l-naphthn7dehydes, CHO CHO 2 6-Dihydroxy-1 -naphthaldehyde crystallised from benzene in dark yellow aaicular prisms melting sharply at 189-190° (Gatter-mann Zoc. c i t . gives bright yellow needles melting and decom-posing a t 185-190°) ; the p-bromophenylkydrazone separated from alcohol in minute yellow scales melting and decomposing at 194-195O (Found Br= 22-93 per cent.).2 6-Dihydroxy-l-naphthylideneaniline orange needles melted a t 250-260° (Gatter-mann gave orange-red needles with green reflex melting indefinitely at 215-235O) (Found N ~ 5 . 3 6 per cent.). 2 7-Dihydroxy-1-naphthaldehyde was obtained with very littl tarry by-product the yield being upwards of 70 per cent.; it crystallised from hot water or dilute alcohol in pale yellow felted needles melting indefinitely a t 156.5-158.5" or from hot benzene in bright yellow silky needles melting a t 159.5-160.5°; crystals from water after drying between porous tiles for several days gave C- 67.33 ; H =-4.87.C,,H,O,,+H,O requires C = 67-00 ; H =4.60 per cent. C,,H,O, requires C'=70*19; H=4*29 per cent. The aldehyde described as 2 7-dihydroxy-1-naphthaldehyde by Gattermann (loc. c i t . ) was stated to melt and decompose a t 210-215O and t o crystallise with a molecule of water which is lost in a vacuum desiccator over sulphuric acid or more rapidly a t 120O. It will be seen below that these properties are approximately those of one of the isomeric aldehydes (a) derived from 1 6-dihydroxynaphthalene. The p-bromophenylhydrazone of 2 7-dihydroxy-1-naphth-aldehyde crystallised readily from dilute alcohol in shimmering, greenish-yellow scales melting and decomposing at 202-203° (Found Br = 22.65 per cent.). 2 7-Dihydroxy-1-naphthylidene-aniline separated from alcohol in lustrous yellow needles melting a t 214-215O (Found N-4.97 per cent.) (Gattermann gave m.p. From benzene C = 69.72 ; H =4*42. 195-196O). Aldehydes from 1 6-Dihydroxynaphthalene, OH OH (a) Main product. ( b ) By-product. The intermediate aldimine hydroohlorides separated rapidly from the reaction mixture and after one and a-half hours the product was washed with ether and decomposed with hot water when a mixture of isomeric aldehydes was obtained together with a small quantity of brownish-red impurity and a trace of tarry matter. The isomerides were separated by extraction with chloroform when the main product (a) remained insoluble whilst the by-product ( b ) passed into solution. (a) 4 7-Di7tydroxy-l-nn~,lzthnl(Eehyde (yield 64 per cent.) crystallised from alcohol in prismatic silvery needles appearing yellowish-brown by transmitted light and green by reflection ; from boiling water it separated in silvery white needles with a yellowish-brown tint.From both solvents the aldehyde retained lH,O, darkened a t 160° and decomposed a t 218O (Found cT=63.89 MORGAN AND VINING DIHYDROXYNAPJFTHALDEHYDES. 185 = 5.03. C11H803,H20 requires C = 64.05 ; H = 4-89 per cent.). When heated in a vacuum a t 120-130° the foregoing preparations lost water becoming browner and devoid of silvery lustre. Over sulphuric acid in a vacuum desiccator this water of hydration was lost very slowly. The anhydrous aldehyde decomposed a t 218O no darkening occurring below 215O (Found C = 70.03 ; H =4*51.CI1H8O3 requires C = 70.19 ; H =4*29 per cent.). 4 7-Dihydroxy-1-naphthaldehyde which dissolved readily in boiling alcohol was less soluble in hot water and very sparingly so in benzene or chloroform. The p-bromophenyZ~ydrazo?2e crystallised from dilute alcohol in golden-yellow needles melting and decomposing a t 185O ; it retained 1H,O (Found Br = 21.34. C,,Hl,02N2Br,H,0 requires Br = 21.30 per cent. After drying in a vacuum a t 120° Br=21.88. C,,H,,O,N,Br requires Br = 22.38 per cent.). 4 7-Dihydroxy-1 -naphthylideneaniline crystallised from alcohol in orange-red scales with a green reflex; it darkened a t 240° but did not melt below 280° (Found N=5*11 per cent.). ( b ) 2 5-Dihydroxy-1-napktkaldehyde (yield about 21 per cent.) was more soluble than its foregoing isomeride dissolving readily in alcohol or ether and more sparingly in water chloroform or benzene ; it orystallised in bright yellow needles darkening at 190° and decomposing between 225O and 230O.A specimen after drying in a vacuum over sulphuric acid for fourteen days had become anhydrous (Found C = 69.97 ; I$=4'76 per cent.). The p-bromophenyl~ydrazone crystallised from alcohol in yellow needles melting and decomposing a t 206-207O; it retained water (Found Br = 21.38. C,7Hl,0,N,Br,H,0 requires Br = 21.30 per cent. After drying as above Br = 22.84. CI7Hl30p2Br requires Br = 22.38 per cent .). 2 5-Bihydroxy-1-naphthylideneaniline crystallised from alcohol in golden-orange needles which melted a t 209-210°' (Found : N=5*76 per cent.).A Possible Third 1Tsomeride.-The filtrate from crystallised 2 5-dihydroxy-1-naphthaldehyde yielded on evaporation a residue, which after recrystallisation from alcohol #darkened a t 170°, melted at 179-184O and gave with aniline a Schiff base crystal-lising from alcohol in orange-yellow needles melting a t 244-246O ; this anit (Found N=5.01 per cent.) depressed the melting points of the two preceding Schiff bases from 4:7- and 2:5-dihydroxy-naphthaldehydes 186 MORGAN AND VINING DIHYDROXYNAPHTHALDEHYDES. OH CHO OH (a) Main product. (44 per cent.) ( b ) By-product. (38 per cent.) Within twenty minutes the yellowish-brown aldiminc hydro-chloride separated and after one and a-half hours the reaction was completed. On treating the crude intermediate product with water a dark brown oily solution was obtained! containing only a trace of insoluble tar.The filtrate heated to the boiling point, yielded a greenish-brown product which was extracted for several days in a Soxhlet apparatus with dry benzene. From this extract the soluble aldehyde ( b ) separated on cooling whilst the reddish-brown residue consisted of the main produot the less soluble aldehyde (a) contaminated with a small amount of dark red impurity which was left undissolved on extracting the latter aldehyde with hot water. (a) 4 6-Dihydroxy-l-naphthaldehyde crystallised from boiling water in which it is sparingly soluble in yellow microscopic needles decomposing without melting a t 265-270° (Found : C=69-42; H=4*11 per cent.). This isomeride which was readily soluble in ether dissolved more sparingly in hot alcohol and was practically insoluble in benzene.The p-brornoph.enyll~ydrazone crystallising from 'dilute alcohol in greenish-yellow scales with silvery lustre melted and decomposed a t 205-2060 (Found Br =22*47 per cent.). 4 6-Dihyd1*oxy-l-naphthylideneaniline was formed less readily than in other cases; the reagents were heated under reflux in alcoholic solution for six hours and water was added when the pro-duct separated as a dark red crystalline powder decomposing without melting at 230-240° (Found N =5.44 per cent.). ( b ) 2 8-Dihydroxy-l-nuphthaldehyde crystallised readily from alcohol in shining yellow scales melting and decomposing a t 203-204O; it was readily soluble in alcohol or ether and sparingly so in boiling water or benzene (Found C=69.48; H=4*55 per cent .) . The p-bromophenylhydrazone separated from dilute alcohol in minute bright yellow needles melting and decomposing at 206-207O (Found Br = 21.88 per cent .) , 2 8-Dih,ydroxy-l-nap~t~,yEide~eaniline crystallised from alcoho ORTHO-CHLORODINITROTOLUENES. PART 11. 187 in lustrous golden-orange needles darkening a t 240° but not melt-ing below 280° (Found N-5.35 per cent.). We desire to express our grateful thanks to the Salter’s Institute of Industrial Chemistry for the grant of a Fellowship t o one of us (D.C.V.) which has rendered this collaboration possible. THE UNIVERSITY OF BIRMINGHAM. EDOBASTON. [Received January 20th 192 1 .

ISSN:0368-1645    

DOI:10.1039/CT9211900177

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

ORTHO-CHLORODINITROTOLUENES. PART 11. 187 XXI. -or tho- Chl orodinit.rlot oluenes. Part I I . By GILBERT T. MORGAN and LESLIE AMIEL JONES. IN the followiiig communication a new derivation of 2-chloro-4 6-dinitrotoluene is indicated and the reactions of the amino-deriv-atives of the 4 5- and 5 6-dinitro-isomerides are further described. (1) Nitration of 2-Chloro-6-nitrotoluene. Hitherto the only compound definitely isolated from the nitra-tion of 2-chloro-6-nitrotoluene is 2-chloro-5 6-dinitrotoluene (D.R.-P. 107505; Cohn Monatsh. 1901 22 475 and T. 1920, 117 787) but a systematic examination of the product has revealed the existence of 2-chloro-4 6-dinitrotoluene (T. 1920, 117 786). 2-Chloro-6-nitrotoluene (30 grams) obtained by chlorinating o-nitrotoluene was added gradually to 450 grams of concentrated sulphuric acid and 60 grams of nitric acid (D 1-42) a t 20° the temperature being afterwards raised to 70° for one h u r .The product when poured on ice yielded a solid and an oil; the former, on crystallising from alcohol yielded 30 grams of 2-chloro-5 6-dinitrotoluene (m. p. 106O) and an oily residue. The oily nitro-compounds were dissolved in warm concentrated sulphuric acid, and the solution on cooling deposited more of the 5 6-dinitro-compound. The acid filtrate on dilution with water yielded an oil which was dissolved in aloohol and the concentrated solution seeded with a crystal of 2-chloro-4 6-,dinitrotoluene when 1.2 grams of this isomeride separated (m. p. 49O). The filtrate when diluted with water yielded an oil which on repeating the successive treatments with concentrated sulphuric acid and alcohol gave a further crop of 2-chloro-4 6-dinitrotoluene (11) identical with th 188 MORGAN AND JONES: product from 2-chloro-4-nitrotoluene (total yield obtained 5-10 per cent.) ; these methods of preparation confirm oompletely the constitution assigned to the compound.CH3 CH 0% CH N=2 NO2 *H + .____ J (1.) (11.) (111.) (2) L 4 c t i o ~ ~ of Uiazoniunt Salts o n 6-C'ltloro-2 4-tolyleitediajiaine. Uemene-5-azo-6-clloro-2 4-tolyle9Lediamin.e (I).-The preceding dinitro-compound reduced with alcoholic stannous chloride and hydrogen chloride yielded 6-chloro-2 4-tolylenediamine (111) the hydrochloride of which on successive treatment with benzene-diazoniuin ohloride (1 mol.) and sodium acetate gave a brown azo-derivative crystallising from benzene and petroleum iu tufts of stout red prisms melting a t 160° (Found N = 21-09.C,,H,,N,Cl requires N=21.49 per cent.). A small dark red residue insoluble in benzene was found to be the dihydrochloride of the azo-com-pound (Found N = 16.84. Cl,H,,N,C1,2HC1 requires N,= 16.80 per cent.). The diacetyl derivative formed on warming the azo-compound with acetic anhydride crystallised from alcohol in silky, orange needles melting and decomposing a t 267O (Found: N = 16.76. C,,H,,O,N,Cl requires N = 16-26 per cent.). The bisazo-compound CH,*C6C1(NH2),(N,*C,H,), produced by treating the diamine with excess of benzenediazonium chloride, separated froin benzene and petroleum in bright red flakes melting at 192O and crystallising wohanged from acetic anhydride (Found N = 23.69.41 -Nitro 6 en2 ene-5-az o-6-chloro-2 4 - t olyle nediamine (I) separated from acetone in dark red'dish-black prisms with green reflex; it melted and decomposed at 240-245O (Found N = 22.63. C,,H17N,C1 requires N = 23.05 per cent.) ORTHO-CKLORODINITROTOLUENES. PART 11. 189 C,,Hl20,N5Cl requires N = 22.91 per cent.). The diacetyl deriv-ative crgstallised from alcohol in red acicular prisms melting and decomposing a t 290° (Found N = 18-09. C,,H,,O,N,Cl requires N=17.97 per cent.). The foregoing azo-compounds give varying shades of orangered in cold concentrated sulphurio acid. (3) Action of Primary Amines on 4:5- and 5:6-Dinitro-o-cAtorot oluenes.6-Chloro-4-nitro-N-met hyl-m-tohidine (V) .-Dry methylamine was passed into a solution of the 4:5-dinitro-compound (IV) dis-solved in absolute alcohol a t 15O. The liquid which immediately became yellow afterwards deposited deep red acicular prisms the filtrate giving a further crop. The yield was quantitative; the recrystallised product melted a t 127O (Found N = 14.81. C,H,O,N,C1 requires 13.96 per cent.). The orientation of the methylamino-group in 6-chloro-4-nitro-N-methyl-m-toluidine was demonstrated as follows 6-chloro-4-nitro-m-toluidine (VIII) (2 parts) was heated with methyl sulphate (1.5 parts) in toluene solution on the water-bath for about four hours. After distilling off toluene in steam and extracting with ether the product crystallised from alcohol in long red acicular needles (m.p. 126-127*) and was identical with the foregoing preparation. The nitrosoamine (IX) prepared from either specimen crystal-lised in yellow plates melting a t 70° and giving Liebermann’s reaction (Found N = 18.35. C,H,03N,C1 requires N = 18.30 per cent .) . 6-Chloro-3-nitro-N-methyl-o-toluidine (VI) produced by the action of methylamine on 2-chloro-5 6-dinitrotoluene (VII) in alcoholic solution separated forthwith in orange acicular prisms softening a t 83O and melting a t 84-87O. The velocity of reaction was not so great as in the case of the 4:5-dinitro-isomeride and the methylamine solution was less coloured ; the final yield was, however almost quantitative (Found N = 13.90. C,H,O,N,Cl requires N = 13-96 per cent .).The orientation of the methylamino-group was demonstrated by heating 6-chloro-3-nitro-o-toluidine (XI) with 1 part of methyl sulphate and 2 parts of toluene for two hours a t llOo. The pro-duot crystallised from alcohol (yield 70 per cent.) was identical with the preparation obtained by the direct action of methylamine on 2-chloro-5 6-dinitrotoluene. The nitrosoamine (X) from each of these specimens produced VOL. CXIX. 190 MORGAN AND JONES: quantitatively in glacial acetic acid solution crystallised from alcohol in pale yellow prismatic needles melting a t 86-87O and giving Liebermann's reaction (Found N = 18-10. C,H,O,N,Cl requires N= 18.30 per cent.). (V) .-2-Chloro - 4 5 - di-nitrotoluene was mixed with aniline (2.5 mols.) in warm alcohol, and the solution heated to boiling for thirty-six hours.The colour changed from yellow to intense red and on oooling the liquid slowly deposited reddish-orange crystals which separated from petroleum in rectangular pyramids melting a t 95-96O. Owing to the difficulty of separating the product completely from the reagents the yield actually obtained was only 50 per cent. of the calculated amount but no isomeride was detected (Found : N = 10.99. C,,HllO,N,C1 requires N = 10.67 per cent.). The orientmation of the phenylamino-group was confirmed by an application of the Goldberg condensation (Ber. 1907 40 4541). 6-Chloro-4-nitro-m-toluidine (3 grams) was heated under reflux with 15 grams of bromobenzene 1.5 grams of dry potassium carbonate and 0.1 gram of cuprous iodide until on cooling the deep reddish-brown solution gave crystals of the phenylated base.The liquid was then distilled in steam the residue extracted with ether and the product fractionally crystallised from alcohol after boiling with animal charcoal the yield being upwards of 50 per cent. After crystallisation from glacial acetic acimd the compound separated in reddish-orange prisms melting at 93O and was identical with the oondensation product of aniline and 2-chloro-4 5 -dinitrotoluene. The nitrosoarnine prepared from either specimen crystallised from alcohol in yellow prisms melting a t 95-96O (Found: N = 14.41. C13Hl,0,N3C1 requires N = 14.40 per cent.). The filtrate from 6-chloro-4-nitro-N-phenyl-m-toluidine obtained by the Goldberg reaction yielded further quantities of this base and a small amount of a deep red crystalline compound melting a t 128-129O and dissolving in concentrated sulphuric acid to an intense bluish-green solution the colour of which disappeared on dilution with water.6-Chl0ro-3-nitro-N-pkenyl-o-totuidine (VI) was produced by treating 2-chloro-5 6-dinitrotoluene with alooholic aniline as in the foregoing preparation. The reaction proceeded very slowly, and after several weeks a 50 per cent. yield of orange rhomboidal plates melting at 108-109° was obtained (Found N = 10.57. C,,HI,O2N,C1 requires N= 10-67 per cent.). The constitution of 6-chloro-3-nitro-N-phenyl-o-toluidine was confirmed by preparing it from 6-chloro-3-nitro-o-toluidine by the 6-Chloro-4-nitro-N-phenyl-m-toluidin ORT'HO-CHLORODINITROTOLUENES.PART II. 191 Goldberg condensation. 3-Chloro-5-nitro-o-toluidine (4 grams), when heated under reflux for ten hours with bromobenzene (6 grams) in 20 grams of nitrobenzene and in the presence of potassium carbonate (2 grams) and a small amount of cuprous iodide gave a 75 per cent. yield of 6-chloro-3-nitro-N-phenyl-o-toluidine. The solution was distilled in steam to remove nitro-benzene and excess of bromobenzene and the remaining solid ex-tracted with ether; the residue from the ethereal solution crystal-lised from light petroleum in orange-red rhomboidal crystals of the phenylated base melting a t 108-109° and identical with the preceding preparation. The nitrosoamine prepared from either specimen crystallised from alcohol in yellow prisms melting a t 9l0 (Found N = 14.37.6-Chloro-3-phenyl-3 4tolylenediamine (XII) produced by re-duction of 6-chloro-4-nitro-A~-phenyl-m-toluidin~ with zinc dust and ammonium chloride crystallised from aqueous alcohol in pink needles melting a t 109-5O. The diuzoirnin/e (XILI) was formed on adding sodium nitrite to a glacial acetic aoid solution and pre-cipitated on dilution with water as a creamy-white flocculent mass ; i t crystallised from aqueous alcohol in tufts of silky pink needles melting a t 119-120° (Found N= 17.58. Cl,Hl,N3C1 requires N=17*25 per cent.). 6-Chloro-2-methyl-2 3-tolylerrediamine (XV) was prepared by reducing the 5 6-nitroamine with zinc dust and ammonium C,,H,,O,N,Cl requires 14.40 per cent.). (VIII.) CH3 Ic /\GI -+ RNH~ ,) \ (XII.) NH 1 92 ORTHO-CHLORODINITROTOLU~NES.PART 11. chloride in aqueous-alcoholic solution and isolated as the hydro-chloride in fine colourless needles sparingly soluble in water. The free base was obtained as an oil. The diazoirnine (XIV) pro-duced in glacial acetic acid crystallised from absolute alcohol in flesh-soloured prismatic needles melting a t 238-239O. The comparative study of 2-chloro-4 5-dinitrotoluene and 2-chloro-5 6-dinitrotoluene in regard t o the displacement of their labile nitro-groups by ammonia and by primary amines (methyl-amine and aniline) is illustrated by the diagram on p. 191 which also indicates the processes employed in proving the wnstitution of the products. (1) 2-Chloro-4 6-dinitrotoluene recently isolated from the nitration product of 2-chloro-4-nitrotoluene has now been obtained to a similar extent in the nitration of 2-chloro-6-nitrotoluene.(2) 6-Chloro-2 4-tolylenediamine the reduction product of the foregoing dinitro-compound yields azo- and bisazo-compounds of the chrysoidine series comparable with those obtained from its isomeride 6-chloro-3 5-tolylenediamine (T. 1902 81 97). (3) The reaction of ammonia and the primary amines (methyl-amine and aniline) with 2-chloro-4 5-dinitrotoluene leads to the displacement of the 5-nitro-group by the aminio radicle (compare Kenner T. 1914 105 2717; 1920 117 852). I n the case of 2-chloro-5 6-dinitrotoluene the basic radicle displaces the nitro-group in position 6 and not the 5-nitro-group situated in a sympa-thetic position with respect to the chlorine and the other nitro-group. The authors desire to express their thanks to the British Dye-stuffs Corporation Ltd. (Manchester) for facilities afforded in the carrying out of this research. THE UNIVERSITY OF BIRMINGHAM, EDGBASTON. [Received December 28th 1920.

ISSN:0368-1645    

DOI:10.1039/CT9211900187

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

TFJX CIONSTITUTION OF THE DISACCHARIDES. PART v. 193 XXI1.-The Constitution of the Disuccharides. Part V. Cellobiose (Cellose). By WALTER NORMAN HAWORTH and EDMUND LANGLEY HIRST. CELLOBIOSE bears much the same relationship to cellulose that maltose does to starch. It is a disaccharide which like maltose, gives rise on hydrolytic cleavage to two molecular proportions of glucose. Pringsheim submitted cellulose to the action of thermophil and other bacteria and observed the intermediate formation of oellobiose in the degradation to glucose (Zeitsch. physiol. Chem. 1912 78 266). These fermentative processes, investigated from the point of view of carbohydrate metabolism in animals are related to the fermentative degradation of starch, and may acquire a new significance in the light of recent endeavours to utilise cellulose as a source for industrial alcohol.Although considerable study has been devoted to the constitu-tion of cellulose the precise mode of linking of the two glucose residues in cellobiose has hitherto remained obscure. The deter-mination of the structure of this disaccharide may however be regarded as an important preliminary to the investigation of the larger problem. I n an earlier paper by Haworth and Leitch (T., 1919 115 809) the constitution of maltose was shown to be represented by formula I. A constitutional formula (11) was provisionally suggested for cellobiose and in the present com-munication this structural representation is supported by experi-mental proof. CH,*OH rCH-0-CH K!H-o--~H I ~ H - O H LdH dH,.OB h+cm CH,*OH ddH.*H ~ H ~ O H 0 /dH*OH r6H I ~ H ~ O H hfia.oa CH*OH 4 H I ~ H O O H ~ H - O H +H.OH ifH 0 .~ H - O H CH*OII (1.1 (11.1 Maltose. Cellobiose (Cellose). Whilst Franchimont first isolated cellobiose in the form of the octa-acetate i t was left to Skraup to recognise the sugar as a disaccharide and the further work of Skraup and Koenig contributed important data on the preparation and properties of the new biose (Ber. 1901 34 1115; Moaatsh., H* 194 HAWORTH AND HIRST: 1901 22 l o l l ) . An extended study of this new disaccharide by Maquenne and Goodwin (Bull. SOC. chim. 1904 [iii] 31 854) and others furnished an alternative procedure for its preparation. I n the main these processes have been followed during the course of the present work.Repeated trials of these methods have revealed however extraordinary variations in yield. For no apparent reason the ainouiit of cellobiose acetate obtained on acetolysis of cellulose was occasionally very small and theref ore it seemed desirable to attempt to improve the recognised procedure. Eventually a process yielding unif ornily good results was devised, details of which are to be found in the experimental part of this paper. Relying rather on colour changes than on duration of reaction the method now described yields results which are dependable. Cellobiose in the form of its iiionopotassium derivative was methylated by treatment with methyl sulphate and aqueous sodium hydroxide until most of the hydroxyl groups were proteoted and, finally the introduction of methyl groups was completed by the aid of Purdie's reagents namely methyl iodide and silver oxide.Under this exhaustive treatment the cellobiose yielded an octa-methyl derivative in which all the hydroxyl positions were occupied hy methoxyl groups. This new derivative was a colourless crystal-line compound melting a t 76-78O having nD 1.4643 and show-ing [a]= +go in water 9 * l 0 in methyl alcohol and 8 . 8 O in ethyl alcohol. It is herein described as heptamethyl metl~ylcello bioside. The coupling of two hexose residues in a disaccharide molecule promeds by the linking of one hydroxyl group in each of the com-ponent hexoses. Ten hydroxyl groups then are available (potenti-ally) two of which are engaged in the above-mentioned union, leaving eight such groups exposed in the disaccharide.In hepta-methyl methylcellobioside these eight pcsitions are protected by methyl groups. One of the eight methyl groups may be eliminated by hydrolysis with dilute acids namely that protecting the par-ticularly labile reducing group of the biose. Hydrolytic cleavage of the disaccharide linking releases two more hydroxyl groups, namely those participating in the union of the two hexose residues. The assignment of positions to these iiewly exposed hydroxyl groups by investigating the methylated hexoses resulting from the scission of the methylated biose provides a solution t o the problem of the constitution of the disacoharide. Heptamethyl methylcellobioside was hydrolysed by digestion for three to five hours with 5 per cent.hydrochloric acid a t SOo. As cleavage products two substances were isolated which were easily separable by extraction with different solvents. 1,ight petroleu THE CONSTITUTION OF THE DISACCHARIDES. PART v. 195 dissolved one component of the mixture and from this solution, tetramethyl glucose of the butylene-oxidic type crystallised in the usual characteristio form and was identified by analysis by specific refraction and rotatory power and by mixed melting-point deter-mination with an authentic specimen. The second component of the mixture was soluble in dry ether and crystallised from this medium in small colourless needles melting a t 115-116°. This compound gave analytical figures corresponding with a trimethyl hexose and showed [nID -4 68.7O in equilibrium in methyl alcohol.These properties are similar t o those of the trimethyl glucose previously isolated by Haworth and Leitch from methylated lactose (T. 1918 113 188) and by Denham and Woodhouse (T. 1914, 105 2364) from methylated cellulose and the constitution is dis-oussed in the former publication. A melting-point determination of this specimen of trimethyl hexose when mixed with speciinem derived from lactose and cellulose established its identity. The products of cleavage of methylated cellobiose were theref ore proved to be butylene-oxidic forms of trimethyl glucose (111) and tetra-methyl glucose (IV) ; consequently the structure of heptamethyl methylcellobioside is indicated by the formula V. He tsmethyl methyfcellobioaide.Tetramethyl Trime thy1 glucose. glucose. The trimethyl glucose was stable in the presence of perman-ganate and additional evidence in support of the constitution previously assigned t o this substance was forthcoming from a study of its behaviour on complete methylation. It must here be recorded however that an embarrassing difficulty arose during the attempt t o introduce t,wo additional methyl groups. One o l the hydroxyl groups displayed a tendency t o resist methylation t o a surprising degree but by repeated application of the methyl-ating agents this hindrance was overcome. Finally 92 per cent, of the original amount of the trimethyl glucose was converted into crystalline tetramethyl-P-methylglucoside and this was identical in every respect with a specimen previously prepared by Purdi 196 HAWOKTH AND HIRST: and Irvine (T.1904 86 1060). Moreover this glucoside gave rise on hydrolysis to the butylene-oxidic form of tetramethyl glucose melting a t 84O and identical in all other respects with an authentic specimen of this compound. Inasmuch as no evidence is available for the assumption that the oxide linking in a sugar is disturbed by the application of the conditions under which the methylation now described was conducted namely with the Purdie reagents the conclusion is drawn that the trimethyl glucose, isolated already from three sources (a) methylated cellobiose, ( 6 ) methylated cellulose (c) methylated lactose possesses the butylene-oxidic structure as shown in the formula (111). Emulsin and cellase hydrolyse cellobiose yielding glucose whilst maltase is without effect.It is not definitely known whether the first of these enzymes owes this property to the presence of traces of the second which is recognised to be specific in its relation to cellobiose. On other grounds however namely the striking similarity in optical properties of derivatives of lactose and cello-biose (Hudson J. 14mer. Chem. Soc. 1916 38 1566) it is to be regarded as highly probable that the linking of the two hexoses in each is similar structurally and stereochemically . For these reasons cellobiose may be considered as glucose-fI-glucoside. Doubtless i t is premature to speculate as to the precise form in which the hexoses (or pentoses) must be represented as comprising the cellulose molecule.Recently various authors have combated the opinion so generally held that hexoses as such are grouped together in cellulose as they are in the di- and tri-saccharides. The cumulative evidence however points definitely to a structural linking for part of the glucose constituent. This is the glucose fragment linked with other residues through both the reducing group and that hydroxyl group attached to the fifth carbon atom from the reducing end of the hexose chain: 0 CH~CH(OH)~CH(OH)~H~CH~CH~-OH I 0 I 0 I This mode of linking is probably largely represented' in the struc-ture of the cellulose molecule. Inasmuch as trimethyl glucose of the constitution represented by formula I11 has been isolated both from methylated cellulose and cellobiose i t seems clear that the above residue forms an essential part of cellulose.A t variance with this view is the opinion expressed by Sarasin (Arch. Sci. phys. nat. 1918 [iv] 46 5 ) who suggests that cellu-lose is composed of polymerised molecules of I-glucosan THE OONSTITUTION OF THE DISACCHARIDES. PART v. 197 EX P E R I M E N T A L . Preparation of Cello biose Octu-acetate. The following is a convenient and rapid procedure for the pre-paration of cellobiose and although based on the methods described by Skraup and Koenig and by Maquenne and Goodwin (Zoc. c i t . ) differs in important respects from the processes out-lined by previous workers. Filter paper in layers of four sheets was kept in a dry atmosphere a t 20° for three days and was then out into pieces of about 1 sq.cm. in size. Twenty grams of this material were stirred into a water-cooled mixture of 80 C.C. of commercial acetic anhydride (85-95 per cent.) containing 11 C.C. of concentrated sulphuric acid. A t this stage the temperature of the viscid mixture was kept just below 20°. When moist filter paper had been used this treatment produced a marked develop-ment of heat whilst paper too well dried failed to become impreg-nated with the reagents. Under the prescribed conditions a viscid paste was formed after stirring the paper with the reagents for five minutes. A bath containing calcium chloride and water having been heated in readiness to a temperature of 120° the vessel containing the viscid paste of impregnated' filter paper was heated by this means and the contents thoroughly stirred.Rapid disintegration and solution of the paper occurred the mixture darkened in colour, and a t about 112" assumed the form of a dark red mobile liquid', which began to boil. This marked the critical stage of the acetolysis process. Immediately the red solution appeared to be changing to black the whole of it was poured into 18 litres of cold water. A pale yellow preoipitate of crude cellobiose octa-acetate separated after ten minutes. It was possible to form an immediate judgment on the success of the experiment in the follow-ing way a sample of the precipitate should dissolve in boiling alcohol but not in the cold; from the hot alcoholic solution white, powdery crystals should be deposited on cooling; under the cover-glass of a microscope slide a drop of the hot solution should give characteristic rosettes of needles affording a rapid and compara-tively trustworthy test for t,he presenoe of cellobiose octa-acetate.The pale yellow precipitate was kept in contact with water for six hours filtered washed with water and dried a t 40°. It was recrystallised by boiling under reflux for half an hour with 300 C.C. of 90 per cent. ethyl alcohol. While still hot the solution was filtered and on cooling there separated colourless minute needles of cellobiose octa-acetate which after twelve hours were collected 198 HAWORTH AND HIRST: washed with dilute alcohol and dried. The melting point was 224-227O but the product may be further purified from alcohol or ethyl acetate.Equally satisfactory results were obtained by using this method of preparation with three times the above quantities. The product weighed 25 t o 35 per cent of the original weight of filter paper and this yield was maintained throughout an extended series of preparations. Removal of Acetyl Groztys.-In this operation 10 grams of cellobiose octa-acetate were moistened with a little absolute alcohol and mixed during constant stirring with a solution of 12 grams of potassium hydroxide in 50 C.C. of absolute alcohol. After keep-ing for two hours the solid potassium cellobiosate which had separated was collected 011 a filter washed with aloohol and dried in a vacuum. The yield was 6 grams but this product contained traces of free alkali. The free disaccharide was isolated from the potassium derivative by the addition of perchloric acid as described by IIaquenne (Zoc.c i t . ) . MetJaylntioia of Cellobiose Isoltction of Heptarizethyl Met hy 1 cello b ios ide . The cellobiose octa-acetate used in the murse of this work melted at 224-227O (uncorr.) showed [aID + 4 1 * 5 O in chloroform and gave by the ebullioscopic method a molecular weight of 685 in chloroform as solvent; the theoretical value is 678. As the methylation was conducted in alkaline solution it appeared unnecessary to isolate the free disaccharide from the potassium derivative obtained by removal of the acetyl groups. Accordingly the potassium cellobiosate (1 1 grams) was dissolved in a little water and subjected to methylation with 38 C.C.of methyl sulphate and 36 grams of sodium hydroxide dissolved in 70 C.C. of water. The procedure was similar to that followed in the case of lactose (Haworth and Leitch Zoc. cit.) and the product consisted essent'ially of a liquid distilling slightly below 200°/0*2 mm. and showing nD 1.4687 whilst the analytical figures were in agreement for hexamethyl rnethylcel7obioside (Found : OMe=48'2. C,,H,O, requires OMe=-49-3 per cent.). I n all 72 grams of oellobiose octa-acetate were deacetylated and passed through the methylation treatment but in order to intro-duce the full complement of eight methyl groups the hexamethyl methylcellobioside was digested on three separate occasions with methyl iodide and silver oxide. On fractional distillation of this product only a minute quantity distilled below 190°/0'25 mm., almost the whole of i t being collected at 190-200°/0~02 mm.as THE CONSTITUTIOK OF THE DISACCHARIDES. PART V. 199 faintly yellow syrup having ?zD 1.4620 This syrup rapidly became completely crystalline and after triturating with light petroleum, the colourless crystals melted a t 76-78O showing n 1.4643 for the superfused solid (Found C = 5.2.7 ; H = 8.37 ; OMe = 53.2. C,,H,O, [C,,H,,O,(OMe),] requires C = 52.86 ; H = 8.37 ; OMe = 54.6 per cent.). The analyses and properties were in agreement with those required for heptamet hyl methylcellobioside. Polari-metric observations were recorded as follows : Solvent. c . c.1 D. Water .................. 1.560 + 9.0" Methyl alcohol ...... 1.629 9.1 Ethyl alcohol......... 1.421 8.8 Acetone ............... 1.803 9.9 klydrolysis of H eptamethyl Methylcellobioside. A preliminary experiment indicated that complete hydrolysis of the fully rnethylated cellobiose was effected after heating with 5 per cent. hydrochloric acid for five hours a t 80-95O Temperatiwe. Time. [a],,. 80" 0 + 9.0" 80 l b minutes 9.0 80 3 hours 48.2 95 5 9 9 76.S 9 5 9 9 ) 76-8 Consequently 10 grams of heptaniethyl methylcellobioside were submitted to a parallel treatment with 500 C.C. of 5 per cent. hydrochloric acid. Thereafter the mineral acid was neutralised with barium carbonate filtered and the solution evaporated t o dryness under diminished pressure. The residue of syrup and salts was extracted several times with boiling ether and yielded 7.7 grams of a dear yellow syrup.This partly crystallised on nucleation with a specimen of butylene-oxidic tetramethyl glucose. Separation of the constituents of the syrup revealed the presence of about equal amounts of tetramethyl glucose and of a trimethyl glucose. The former was isolated by repeated digestion with boiling light petroleum from which on cooling the character-istic crystals of the compound were deposited. These melted a t 88-89O showed [a], +83-3O in equilibrium in water and the mixed melting-point determination and other properties were also in full agreement with those recorded for butylene-oxidic tetra-methyl glucose (Found OMe= 53.0. Calc. OMe=52*4 per cent.). The residual syrup from which the above compound was separated was dissolved in boiling ether.lilrom the cooled solution there were obtained minute colourless needles melting a t H* 200 THE CONSTITUTION OF THE DISAC(3HARIDES. PART V . 115-116O which gave analytical figures for a trimethyl glucose (Found OMe=41-2. Calc. OMe=41-9 per cent.). The crystalline form solubility and melting point were similar to those recorded for the trimethyl glucose isolated by Denham and Woodhouse (loc. c i t . ) from methylated cellulose and by Haworth and Leitch (loc. cit.) from methylated lactose and mixed melting-point 'determinations with these specimens indicated their identity. [alD (inikial value) + 105'0O in methyl alcohol. (equilibrium value) + 68'. 7 O (after catalysis). The values given by Haworth and Leitch were respectively + 112.9O (initial) and 69-1O (final).This view was supported by polarimetric observations : Conversion of the Crystalline Trimethyl Glucose i n t o Yetramethyl Glucose (Butylene-oxidic). When the trimethyl glucose was digested with methyl iodide and silver oxide i t was found that the product remained incom-pletely methylated. On twice repeating the operation the boiling point (l2Oo/O*9 mm.) and refraotive index (nD 1.4512) of the product showed higher values than were anticipated (b. p. 1 1 6 O 0.9 mm. and nD 1.4454) indicating that methylation of the whole of the material was incomplete. Only after five methylations was the major portion (92 per cent.) of the product obtained as a liquid distilling a t 115-117°/0.9 mm. and showiiig n 1.4457, which values correspond with those recorded for tetramethyl J3-methylglucoside (Haworth T.1915 107 12). The distillate partly crystallised on nucleation with an authentic specimen of this substance and after draining on porous tile the crystals melted a t 38-39O as previously observed by Purdie and Irvine (Zoc. cit.) and a mixed melting-point determination showed no depression (Found OMe = 61.3. Calo. OMe= 62.0 per cent.). The remaining 8 per cent. of the quantity of distilled product showed nD 1-4492 and contained only 5'7.3 per cent. of methoxyl. Hydrolysis of the crystalline tetramethyl 8-methylglucoside pro-ceeded normally on 'digestion with 8 per cent. hydrochloric acid for three hours. On isolating the free sugar in the usual way it crystallised readily and completely melted a t 80-84O and showed nD 1.4585 for the superfused solid which is comparable with the value 1.4588 previously observed (Haworth Zoc. c i t . ) . Moreover a mixture with an authentic spwimen of butylene-oxidic tetramethyl glucose melted a t 80-83O. The specific rota-tory power in equilibrium in water was + 83O. All the above An Example of Steric Hindrance DERIVATIVES OF GALLIC ACID. PaRT XI. 201 properties are in good agreement with those previously quoted in the literature for this compound. The authors are grateful to the Carnegie Trust for a Scholar ship under the terms of which this investigation has been conducted. UNITED COLLEGE OF ST. SALVATOR AND ST. LEONARD, UNIVERSITY OF ST. ANDREWS. [Received January 3rd. 1921.

ISSN:0368-1645    

DOI:10.1039/CT9211900193

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

DERIVATIVES OF GALLIC ACID. PaRT XI. 201 XXII1.-Derivatives of Gallic Acid. Part 11. Gallic Acid (and the Cresotic Acids) and Chloral. By RUPCHAND LILARAM ALIMCHANDANI and ANDREW NORMAN MELDRUM. THE condensation of chloral with methoxybenzoic acids was studied first by Fritsch (Annalen 1897 296 358; 1898 301 360) and the study has been continued by Meldrum (T. 1911 99 1712), by Bargellini and Molina (-41% R . Accad. Lincei 1912 [v] 21, ii 146) and by the authors (T, 1920 117 964). Little if any-thing has been done on the condensation of chloral with the hydroxybenzoic acids. The authors undertook to investigate the condensation with gallic acid; then in order to throw light on the results they obtained they found it desirable to extend the work to the cases of p- o - and m-cresotic acids." The condensation with gallic acid was undertaken as a possible means of ultimately obtaining 3 4 5-trihydroxyphthalic acid (11).The trimethyl ether of this acid has been synthesised by Bargellini and Molina (Zoc. c i t . ) and by the authors (Zoc. ,it.),+ but has not * The work with m-cresotic acid has led to the isolation of three substances, and these are still under investigation. The results wiil be communicated later when the constitution of the substances has been established. t The same acid has been synthesised recently by Herzig and Brunner (AnnuZen 1920 4Zl 283). In each case the same synthetic method was employed and the initial material was the trimethyl ether of gallic acid (or its ester). Herzig and Brunner found it necessary to resort to methylation in the course of the work.Alimchandani and Meldrum had the same experience previously. The explanation is that in the condensation with chloral the sulphuric acid that is employed as condensing agent can hydrolyse the trimethyl ether of gallic acid to the dimethyl ether namely syringic acid, which then condenses with chloral. Herzig and Brunner do not state that this explanation had already been given and do not give any explanation themselvee of the necessity for methylation. Bargellini and Molina found that t,he condensation proceeded without dieturbance. ' H** 202 ATAIMOHAND AN1 AND ilfELDRUM. been completely hydrolysed. The constitution 11 used to be ascribed to pyrogalloldicarboxylic acid but this acid is believed now to have the constitution I11 (Voswinckel and de Weerth Ber., 1912 45 1242; Feist AT&.Yh.nrm. 1907 245 586; 1918, 256 1). It was hoped t o synthesise 3 4 5-trihydrosyphthalic acid and to compare i t with pyrogalloldicarboxylic acid directly. Although this hope was not fulfilled in the course of the work that was carried through compounds were obtained that merit description. From chloral and gallic acid with sulphuric acid as the con-densing agent three different substances can be obtained. When excess of gallic acid is used the product is 3 4 5-trihydroxy-2-tri-c h t l o r o m e t ~ ~ l p ~ ~ t h n l i d c ( I ) . This substance would seem capable of oonversion by hydrolysis into a phthalidecarboxylic acid and this, ultimately into 3 4 5-trihydroxyphthalic acid (11).Hydrolysis, however did not lead to any definite product; the trichloromethyl-phthalide contains the three hydroxy-groups of the gallic acid and behaves much like this acid on treatment with alkali giving material of indefinite composition. 01'1 OH 013 HO/\ €IO/'\ HO/\,CO~H HO! Ico,€r HOI \/ CO,fi HO(,!-YO I \/ C02H CC1;CH-O (1.) (11.) (IIJ.) The production of a trichloromethylphthalide is the usual result of condensiiig chloral with a methoxyhenzoic acid. A condensa-tion of quite another type ocours on using gallic acid and excess of chloral. Much difficulty was experienced in arriving at the formula and constitution of the resulting substance. Ultimately, i t was found that the substance can be obtained by the condensa-tion of chloral with the substance I the reaction proceeding according to the equation C91T50,C1 + 2CC13*CH0 = C13H506C19 + H,O, Once this formula had been arrived at i t was easy to devise con-stitution I V for the substame on the basis of the following behaviour : (1) When treated with a cold alkaline solution the substance does not decompose whilst I under the same treatment darkens as readily as gallic acid.(2) It gives rise to disodium dipotassium dimethoxy- diacetyl, and dibenzoyl derivatives DERIVATIVES OF GALLIC AOID. PART 11. 203 (3) With ferric chloride solution it gives a deep green coloration, such as is given by catechol. These properties might well be expected of a substance contain-ing the two hydroxyl groups shown in IV. What could not have been expected is the great stability which the substance shows under treatment with alkali.It is true that when boiled with an alkaline solution the substance decomposes without yielding definite products; in this case decomposition must begin at the two hydroxyl groups. 0 1 3 the other hand the dirnethoxy- and. dibenzoyl derivatives of the substance can be prepared from i t in the usual manner that is in the presence of alkali. Moreover, these derivatives and the diacetyl derivative as well do not decom-pose when heated with sodium hydroxide solution. Evidently the phthalide ring with its >CH*CCl group and the new ring con-taining two oxygen atoms and two >CH*CCl groups are each remarkably stable. CCl, I CH /\ 0 0 CCI, bH CCl,*CH-O (IV.1 (V.) A third product is obtained from gallic acid when a large excess of ohloral is used; it is found in greater or less amount whenever the substance I V is prepared. Being almost insoluble in organic liquids i t can be separated from IV. Because of this insolubility, and because it gives off the pure state. When yields the substance IV derivative. Its whole constitution V. p-Cresotic acid (VI) equation chloral on keeping it is not obtained in treated with a cold alkaline solution it and chloral. It does not give an acetyl beh;tviour is in accordance with the reacts with chloral according to the C,H,O + 2CC1,eCHO = C1,H80,C1 + R,O. Constitution VIT has been assigned t o the substance in view of the facts that with ferric chloride it does not give the violet zolor-ation characteristic of salicylic acids and that it is acid t o litmus and forms saits easily 204 AIJM(YHANDAN1 AND MELDRUM : Thus the same heterocyclio ring containing two oxygen atoms and two >CR*CCl groups is present in the substances IV and VII.Like IV V I I is remarkably stable under treatment with a hot alkaline solution. When boiled with zinc dust and acetic acid V I I gives rise to the substance VIII which is a derivative of phenylethylidene chloride. I n this connexion i t is to be recalled that Pinner (Bey. 1898 31 1935) using the same reagents con-verted the two -CCl groups in hexachlorodimethyltrioxine (XV) into *CHCl groups the trioxine ring itself undergoing no change. In the present case the heterocyclio ring breaks down and the resulting *CH(OH)*CCl group is converted into *CH,*CHCl,.Compound VIII when fused with potassium hydroxide gives rise to IX. Both these compounds V I I I and IX give with ferric chloride the violet coloratioii characteristic of salicylic acids. CCI, 6H /\ 0 0 I OH I CH*CCl OH CO,H/' CO,H/\/ CO,H/\ICH,*CHCI L c=, (VIII.) II CH3 0 CH3 (VI. ) (VII. ) C0,H CH,*CO,H 6 cJa3 (TX. 1 The reaction of chloral with o-cresotic aoid (X) proceeds accord-ing t o the equation 2C,H,O + CCl,*CHO = C,,H,,O,Cl + H,O. The product has the constitution XI the assumption being made that chloral attacks the cresotic acid molecule a t the para-position with respect to the hydroxy-group. The substance XI is t o be, compared with that which Causse (Bzc71.Soc. ehim. 1890 [iii] 3, C0,H CO,H CO,H f XI. ) (=I. DERIVATIVES OF GALLIC ACID. PART 11. 205 861) obtained from chloral and resorcinol and which Hewitt (T., 1896 69 1265) showed to have the constitution XII. Thus the condensation of a hydroxybenzoic acid with chloral can take place in a t least three different ways: ( A ) As in the case of a methoxybenzoic acid a trichloromethyl-phthalide is formed; gallic acid when used in excess gives the trichloromethylphthalide (I). ( B ) A new heterocyclic ring is formed; p-oresotic acid gives rise to the substance VII. Gallic acid also when taken with excess of chloral gives the compound IV containing the new heterocyclic ring along with the phthalide ring. The new ring is remarkable for stability under treatment with alkaline solution.Under this treatment the substance V I I is not hydrolysed. The substance IV is of course vulnerable to attack by alkali by reason of its two hydroxy-groups but the dimethoxy- diacetyl and dibenzoyl derivatives of I V are as stable as the substance VII. Here the stability also of the phthalide ring and of the CCl group attached to it is worthy of note. Chloral has a strong tendency to take part in the formation of heterocyclic rings. Wallach obtained substances which he termed chloralides by the condensation of chloral with glycollic malic, tartaric and trichlorolactic aoicls. The first known chloralide (XIII) he obtained from chloral and trichlorolactic acid (Annalen? 1878 193 1). From salicylic acid he obtained the substance XIV.The heterocyclic rings in XI11 and XIV however are different from one another CCl, 6H /\ 0 0 d0 -&H* CC1, (XIIT.) cc1, /\ \/ 6H 0 0 CCl,-bH bH-CCl, -and from the heterocyclic ring which /\ 0 0 /\ 0 0 I CH I I CH*CCI co I \/ 0 (XV.) \i\ t i \/ (XIV.) CH3 A CH;CH CH*CH, \/ (XVII.) 6H 0 1 1 0 0 (XVI. 206 ALIMOHANDANI AND MELDRUM : is present both in I V and VII. The nearest approach to this new ring is to be found in hexachlorodirnethyltrioxine (XV), which Pinner obtained by the action of chloral on formaldehyde in the presence of sulphuric acid ( l o c . cit.). Various meta-chlorals have been described one of whicli may have the constitution XVI. Both XV and XVI have the same ring structure as paraldehyde (C) In the oase of o-cresotic acid two molecules of the acid condense with one molecule of chloral and the substa.nce XI is formed.No closed chain is produced; formation of the new heterocyclic ring is impossible owing to the position of the hydroxy-group and formation of the phthalide which appears possible, does not take place. (XVII) . E X P E R I M E N T A L . 3 4 5-Trihydroxy-2-trichloromethylphthalide (I). Gallic acid (24 grams) chloral hydrate (10 grams) and sulphuric acid (100 C.C. of 94-95 per cent.) were mixed together and the solids dissolved by vigorous shaking. After about forty-eight hours the liquid was cooled by a freezing mixture when some gallic acid separated. The mixture was filtered the filtrate poured on ice and the white solid which separated was collected washed, and crystallised from hot water; the yield was 10 grams.On recrystallisation from a mixture of acetone and chloroform it separated in double pyramids melting a t 210-212° (Found : C1=35*55. The triacetyl derivative was prepared by the aid of acetic anhydride in the presence of a few drops of concentrated sulphuric acid. It crystallised from absolute alcohol in white lustrous silky needles melting at 1 7 2 O (Found C1= 25.12. C,,H,,08Cl requires C1= 25-02 per cent .) . Hydrolysis of these substances did not lead t o definite results. When treated with an alkaline solution the former 'darkened a t once and no new substance could be isolated from the solution. When it was heated with water in sealed tubes part was unchanged and part was converted into tarry products.No better result was obtained by heating the substance with. water and lead oxide or with water and potassium hydrogen sulphate. C,H,O,Cl requires C1=35.54 per cent.). Luctone of 7 8-Dihyclroxy-2 4-bist.1.ichlol.omethy~-6-P-ti.ic~~l~~~(~-Gallic acid (24 grams) chloral hydrate (40 grams) and sulphuric acid (200 C.C. of 94-95 per cent.) were mised together when the a-hydroxyethyl-1 3-~enxdioxine-5-carbozylic Acid (IV p. 203) DERIVATIVES OF GALLIC ACID. YART 11. 207 solids dissolved. After about twenty-four hours a gelatinous mass separated floating on the surface of the liquid the amount of which increased in course of time. On the fourth day the solution, together with the floating mass was poured on ice when a white solid separated which was collected washed with water and dried on porous tile; the yield of the crude product was 40 grams.It ci-ystallised from either methyl alcohol or ethyl alcohol in trans-parent stout plates containing alcohol of crystallisation. It effloresced at the ordinary temperature lost alcohol com-pletely a t 100-llOo and then melted and decomposed a t 277-279O (the melting point being determined quickly) (Found MeOH = 10.00. C13H,0,Clg,2MeOH requires MeOH = 10.00 per cent. EtOH = 13.62. C,,H,0,C!lg,2EtOH requires EtOH= 13.76 per cent. C1= 5556. C13E150tjC19 requires C1= 55.43 per cent .). Disodium Compound.-The substance was dissolved in very dilute, cold sodium hydroxide solution and t o this a concentrated solu-tion of the alkali was added.A yellow sodium compound separated which was collected washed with a small amount of water and then crushed on a porous tile (Found Na=7-47. C,,H30,Cl,(O~a) requires Na = 7.42 per cent .). Dipotassium Compound .-A yellow potassi.zm compound was obtained in the same manner as the sodium compound (Found: K = 11.76. Dimethoxy-derivatiue .-The substance (5 grams) was dissolved in methyl sulphate (20 c.c.) with heating To this a solution of potassium hydroxide was added when a vigorous reaction took place which was cheoked by cooling. Addition of the alkali was continued until the solution was faintly alkaline. A brown, sandy material separated which was collected and dried. It crystallised from glacial acetic acid in colourless prismatic needles, and after recrystallisation from acetone melted a t 192-193O (Found C1=52-80.The diacety7 derivative was prepared by means of acetic anhydride in the presence of a few drops of concentrated sulphuric acid. It crystallised from glacial acetic acid in clusters of pris-matic needles and when recrystallised f roni absolute alcohol. melted a t 217-219O (Found C1= 48.31. C,,HgO8Cl9 requires Cl = 48.37 per cent.). Di benzoyZ Derivative .-The substance (2 grams) was dissolved in benzoyl chloride (10 0.c.) with heating. To this a slight excess of sodium hydroxide solution was added when a vigorous reaction took place and an oil separated which solidified when treated with water. When crystallised from glacial acgtic acid and then from absolute alcohol it formed slender prismatic needles melting at C,3H30,Clg(OK) requires K = 11.96 per cent.).C,,Hg06Clg requires C1= 52-85 per cent.) 208 A.LIM.CHANDAN1 AND MELDRUM: 234-236O (Found C1= 40.5. C,7H,,08C1 requires C1= 40.7 per cent .) . Luctone of 7 8-P~BTrichloroethylidenedioa.1~-2 4-bistrichloro-methyl - 6-B~~-trichloro-a-htydroxyethyl- 1 3 - benzdiolmne - 5 -carboxylic gcid (V p. 203). Gallic acid (20 grams) chloral hydrate (60 grams) and sulphuric acid (250 C.C. of 94-96 per cent.) were mixed together and on vigorous shaking most of the solid dissolved. The impurities were filtered off. During the course of three days a solid separated, which was collected and dried on porous tile. The dried material contained a quantity of the substance IV which was removed by extracting the material thrice with acetone.The yield of the insoluble product was 10 grams. It is an amorphous powder insoluble in most organic liquids. It begins to char a t about 250° and when quickly heated melts a t 268-270O (Found C1= 64.3. Cl5H4O6Clll requires C1= 60.3 per cent .). 6-Methyl-2 4- bistrichloromethyi-l 3-ber~zcEioxine-8-carboccylic ,4 cid (VII p. 204). p-Cresotic acid (20 grams) chloral hydrate (40 grams) and sulphuric acid (150 C.C. of 98-100 per cent.) were mixed together. On vigorous shaking and slight warming the solids dissolved. After five days the solution was poured on ice when a yellowish-white solid separated whiclh was collected washed with water and dried. The product was a mixture of the condensation product and unaltered p-cresotic acid.The latter was removed by grind-ing the material with a small quantity of absolute alcohol, in which the cresotic acid was extremely soluble and the derivative sparingly so; the yield was 20 grams. The substance was sparingly soluble in most organic solvents; from acetone it crystallised in fine snow-white needles melting and decomposing a t 385-286O (Found C1= 49.7. C,,H,O,Cl requires C1= 49.9 per cent .) . Sodium Sul t .-The substance was dissolved in dilute sodium hydroxide solution and to this a concentrated solution of the alkali was added when a white crystalline sodium salt was obtained (Found Na=5.29. C,2H70,C1,Na requires Na=5-11 per cent.). 4-Hydroxy-5-&3-dichloroethyl-m-tolztic 9 cid (VIII p. 204) The substance (VII) (2 grams) and glacial acetic acid (40 c.c.) were boiled together but the solid did not dissolve completely DERIVATIYES OF QA.LLIC ACID.P-T 11. 209 To the hot mixture zinc dust (10 grams) was added in small amounts when a vigorous reaction took place and the undissolved material dissolved freely. Heating was continued for about an hour after which time water (60 o.c.) was added and the mixture boiled for fifteen minutes more. The hot solution having been filtered gave on cooling white needle-shaped crystals which, when recrystallised from benzene melted a t 205-207O (Found : C1= 28.8. It crystallises in slender, glistening needles (Found H,O = 11.7. C2,H180,C14Ca,4H,0 requires H,O=11*9 per cent.). The anhydrous salt was also analysed (Found Ca = 7.53.C,H,,O,Cl,Ca requires Ca = 7.48 per cent.). C,,H,,O,Cl requires Cl= 28.5 per cent.). The calcium salt is characteristic. 6-Hydrozy-5-carboq-m-tolylacetic d cad (IX p. 204). The substanoe just described (2 grams) and potassium hydroxide (10 grams) were fused together in a nickel crucible a t 250-260° for about half an hour. The product was dissolved in water and acidified when brown-coloured needles were obtained. From acetic acid it crystallised in white soft needles melting and decom-posing a t 2 5 7 O (Found C = 57.3 ; H = 4.9. C,,H,,O requires C=57.1; H=4*8 per cent.). The silver salt was prepared (Found Ag ='50*8. CloH80,Ag, requires Ag = 50.9 per cent.). PBB-Trichloro-4 4~-dihydroxy-aa-di-m-tolyEethane-5 5'-dicarboqlic Acid (XI p. 204). o-Cresotic acid (5 grams) chloral hydrate (5 grams) and sulphuric acid (50 C.C. of 94-95 per cent.) were mixed together, when the solids dissolved easily. After three days the solution was poured on ice when a white solid separated which was collected washed with water and dried on a porous tile; the yield of the crude product was 7 grams. It was sparingly soluble in most organic solvents and separated from glacial acetic acid as a white, microcrystalline powder melting and decomposing a t 283-285O (the melting point being determined quickly) (Found C1= 24-32. C,8H,,0,C13 requires C1= 24.55 per cent.). C,8H,30&13Ca~ requires Ca = 8.5 per cant.). The calcium salt was prepared (Found Ca == 8.3. MADHAVLAL RANCHHODLAL SCIENCE INBTITUTE, AHMEDABAD INDTA. [Received. November 27th. 1920.1

ISSN:0368-1645    

DOI:10.1039/CT9211900201

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

210 SINUH AND LAL STUDIES IN SUBSTITUTED XX1V.-Studies in Substituted Quaternary Azonimn Compounds containing an Asymmetric Nitrogen Atom. Part IV. Additive Compounds of Thiocarbamide with Axonium Iodides. By BAWA KARTAR SINGH and MIRI LAL. THE object of this investigation was to determine the ratio in which thiocarbainide unites directly with substituted quaternary azonium iodides of different constitution. Werner and Atkins (T. 1912 101 1988) found that whilst the simplest ammonium bases and primary amines form tetrathjo-carbamide derivatives the tertiary amines and quaternary ammonium bases appear always to produce dithiocarbamide compounds. The substituted quaternary azonium iodides also form dithio-carbamide derivatives thus resembling the quaternary ammonium compounds in their capacity to yield additive compounds with thiocarbamicie.Exceptions to the above rule are found in phenylbenzylallyl-azonium iodide which forms a monothiocarbamide conipoun'd and the phenyldimethyl and phenyldiethyl compounds which do not unite a t all with thiocarbamide. The thiocarbamide additive compounds described below have coniparatively high melting points which seems to imply increased stability of both thiocarbamide and azonium iodides in these sub-stances. The azonium iodides have comparatively low deconi-position temperatures. E x P E R I M E N T A L. Preparation of Additive Compounds of Thiocarbumide and 9 xon+um Iodides. The general method adopted a t first in the preparation of these compounds consisted in mixing hot alcoholic solutions of thiocarb-amide and azonium iodides.On cooling the additive compound separated and' was purified by recrystallisation from alcohol. Better yields and purer products however were obtained by heat-ing a concentrated alcoholic solution of the components containing an excess of thiocarbamide. The solution was cooled and then filtered. To the filtrate alcohol-free dry ether was added which precipitated the compound QUATERNARY A4ZONIUM COMPOUNDS ETC. PART IV. 21 1 The azonium iodides used were prepared and purified according to methods already described (Singh T. 1913 103 604; 1914, 105 1972; 1920 117 1202). The results obtained are given below. T hi0 car bamide and 1% e rc y lme t Ib y le t hp Lu z oni u m l o d id e .-This compound forms fine silky needles melting a t 192-193O (Found: I = 30.24 29.7.PhMeEtl[NH,]NI,2CH,N,S requires I = 29-54 per cent .) . Thiocar bamide and Pkeny l b enz ylme thy laz onium Iodide .-The additive compound was obtained as a white crystalline substance melting a t 2 1 1 O (Found I = 25.68 26.06. Ph[CH2Ph]Me[NH,]NI 2CH4N,S requires I = 25.81 per cent.). Iod id e .-The double compound was obtained as a white crystalline substance melting a t 180-181° (Found I = 24-72 24-28. Ph[CH,Ph]Pr[NH,]NI 2CH4N,S requires I = 24.42 per cent .). Z’hiocar bamide and I’henpl ben+yZallyluzoniurn Iodide.-This compound forms fine silky needles melting a t 187-188O (Found : I = 28.41. Ph[CH,Ph][C,H,][NH,]NI,CH,N,S requires I = 28.73 per cent .). It is noteworthy that phenyldimethylazonium iodide and phenyl-diethylazonium iodide do not give any compounds with thiocarb-amide though in each case the azonium iodide was boiled with varying proportions of thiocarbamide in alcoholic solution. This may be due to the presence of two like alkyl radicles in the azonium iodides. Werner and Atkins ( b e . cit.) also found that tetramethylammonium iodide does not exhibit any tendency to unite with thiocarbamide. The iodine was estimated in all the above compounds by boiling with an excess of iron alum and dilute sulphuric acid and titrating the liberated iodine with a standard solution of sodium thiosulphate. T hio car barnide and Ph enyl b e n zylpro pylu zon i u m THE CHEMIO.U LABORATORY, GOVERNMENT COLLEGE LAHORE, PUNJAB INDIA. [Received JarLuary 2Wh 1921.

ISSN:0368-1645    

DOI:10.1039/CT9211900210

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

21 2 BARROW AND GRXFFITHS CONDENSATION OF XXV.-Condensation of p-Nitrobenxyl Chloride with A New Mode of F o r m a t i o n Nitroso-compounds. of N- Oximino-ethers. By FRED BARROW and EVAN DALTON GRIFFITHS. DURING the course of an investigation on the action of alcoholic potassium hydroxide on p-nitrobenzylidene chloride it was found by one of us that this substance and p-nitrobenzyl chloride readily undergo condensation with aromatic aldehydes under the influence of the condensing agent with the formation of derivatives of stilbene oxide. Thus for example p-nitrobenzylidene chloride con-denses with y -nitr obenzaldehy de yielding a-chlo r o -pp/-d ini t r o -stilbene oxide (I) whilst y-nitrobenzyl chloride gives rise to two stereoisomeric cis- and trans- ypl-dinitrostilbene oxides (11).In view of the remarkable ease with which these condensations take place a study of the behaviour of the nitro-substituted benzyl-idene and benzyl chlorides towards other classes of compounds con-taining an unsaturated group is being undertaken by the authors, and in the present communication the interaction of p-nitrobenzyl chloride with aromatic nitroso-compounds is described. It has been found that nitrosobenzene and also the p-nitroso-derivatives of the mono- and di-alkylanilines readily condense with p-nitro-benzyl chloride in the presence of alcoholic potassium hydroxide, yielding A‘-aryl ethers of p-nitrobenzaldoxime thus : N0,°C6H4*CH2CI + NPh:O = HCI + N02*C6H4*CH:NPh:0. The constitution of these condensation products was first estab-lished by the investigation of the behaviour of the N-phenyl ether prepared from p-nitrobenzyl chloride and nitrosobenzene towards hydrochloric acid.When heated with concentrated hydrochloric acid the phenyl ether is rapidly hydrolysed to p-nitrobenz-aldehyde and p-chloroaniliue. The initial products of the hydro-lysis evidently consist of the aldehyde and P-phenylhydroxyl-amine the latter compound being subsequently converted into p-chloroaniline by the further action of the hydrochloric acid : N0,-C6H4*CH:NPh:0 + NO,*C,H4*CHO + C,H,*NH*OH -+ C6H4C1*NH, P-NITROBENZYL CHLORIDE WITH NITROSO-COMPOUNDS. 21 3 Further evidence as to the constitution of these condensation products is furnished by the direct synthesis of the N-phenyl ether, by the interaction of 8.phenylhydroxylamine and p-nitrobenz-aldehyde in alcoholic solution : NO,*C,H,*CHO + NHPh*OH + H,O + NO,*C,H,*CH:NPh:O.It has been shown by Semper and Lichtenstadt (Ber. 1918 51, 928) that the N-methyl ether of phenyl p-tolyl ketoxime exists in two stereoisomeric forms corresponding with the syn- and anti-modifications of the oxime. In view of the existence of these isomerides and also of the simultaneous formation of cis- and trans-py I-dinitrostilbene oxides by the condensation of p-nitrobenzyl chloride with p-nitrobenz-aldehyde the crude AT-p-dimethylaminophenyl ether prepared from p-nitrobenzyl chloride and p-nitrosodimethylaniline was sub-mitted t o a careful fractional crystallisation from alcohol in order to ascertain whether stereoisomeric ethers are formed in this con-densation but no evidence of the existence of a second isomeride was obtained.The interaction of p-nitrobenzyl chloride with nitroso-compounds is analogous to that of the aliphatic diazo-compounds described by Staudinger and Mie-scher (Ilelv. Chim. Acta 1919 2 554) who have shown that diazo-compounds of the type I l>C<E readily react with nitrosobenzene and its derivatives yielding N-oximino-ethers. Attempts to condense p-nitrobenzyl chloride with nitrosoamines, NRR’*NO were unsuccessful the nitrosoamine being recovered unchanged whilst the p-nitrobenzyl chloride was converted by the alcoholic potassium hydroxide into ppl-dinitrostilbene. The reaction described in the present communication is being extended to other nitroderivatives of benzyl chloride.It has also been found that aromatic chloro-ketones of the types CH,Cl*COR and CHRCl*COR react with nitroso-compounds yielding LN-oxirnino-ethers . N N EXPERIMENTAL. p-Nitrobenzaldoxime-N-phenyl Ether NO,*C,H,*CH:NPh:O Equimolecular proportions of p-nitrobenzyl chloride (8.6 grams) and nitrosobenzene (5.4 grams) were dissolved in the minimum amount of alcohol a t the ordinary temperature and the solution treated with one and a-half times the theoretical amount of potassium hydroxide (4-2 grams) dissolved in methyl alcoho 214 BAItROW AND GRIP'BI'L'HS CONDENSATION OF (30 c.c.). On the addition of the alkali the colour of the solutioii rapidly changed from green t o light reddish-brown and the oximino-ether separated in the form of a yellow crystalline powder.After remaining for one hour a t the ordinary temperature this was filtered freed from potassiiim chloride by washing with water, and crystallised from alcohol. p-Yitro benzaldorcime-N-phPnyl et he?- crystallises iii small slender, pale yellow leaflets which melt at 1 8 2 O (Foiind C = 64-46 ; H=4*29; N=11.50. C,,H,,,O,N requires C=64.46; PI=4.13; N = l l * 5 7 per cent.). It is very readily soluble in chloroform, moderately so in alcohol and benzene and almost insoluble in ether and light petroleum. Hydrolysis.-The AT-phenyl ether (4 grams) was heated for half an hour with concentrated hydrochloric acid (30 c.c.) and after cooling the solid which separated was purified by crystallisation from dilute alcohol and theii from water.It crystallised in almost colourless needles melting a t 1 0 6 O and was ident,ified as y-nitro-benzaldehyde. The acid solution from which the p-nitrobenz-aldehyde separated was extracted with ether and then rendered alkaline by the addition of sodiuni hydroxide. The base thus liberated gave with acetic anhydride an acetyl derivative melting a t 1 7 5 O which was identified as 2'-chloroacetanilide. C'onderzsation of p-i~itrobe?kzaldehycle ccizd P - 1 ~ ~ ~ e i L ~ l l ~ y d ~ o ~ y ~ a n b i n For the purpose of comparison p-iiitrobenzaldoxinie-N-phenyl ether was prepared by the condensation of p-nitrobenzaldehyde and /3-phenylhyd toxylamine. Equimolecular proportions of the two components were separately dissolved in alcohol and after mixing the solution was heated for half an hour on the water-bath.The solid which separated crystallised f roni alcohol in pale yellow leaflets meltiiig a t 1 8 2 O and was identical with the I\'-phenyl ether obtained by the condensation of p-nitrobenzyl chloride and nitrosobenzene. p-Nitrobenzaldoxime-N-p-~~?~~~thylurnlnoyhenyl JZl'ther, NO,*C,H,*CH:NO.C,H,*NMe~. 28.6 Grams of p-nitrobenzyl chloride and 25 grams of p-nitroso-dimethylaniline were dissolved in the minimum amount of alcohol a t the ordinary temperature and to the solution was added the theoretical amount of potassium hydroxide (9.3 grams) dissolved in methyl alcohol (66 c.c.). On the addition of the alkali the colour of the solution rapidly changed from green t o dark chooo Y- NlTROBENZYL CHLOR CDE WI'I'H NITROSO-COMPOITNDS.2 15 late-red and the osimino-ether separated in the form of a brownish-red powder. After one hour this was filtered washed with water and purified by crystallisation from a mixture of equal parts of pyridine and alcohol the yield being 33.5 grams. The N-p-rlimeth?/Znminophe?~yl ether crystallises in chocolate-red needles melting a t 20lo and forms a yellow hydrochloride and sulphate (Found C = 63.09 ; H = 5.20; N = 14.58. C15H1503N3 requires C=63*15; H=5.26; N=14*74 per cent.). It is readily soluble in chloroform and pyridine sparingly so in benzene and alcohol and practically insoluble in light petroleum and ether. When heated with concentrated hydrochloric acid it readily under-goes hydrolysis with the formation of 2'-nitrobenzaldehyde.p -A- i t 1-0 Z e 11 xuldo xi m e-N - p -die t h y lami no p h e n y 1 E t her, NO,* C,H CH :NO.C,H,*NE t . This was prepared from p-nitrosodiethylaniline and p-nitro-benzyl chloride in a similar manner t o that described above and purified by crystallisation from a mixture of equal parts of pyridine and alcohol. It crystallises in large lustrous bright crimson leaflets which melt a t 167O is readily soluble in pyridine and chloroform but only sparingly so in alcohol and benzene (Found: C = 64-99 ; €1 = 6-01 ; N= 13.34. C,,H,,O,N requires C = 65.17 ; H = 6.07 ; N = 13.42 per cent.). p-.Sitro b enznlcloxime-N-p-d i-n-propylaminoph en,y7 Ether, NO,*C,€I,*CH:NO*C,H,.NPr.,. This compound is obtained in a similar manner from p-nitrosodi-wpropylaniline and separates from alcohol in felted slender dark crimson needles melting a t 135-138O with previous softening a t 130° (Found N= 12.31.C1,,H,,0,N3 requires N = 12-32 per cent.). It is much more readily soluble in the common solvents than the dirnethylarninopheiiyl ether. p-n'itro berizaldoxi,ne-N-p-ethylami?~ophenyl Ether , N0,*CGH,.CH:NO*C,H4*NHEt. The condensation of p-nitrosoethylaniline prepared by Fischer and Hepp's method ( B e y . 1886 19 2993) with p-nitrobenzyl chloride was carried out at 4 5 O the reaction proceeding only slowly at the ordinary temperature. The p-ethylamin.ophen y l ether crystallises from benzene in deep crimson needles melting at 1 6 8 O (Found C = 63.08 ; H = 5-25 ; N=14*53. C,,H,,O,N requires C=63*15; 13=5-26; N=14.7 216 PERKIN AND TUCKER THE OXIDATION OF CARBAZOLIC.per cent.). It is readily soluble in alcohol an’d chloroform less so in benzene and only very sparingly so in ether and light petroleum. p-Nitro benzaldoxime-N-p-nitrophenyl Ether, NO,*C,H,*CH:NO* C,H,-NO,. The p-nitronitrosobenzene required for the preparation of this compound was obtained by the addition of the calculated amount of Caro’s acid to a well-cooled solution of p-nitroaniline in con-centrated sulphuric acid and was separated from the accompany-ing p-dinitrobenzene by fractional distillation in steam (compare Bamberger Rer. 1903 36 3808). The product resulting from the condensation of the nitroso-compound with p-nitrobenzyl chloride consisted of a mixture of the oximino-ether and pp’-dinitrostilbene, which was readily separated by taking advantage of the sparing solubility of the latter compound in alcohol. The p-nitrophenyl ether separates from alcohol in the form of a pale yellow crystalline powder which melts a t 1 8 6 O (Found: N=14.54. CI3H9O5N3 requires 14.63 per cent.). It is readily soluble in chloroform moderately so in alcohol and benzene and insoluble in ether and light petroleum. CHEMISTRY DEPARTMENT, BIRKBECK COLLEUE. [Receiwd J a w a r y 27th 1921.

ISSN:0368-1645    

DOI:10.1039/CT9211900212

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

216 PERKIN AND TUCKER THE OXIDATION OF CARBAZOLIC. XXV1.-The Oxidation of Carbaxole. By WILLIAM HENRY PERKIN jun. and STANLEY HORWOOD TUCKER. FISSION of the rings of carbazole by oxidation methods has been attempted by several investigators. Graebe and Glaser (Ber., 1872 5 12) found that chromic acid in glacial acetic acid solution reacted very vigorously with carbazole but the product isolated from the reaction mixture contained chromium and was not identifiable. It is noteworthy that Branch and Smith (Y. Amer. Chem. SOC. 1920 42 2405)" obtained by the oxidation of carb-* The results contained in the following communication were reported to the Department of Scientific and Industrial Research in October 1919. The above-mentioned paper by Branch and Smith made it advisable to publish an account of our experiments although they are at present somewbat incomplete PERKEN AND TUCKER THE OXLDATION OF CARBAZOLE.217 azole with silver oxide in benzene solution a product which contained silver. The only instance of ring-rupture in the carbazole molecule is that recorded by Padoa and Chiaves ( A t t i . R. Accad. Lincei 1908, [v] 16 ii 762) who found that 2:3-diethylindole is produced by heating carbazole with nickel in an atmosphere of hydrogen under pressure. Wieland and Gambarjan (Ber. 1906 39 1506) stated that carbazole was unaffected by shaking in benzene solution with lead peroxide and also that a cold 3 per cent. solution of potassium permanganate in acetone had similarly no action on carbazole. We find however that when finely divided potassium perman-ganate is added to a hot solution of carbazole in pure acetone a vigorous reaction sets in and from the reaction mixture three pro-ducts have been isolated two well-defined crystalline compounds melting respectively a t 220-22lo and 265O and an amorphous substance.The amorphous substance which is the main product of the oxidation has not yet been analysed. It is evidently of high molecular weight since it cannot be distilled. The two crystalline compounds which we distinguish provisionally as ( A ) and ( B ) both possess the formula C,,H,,N, and are undoubtedly isomeric dicar bazyls analyses and molecular-weight determinations by the cryoscopic method in the solvents benzene and naphthalene being in complete support of this formula.There is no indication a t present that either a benzene ring or the pyrrole ring of the carbazde molecule can be opened by oxida-tion methods. During our oxidation experiments carried out on a large scale no amino-acids could be detected by means of the copper or lead salts. Dicarbazyl ( A ) m. p 220-221° crystallises from benzene in large crystals which evidently contain benzene of crystallisation, and these when exposed to the air fall t o powder. It crystallises particularly well from acetone and the measurement of the crystals thus obtained is ‘described on p. 222. Dicarbazyl ( B ) m. p. 265O crystallises from a mixture of acetone and benzene in minute glistening crystals which are not suitable for measurement. Carbazole fused with anhydrous oxalic acid gives a melt which is first green and finally becomes blue.This “carbazole-blue” test when applied to dicarbazyl ( A ) of melting point 220° gives a distinctly green melt which turns blue; but the isomeride ( B ) , melting a t 265O gives only a faintly blue melt different in tint from the above. It is remarkable that neither of these two substances forms 218 PERKIN AND TUCKEB THE OXIDATION OF CARBAZOLE. picrate for carbazole and its homologues combine readily with picric acid. The behaviours of the dicarbazyls ( A ) and ( B ) towards concen-trated sulphuric acid are characteristic and different from that given by carbazole. Both react with nitric acid) to give yellow nitro-derivatives and are violently attacked by bromine. Three possible types of isomeric dicarbazyls may be distinguished, namely : (1) i?A‘-Dicarbazyl in which the linking of the carbazyl radicles is effected through nitrogen.There is only one possibility in this case namely the hydrazine formula, (2) CC-Dicarbazyls in which the union of the carbazyl radicles is attained through carbon. In this class there are several possible modes of linking but the two most probable are represented by (I) and (11) since in the carbazole molecule the hydrogen atoms 3 S’-Dicarbazyl. (I. 1 1 1’-Dicarbazyl. VJ*) in the positions 3 and 1 are the most reactive and hydrogen in position 3 is more reactive than hydrogen in position 1. (3) CN-Dicarbazyls in which the carbon of one carbazyl radicle is linked to the nitrogen of the other for example formula (111).(111.) VV-1 Relevant t o the discussion of the constitution of the dicarbazyls is the observation of Wieland ( B e y . 1913 46 3300) that by the oxidation of diphenylainine with sodium dichromate in glacial acetic acid solution in the presence of sulphuric acid diphenyl-benzidine (IV) is obtained. The behaviuur of carbazole under these conditions is being investigated PERmN AND TUCKER THE OXIDATION OF CARBAZOLE. 219 It is probable that the two modifications of dicarbazyl obtained under the conditions described in this research are not the first products of the oxidation of carbazole. It is well known that the hydrogen atoms in positions 1 and 3 in carbazole are particularly susceptible t o substitution and several instances of wandering to these positions of groups attached to the nitrogen atom of the imide group have been observed.Thus H. Schott (D.R.-P. 134983 1901) has shown that N-nitrosocarbazole is converted into 3-nitrosocarbazole when it is dissolved in glacial acetic acid and the solution treated with concentrated hydrochloric acid. Again, Ciamician and Silber (Gazzetta 188’2 12 272) have found that, under certain conditions of temperature the potassium salt of carbazole-N-carboxylic acid breaks down and the carboxyl group wanders to one of the benzene rings giving rise to carbazole-l-carboxylic acid. Wieland and Gambarjan (Ber. 1906 39 1503 ; Wieland Ber., 1913 46 3296) have moreover shown that in the preparation of diphenylbenzidine from diphenylamine by oxidation with sodium dichromate as above mentioned tetraphenylhydrazine, is first produced; the sulphuric acid present then converts tetra-phenylhydrazine by intramolecular rearrangement into diphenyl-b enz id in e .Hence the oxidation of carbazole may lead in the first place to the production of NN-dicarbazyl and this may then undergo intra-(CGH,),N*N(C,Hd,, I niolecular change with subsequent formation of the dicarbazyls ( A ) and (B). Wieland and his collaborators (Ber. 1906 39 1506; Annalen, 1912 392 184) have attempted without success to prepare NN-dicarbazyl (named by Wieland and his collaborators (‘ bis-diphenylen-hydrazin ”). There is no evidence to show that this compound has been prepared or that it is even capable of exist-ence; nor is there any evidence that any dicarbazyl had been pre-pared before the publication of the present investigation.At the present stage of this research it is not possible to say t o which type or types the two crystalline dicarbazyls which we have isolated belong except that one of them must belong to type I1 or 111. With the object of testing whether either compound was the hydrazins attempts were made to reduce these tw 220 PERKIN AND TUCKER THE OXIDATION OF CARBAZOLE. substances but without success. It was t o be expected' that the hydrazine NN-dicarbazyl would readily reduce t o carbazole. Practical exigencies for example sparing solubility of these two compounds in alcohol glacial acetic acid and other solvents have given rise t o difficult experimental conditions and the question is being further investigated.Attempts t o acetylate the two crystalline dicarbazyls ( A ) and ( B ) have met with only partial success. Both compounds crystal-lise unchanged from acetic anhydride but when boiled with excess of anhydride in the presence of a catalyst (for example a trace of concentrated sulphuric acid zinc chloride potassium hydrogen sulphate etc.) the solution darkens and on addition of water, substances are precipitated which are in part soluble in hot alcohol. The investigation of these alcohol-soluble products is in progress. Acetyl chloride and benzoyl chloride appear t o have merely a solvent action but this behaviour cannot be taken as proof of the absence of the imide group in these dicarbazyls since carbazole itself reacts only imperfectly with these reagents.E X P E R I M E N T A L . Oxidation of Carbazole with Potassium Permanyanate in ,4cetone Solution.-To a boiling solution of 100 grams of carbazole in 700 grams of pure acetone are added 100 grams of finely powdered potassium permanganate in two portions of 50 grams each. [The acetone was purified by boiling with excess of potassium permanganate until the purple colour persisted and was then distilled from the mixture.] On addition of the first 50 grams a vigorous reaction sets in and continues for about three t o five minutes; the permanganate colour will then have practically disappeared. The remaining 50 grams of potassium permanganate are now added to the acetone solution and the whole boiled vigorously on the water-bath with occasional shaking, until the pink colour is discharged which is the case in about one hour.After filtering from the brown manganese precipitate the faintly pink filtrate commences a t once to deposit a small quantity of a white pasty precipitate which soon becomes crystalline. The manganese precipitate is washed with hot acetone the residue is returned t o the flask and extracted twice with boiling acetone (about 400 grams each time). ' The united acetone filtrates which exhibit a deep violet fluorescence are now evaporated. The manganese precipitate was extracted twice with benzene, but yielded only 1-2 grams of syrup and this was added to the acetone extract. In order to determine whether any other sub PERKIN AND TUCKER THE OXIDATION OF (TARBAZOLE.221 stances such as amino-acids had been formed during the oxida-tion the residue in one case was extracted several times with boil-ing water until the filtrate was nearly neutral. The united filtrates were rendered slightly acid with acetic acid the liquid heated to boiling filtered from a slight amount of a dirty brown precipi-tate and solutions of copper acetate and of lead acetate added t o portions of the filtrate but no copper or lead salts of organio acids separated. Some of the manganese residue was mixed into a paste with water and sulphur dioxide gas passed until the manganese had dissolved but only a slight amount of a dirty brown precipitate separated. Examination of the Acetone Extract.-The brown oxidation mass obtained as described above which weighs 110 grams is dis-solved in boiling benzene.The deep red-coloured liquid which exhibits a violet fluorescence deposits white crystals on keep-ing. These are washed with a little cold benzene. The crystals melt a t 210° with evidence of softening a t 200° and weigh 24 grams. This substance (called dicarbazyl A ) is first recrystallised from benzene and finally twice from acetone and is thus obtained in large crystals (20 grams) melting a t 220-221O (224-225O corr.). The combined mother liquors of dicarbazyl ( A ) are mixed with twice the volume of acetone and the solution so obtained is allowed to remain several days. A white microcrystalline substance (dicarbazyl B) about 5 grams gradually separates and melts a t about 252O but after recrystallisation a t 2 6 5 O (270O corr.).From the acetone-benzene filtrate the excess of acetone is distilled off, and to the residual liquor sufficient alcohol (97 per cent.) is added t o cause a slight precipitation. The mixture is then poured into warm alcohol when a pink pasty mass separates which on stirring vigorously disintegrates into a faintly pink flocculent substance. The s&pension in alcohol is allowed to remain twenty-four hours, collected ground up with a further quantity of alcohol again collected and washed. The dried mass weighs 70-75 grams and melts at about 175-200° and is conveniently designated (C). All efforts to improve the yields of the two crystalline products have been unavailing. The use of a smaller quantity of acetone causes the potassium permanganate to cake and become coated with manganese oxides.On the other hand the addition of larger quantities of acetone has no beneficial result. The employment of a larger quantity of permanganate also does not improve the yield, whilst if a less amount is used unchanged carbazole remains. Examination of Dicarbazyl (A) melting at 220-221O. 222 PERKIN AND TUCKER THE OXIDATION OF CARBAZOLE. This substance separates from benzene in fine large crystals but when these are removed from the solution and exposed t o the air they a t once commence t o become opaque a t the ordinary temperature and do so more rapidly in the steam-oven falling t o a white powder. The melting point is unaltered by this change, which indicates that dicarbazyl ( A ) crystallises with benzene of crystallisation.Apparently the substance crystallises froin toluene and sylene without added solvent (Found C = 86.7 ; M = 4.9 ; N = 8.4. C24H1FN2 requires C = 56.7 ; H = 4.8 ; N = 8.4 per cent.). Dicarbazyi (A) is soluble in acetone but not as easily as in the other solvents just mentioned. The solution in acetone exhibits a pale violet fluorescence and deposits large crystals possessing a faint steel-blue tint. It is soluble with difficulty in D;crrrhnzyZ (:I). alcohol and in glacial acetic acid but easily so in nitrobenzene and in naphthalene. On account of the peculiar behaviour of dicarbazyl ( A ) towards benzene the molecular weight was determined in naphthalene by the cryoscopic method and gave M=336 and 327. C,H,,N, requires M = 332.The crystallographic examination of dicarbazyl ( A ) melting a t 220-221° was kindly carried out by Miss M. W. Porter with the following result : The crystals from acetone are orthorhombic and show the follow-ing two forms m{llO} ~ ( 1 1 1 ) . The habit is pyramidal as shown in the figure. The mean results of measurement f o r m ( l l 0 ) are azimuth (4) = 60°2/ polar distance ( p ) =- 90'0' ; for p { 11 1 } azimut PERKIN AND TUCKER THE OXIDATION OF CARBAZOLE. 223 ( 4 ) = 60°1’ polar distance ( p ) =55O11’. Axial ratios a b c = 0.5766 1 0.7182. Several attempts were made to acetylate dicarbazyl ( A ) , for example by dissolving the substance (0.5 gram) in acetic anhydride (4 c.c.) and boiling the solution for one hour. On cooling large crystals of unchanged substance (m.p. 219-220O) separated and on pouring the acetic anhydride filtrate into water there was no further precipitate. Again separate portions of 0.5 gram of substance were dis-solved in 4 C.C. of acetic anhydride and to these solutions were added (1) a drop of concentrated sulphuric acid (2) 0.1 gram of fused potassium hydrogen sulphate and (3) 0.1 gram of fused zinc chloride and the solution was in each case boiled for one hour. The solutions in experiments (1) and (3) darkened but (2) remained colourless so the boiling was continued in this experiment for another two hours. No crystalline matter separated from the solutions on keeping but when poured into water (1) and (3) gave a pink solid and a considerable amount of iridescent scales.The addition of water to (2) brought about the precipitation of a white solid possessing a silky sheen. These pre-cipitates were soluble in alcohol but were quite indefinite and all attempts to isolate any satisfactory product from any of them were unsuccessful. Dicarbazyl ( A ) dissolves in cold concentrated sulphuric acid to an emerald-green solution which on warming turns dirty green, but addition of a drop of concentrated nitric acid t o the solution in sulphuric acid turns the colour to a deep blue. The solution in toluene treated with concentrated nitric acid in the cold and allowed t o remain deposits yellow crystals which do not melt a t 330O. Bromine reacts violently with dicarbazyl ( A ) as i t does with carbazole but the products have not yet been investigated.When fused with anhydrous oxalic acid dicarbazyl ( A ) yields a deep green melt which finally turns blue. No picrate could be obtained by treating either the benzene, acetone or alcoholic solution of dicarbazyl ( A ) with picric acid. If carefully heated dicarbazyl ( A ) melts without decomposition, but finally chars and gives a white sublimate different in appear-ance from the fluffy sublimate of carbazole. Examination of the Dicarbaxyl (B) melting at 265O.-some difficulty was experienced in the isolation of this substance from the crude material melting a t 252O (see p. 221) but a pure pro-duct was obtained by dissolving the crude substance in hot benzene 224 PERKIN AND TUCKER THE OXIDATION OF CARBAZOLE pouring into twice the volume of cold acetone and allowing to remain until the fine crystalline deposit had completely separated.Several repetitions of this process followed by a crystallisation from acetone in which the substance is only slightly soluble gave small bright white crystals melting without preliminary brown-ing a t 265O (270O corr.) to a clear yellow melt which almost immediately blackens. Subsequently in an attempt to acetylate this dicarbazyl ( B ) by boiling with acetic anhydride it was found that it crystallises well from this reagent and the melting point may be raised in this way to 268O ( 2 7 3 O corr.). Dicarbazyl (B) also crystallises from nitrobenzene aniline pyridine toluene and xylene (Found : C=86-7; H=4*8; N=8*1. The molecular weight by the cryo-scopic method in benzene gave M=330.C,HI,N2 requires (2 = 86.7 ; H = 4.8 ; N = 8.4 per cent. Attempts to acetylate dicarbazyl ( B ) under a variety of conditions gave results very similar to those obtained in the case of dicarbazyl ( A ) and no definite product could be isolated in any case. Dicarbazyl (B) is insoluble in cold concentrated sulphuric acid, but on warming dissolves with a dull green colour unaffected by the addition of a drop of concentrated nitric acid. Bromine reacts violently with the substance. Fusion of the substance ( B ) with anhydrous oxalic acid gives a faintly blue melt different in tint from that of carbazole-blue. Molecular weight = 332 .) . Attempts to prepare a picrate were unsuccessful. The Amorphous Substance (C) ,-All attempts to crystallise the flocculent faintly pink material precipitated by addition of alcohol t o the benzene filtrates of ( A ) and ( B ) as explained on p.221, have proved futile. By several repetitions of this process or by the addition of alcohol to the ethereal solution a cream-coloured, amorphous powder having a peculiar silky sheen is obtained but the amount is small. The substance (C) is very soluble to red-coloured solutions in benzene toluene pyridine chloroform and in hot ethyl acetate ; it is also somewhat soluble in hot methyl and ethyl alcohols glacial acetic acid acetic anhydride aniline and ether less so in acetone, light petroleum and in amyl alcohol; it is insoluble in dilute sodium hydroxide ammonium hydroxide and concentrated hydrochloric acid solutions.The substance (C) is precipitated from its solutions by alcohol water and by acetic acid. Crystal-lisation could not be effected from any of these solvents or from mixtures of them. Soxhlet extraction experiments using acetone, glacial acetic acid and methyl alcohol failed to separate any con-stituent other than traces of the dicarbazyls ( A ) and (B) THE INFLUENCE OF MERCURY SULPHUR ARSENIC ETC. 225 As (C) could not be obtained in a sufficiently definite crystalline condition no further work has in the meantime been done with it. It dissolves in concentrated sulphuric acid with a green colour, and does not give a picrate. The substance reacts violently with bromine and readily with nitric acid to give yellow derivatives. Summary. By the action of potassium permanganate on carbazole in boil-ing acetone solution three substances are produced two crystal-line compounds melting a t 220-221° and a t 265O respectively and an amorphous substance. The two crystalline compounds are isomeric dicarbazyls and combustion analyses and molecular weight determinations in naphthalene and in benzene show that they have the molecular formula C,,HI6Nz. All three substances can be nitrated and brominated and react under certain conditions with acetic anhydride. They do not form picrates. The authors desire to express their indebtedness to the Depart-ment of Scientific and Industrial Research for a grant which has enabled this research to be carried out and also wish to thank Miss M. W. Porter and Mr. T. V. Barker for the crystallographic measurements contained in this paper. ORGANIC CHEMISTRY LABORATORY, OXFORD. [Receiued January 2l& 1921.

ISSN:0368-1645    

DOI:10.1039/CT9211900216

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

THE INFLUENCE OF MERCURY SULPHUR ARSENIC ETC. 225 XXVI1.-The InJ~erace of Nercury SuZphw Arsenic, and Zinc on the Catalytic Activity of Platinum. By EDWARD BRADFORD MAXTED. IN a previous communicatioii (T. 1920 117 lSOl) it has been shown that the inhibitive effect of varied proportions of lead on the catalytic activity of platinum is within certain limits directly proportional to the concentration of the inhibitant. In view of this result and with the object of examining the general applica-bility and limits of a linear poisoning law the work has now been extended to other inhibitants. I n order to eliminate the necessity of obtaining the value repre-senting the activity of the catalyst from the equation to the absorp-tion curve by a process of differentiation great care has been given VOL.CXIX. 226 MAXTER THE INFLUENClC OF MERCUIGY SULPHUR ARSENIC!? to the purity of all materials employed) for the measurements. Absorption curves were thus obtained of the form which Armstrong and Hilditch (Proc. Roy. SOC. 1919 [ A ] 96 137 322) have recently shown to be typical of pure substances this form being found also to be unaltered by the presence of inhibitants of the type studied. Froin these curves the catalytic activity of the preparation could be read off directly without recourse to mathe-matical treatment. The conditions have thus been rendered more precise and i t has been possible to plot with considerable accuracy the form and limits of the poisoning curves. As before the variation of the catalytic activity of the platinum has been followed by measuring its activity when employed for catalytic hydrogenation under standard conditions.I n general, the existence of a linear poisoning law has been coqfirrned. The limits between which this holds woulld appear to be from a zero concentration of inhibitant up to a region in the neighbourhood of total extinction where a point of inflexion occurs from which the poisoning curve slopes far less steeply towards complete inactivity. The existence of this small residual activity which is thus more resistant to the action of poisons than the main por-tion of the original activity possessed by the catalyst is of con-siderable theoretical interest and renders the poisoning curve somewhat analogous in form to that of the absorption curves themselves in which the predominating linear portion is followed by a point of inflexion and a sudden decrease in the rate of absorption of gas.EXPERIMENTAL. The oleic acid which was employed as a standard unsaturated substance for measuring the catalytic activity of the various prepar-ations was owing to difficulties in obtaining supplies of acid of sufficient purity prepared by shaking purest commercial olive oil with hot distilled water containing a small quantity of sodium hydroxide as in the usual procedure for refining oils by means of alkali. The aqueous phase was rejected and after frequent washing with boiling water the oil was dried and treated a t looo with fuller’s earth the latter treatment being repeated several times. Hydrolysis was next effected in the usual manner by means of sodium hydroxide in dilute alcoholic solution and the almost completely colourless fatty acid obtained was after renewed wash-ing with boiling water and subsequent drying in a current, of nitrogen found to be sufficiently free from impurities to give an absorption curve of the form described.In view of the requirements merely for a homogeneous “ catalytically pure, AND ZINC ON THE CATALYTXC ACTfVfTY OF PLATINUM. 227 unsaturated) substance i t was not considered necessary to eliminate, by subsequent treatment the small quantities of homologues of oleic acid which are present in olive oil and consequently in the oleic acid obtained. It was found possible to obtain both platinic oxide and glacial acetic acid of sufficient purity the properties of the solvent being found t o be practically unaffected from a catalytic point of view, by subsequent purification in the laboratory.In order to avoid the possibility of complications due to the Time in minutes. coagulation of a colloidal metal by electrolytes the catalyst was employed in a non-colloidal form. The platinum was however, in view of the greater susceptibility of catalysts of relatively high activity t o the action of poisons prepared in a condition as active as possible by carefully grinding the platinic oxide in a n agate mortar in order t o expose a large catalytic surface the subsequent reduction to metal being effected a t a relatively low temperature. Form of t h e ,4 bsorption Curve. Previous to the recent work of Armstrong and Rilditch (loc.c i t . ) the course of the catalytic hydrogenation of liquids in the I 228 MAXTED THE IXFLUENCE OF MERCURY SULPHUR ARSENIC, presence of unpoisoned catalysts had been supposed to follow the unimolecular law any deviation from this being attributed to the presence of impurities. These investigators have shown that with increasing purity of the reacting substances absorption of hydrogen takes place more rapidly than is required by a unimolecular reac-tion and that the true form for reactants in a state of purity is that to which reference has already been made. The present measurements in the absence of catalyst poisons, confirm Arnistrong and Hilditch’s observations and the general form of the reaction curve would appear also to be unaffected by the presence of inhibitants of the nature studied this being how-ever not the case with “clogging” poisons probably of an albuminoid or resinous nature such as are found in oils and other organic liquids of ordinary purity.The independence of the form of the reaction curve on the presence of catalyst poisons of the first class is illustrated by the results collected in the figure which summarises the course of the absorption of hydrogen by a system consisting of 3 C.C. of oleic acid dissolved in 9 C.C. of acetic acid, both alone and in the presence of the poisons indicated the catalyst consisting of 0.005 gram of platinum. The linear character of this curve is further exemplified by the various results detailed in connexion with the work about to be described.The measurements of the rate of absorption of hydrogen were in every case carried out a t 50° in the apparatus previously employed for lead. It will be seen that both with and without poisons the point of inflexion occurs in a region immediately preceding saturation . Action of Mercury. The solution employed for poisoning was made by dissolving 0.108 gram of pure mercuric oxide in 100 C.C. of glacial acetic acid. Each C.C. thus corresponded with 1 mg. of mercury. On adding varied amounts of this solution to a reaction mix-ture consisting of 3 C.C. of oleic acid 0.005 gram of platinum and sufficient glacial acetic acid to amount with that added with the poison t o 9 C.C. of the solvent the following results were obtained. The temperature both for this series and for all measurements recorded in the present paper was 50°.The absorption curves are in each case approximately linear, and the activity can consequently be read off directly without differentiation. As before the relative activity of the platinum, in the presence of the given concentration of inhibitant has been taken as being proportional to the number of C.C. of hydrogen absorbed per minute. I n cases where the reaction curve obtaine AND ZINC ON THE CATALYTIC ACTIVITY OF PLATINUM. 229 c c . of Hg solu-tion added. 0.0 0.25 0.5 0.75 0.95 1.25 1.6 2.0 3.0 TABLE I. Hydrogen absorption in C.C. after 1 min. 2 mins. 3 mins. 21.0 43.6 67.0 16.0 32.5 49.1 10.8 21.8 33.0 5.6 11.8 17.5 2.8 5.2 7.6 2.2 4.0 6.0 1.3 2.8 4.3 1.0 1.8 2.6 0.1 0.3 0.4 4 mins.5 mins. 90.0 113.0 65.6 81.6 43.6 54.6 23.0 20.4 10.2 12.6 8.0 9.8 5-8 7.2 3.3 3.8 0.6 0.8 7 10 mins. 185.0 156.4 104.6 51.2 24.8 13.9 1.8 --has been found to deviate subsequently from the ideal linear form, the rate of absorption of hydrogen during the first few minutes has been taken as being representative of the activity of the platinum. The variation of catalytlic activity with the mercury content of the system is given in table 11. TABLE 11. Mercury con tent of system in milligrams. 0.0 0-25 0-5 0.75 0.95 1.25 1.5 2.0 3 0 Ratio gram-atom Hg to gram-atom Pt. 0.0 0.049 0-097 0.14G 0.185 0-244 0-292 0.390 0.585 Catalytic activity of platinum in terms of C.C.H fixed per min. 22.5 16.5 11.0 5.7 i.6 (point of inflexion) 2.0 1.4 0.9 0.15 If the above results are plotted graphically the inhibitive effect is seen to be a linear function of the mercury present in the react-ing system almost down to complete extinction of activity in which region a point of inflexion occurs. The residual catalytic activity then decreases less rapidly with increasing mercury con-tent than is required by the linear law. With sulphur this point of inflexion would appear to be somewhat lower than with mercury, whilst arsenic and zinc deviate from the linear course a t a slightly higher point. I n the case of lead under the conditions previously employed, the presence of a similar point of inflexion in the neighbourhood of total suppression of activity was not noted but i t is possible that the existence of a small residual activity was masked b 230 MAXTED THE INFLUENCE OF ETBRCURY SULPHTTR ARSENIC, “clogging” poisons by the action of which catalytic activity of a relatively low magnitude would be quickly suppressed.Action of Sulphur. The poison was here employed in the elementary form dissolved in glacial acetic acid in which solvent sulphur is slightly but sufficiently soluble. Each C.C. of the solution contained 0.069 mg. of sulphur. The progress of the absorption of hydrogen by a reacting system of the usual nature in the presence of various concentrations of the inhibitant is suniimarised in table 111. T ~ L E 111. C.C.of sulphur solution added. 0.0 0.2 0.4 0.8 1.2 1.6 2.0 2.2 2.7 2.9 3.5 N 7 1 min. 21.0 20.0 18.2 15.0 10.8 5.5 1.6 1.0 0.6 0.5 0.2 Hydrogen absorption in C.C. after -_ - - ~ ~- _ -2 mins. 3 mins. 4 mins. 5 mins. 43.6 67.0 90.0 113.0 39.5 60.0 80.6 100*5 36.2 53.8 70.0 86.0 29.2 42.5 55.0 66.5 21.8 32.8 42.6 52.1 11.8 18.1 “4.1 “9.8 3.0 4.6 6.1 7.4 2-1 3.2 4.4 5.5 0-9 1.2 1-5 1.7 1.0 1.5 1.9 0.4 0-6 0.7 0.9 -_ ~ _ _ 10 mins. 185.0 182.0 162.0 118.5 93.1 52.5 13.7 10-8 3.2 -_ The variation of catalytic activity with the concentration of the inhibitant is given in table IV. TABLE IV. Sulphur content of system in milligrams. 0.0 0.014 0.028 0.055 0.083 0.110 0.138 0.154 0.186 0.20 0.24 Ratio gram-atom S to gram-atom Pt.0.0 0.017 0.034 0.087 0.101 0.134 0.168 0-188 0.228 0.244 0.293 Catalytic activity of platinum. 22.5 20.0 18.0 15.0 10.8 6.0 _ . 1.5 (point of inflexion) 1-1 0.5 0.5 0.2 The above results when plotted graphically form a curve similar to that obtained for mercury AND ZINC ON THE CATALYTIC ACTIVITY OF PLATINUM. 231 Action of L4rse7tic, The inhibitive solution was made by dissolving 6.6 milligrams of arsenious oxide in a small quantity of distilled water and diluting to 50 C.C. with acetic acid. Each C.C. thus contained 0.1 mg. of arsenic. A series of measurements of the rate of absorption of hydrogen by 3 C.C. of oleic acid in the presence of 0.005 gram of platinum and various known concentrations of arsenic is summarised in table V the experiments being carried out under the usual conditions.TABLE V. C . C . of arsenic Hydrogen absorption in C.C. after solution added. Cmin. 2 mins. 3 mins." 4 mins. 5 r n i n s x i n s . 0.0 0.25 0.5 0.75 1.0 I .25 2.5 2.25 9.5 3.0 3.5 20.0 17.2 12.5 11.5 9.0 6.6 4.9 2.0 1.3 0.6 0.3 40.1 35.0 26-0 23.0 18.0 13.0 10.0 4.0 2.0 1.1 0.5 59.8 52-3 39.7 34.5 27-5 19.0 14.9 5.8 3.4 1.6 0.7 79.0 69.8 53.6 45-5 37-0 24.7 19.5 7.9 4.2 1.9 0.8 96.4 86.0 67.5 56.5 46.0 30.0 23.8 9.7 4.9 2.2 1.0 181.0 170.5 134.2 107.0 88.7 48-0 39.6 -The variation in catalytic activity of the platinum with increas-ing arsenic content is given in table VI the poisoning curve being of the usual linear type.T~BLE VI. Arsenic content of system in milligrams. 0.0 0.025 0.05 0.075 0.10 0.125 Ratio gram-atom As to gram-atom Pt. 0.0 0.01 3 0-02 6 0.039 0.052 0.065 Catalytic activity of platinum. 20.0 17.5 13-7 11.5 9.2 6.6 0.15 0.078 4-9 (point of inflexion) 0.225 0.1 17 2.0 0.25 0.13 1.1 0.30 0.156 0.6 0.35 0.152 0.3 Action of Zine. zinc oxide in 100 c.c of glacial acetic acid. The zinc solution was made by dissolving 8.3 milligrams of pure Each C.C. thus oorre 232 THE INFLUENCE OF MERCURY SULPHUR ARSENIC ETC. sponded with 0.067 ing. of zinc. 'The absorption of hydrogen by 3 C.C.of oleic acid under the usual conditions is summarised in table VII. TABLE VII. C.C. of zinc: solution added. 0.0 0.5 1.0 1.5 1.8 3.0 4-0 Hydrogen absorption in C.C. after r A \ 1 min. 2mins. 3 mins. 4mins. 5mins. 10 mins. 20-0 40.1 59.8 79.0 96.4 181.0 14-4 28.5 41.7 53.7 65.4 126.7 9.8 18.8 28.0 37.2 46.2 90.4 4.7 9.2 13.7 18.2 2 2 - 5 41.7 3.6 7.0 10-3 13.5 16.6 I 0.5 0.9 1.3 1.7 2.1 0.0 0.2 0-3 0.4 0.4 I -The variation in the activity of the platinum catalyst with the various concentrations of zinc is sunimarised in table VIII. The poisoning curve is of the usual linear form the point of inflexion being however somewhat higher both for zinc and arsenic than for mercury and sulphur. TABLE V I I I .Zinc content of system in milligrams. 0.0 0.034 0.067 0.10 0.12 0.20 0-27 Ratio gram-atom Zn to gram-atom Pt. 0-0 0.02 0.04 0.0 6 0-07 1 0-12 0.16 Catalytic activity of platinum. 20.0 14.3 9.2 4.5 (point of inflexion) 3.6 0.4 0.1 It has previously been mentioned that in order to avoid the complication which would be introduced by the possible coagula-tion of a colloid in the presence of an electrolyte non-colloidal in place of colloidal platinum was employed for the present measure-ments. I n view however of the rather indefinite line of demarca-tion between colloidal and finely divided non-colloidal particles, such as are formed by the reduction of carefully ground metallic oxides in an oil suspension it was considered interesting to investigate in addition the influence of an acetate containing a metallic ion which from its nature would not be expected to be poisonous.It may perhaps be stated however that in no case did visible flocculation occur during any of the measurements described in the present paper although this frequently occurring pheno-menon has been noticed in another series carried out a t a higher temperature OX REDUCTION BY METALS I S ACID SOLUTIONS. PART I. 233 The indifference of the ctrtlytic activity of platinum to the presence of a metallic acetate of this nature even in concentratlion somewhat in excess of those employed for the poisoning experi-ments is illustrated by the following measurements of the relative rates of absorption of hydrogen by 3 C.C. of oleic acid and 0.005 gram of platinum with and without the addition of potassium acetate. The experiments were carried out at 50° under the usual conditions. The potassium acetate solution was made by dis-solving 50 milligrams of the salt in 100 c . ~ . of acetic acid 1 c . ~ . of this solution being added t o the reacting system. Reference t o table I X shows that no inhibition of activity occurred. TABLE IX. Time in minutes . 1 2 3 4 5 6 7 Hydrogen absorption in C.C. System without System with potassium acetate. potmsium acetate. 23.6 24.0 48.0 49.0 72.0 75.0 96.3 100.4 121.0 127.0 146.0 153.0 168-0 178.0 A \ It is a matter of experience that catalysts of relatively high activity are more susceptible to poisoning than those of lower activity and for this reason work is being carried out on the influence of variations in the surface of the catalyst on the slope of the various poisoning curves. CHARLES STREET, WALSALL

ISSN:0368-1645    

DOI:10.1039/CT9211900225

出版商:RSC    

年代:1921

数据来源: RSC

摘要:

OX REDUCTION BY METALS I S ACID SOLUTIONS. PART I. 233 XXVIII-On Reduction by Metals in Acid Solu-tions. Part I. The Reduction of Acid Ferric Sulphate Solutions by Zinc and Magnesium. By SAMUEL SUGDEN. THE reduction of ferric sulphate solutions containing free sulphuric acid by zinc magnesium and iron was studied by Thorpe (T., 1882 41 287) who found that the amount of reduction brought about by the metal increased as the ferric sulphate concentration increased and ,decreased with increasing concentrations of free I 234 SUGDEN ON REDIJCTIOX BY METALS IN acid. The solutions examined were moderately dilute arid only covered a small range of concentrations. I n the present paper similar experiments are described covering nearly the whole available range of concentrations.These reveal a marked difference between the behaviour of zinc and magnesium. E X P E R I M E N T A L . Materials.-Zinc foil 0.0107 cm. thick cut into pieces 0.19 cm. by 9-72 cm. was used; 0.3268 gram was weighed out for each experiment. The initial surface area of this amount of the met,al was calculated to be 9.3 sq. ern.%- The foil contained 99.7 per cent. of zinc weighed as oxide. A trace of lead was present but no other metallic impurities were detected. The magnesium was in the form of ribbon 0.0158 cm. thick. This was cut into pieces 0.27 cm. by 0.70 cm.,-f and 0.1216 gram used in each experiment. The initial surface area was calculated to be 9-5 sq. cm. The ribbon contained 98-7 per cent. of mag-nesium weighed as pyrophosphate. A minute amount of silica was found on examination of a solution of the metal in nitric acid but no metallic impurities.Considerable difficulty was experienced in purifying the ferric sulphate in large amounts. The method finally adopted was as follows. Concentrated sulphuric acid was added to a filtered hot solution of 1 kilo. of the commercial salt in 1 litre of water until a faint turbidity was produced. After keeping for several hours, the viscous mother liquor was filtered from the granular crystals of the acid sulphate which were washed with alcohol and dried in air. About G kilo. of the hydrated acid sulphate were prepared in this manner. A stock solution was prepared by dissolving the acid sulphate in water and addiiig the requisite amount of ferric hydroxide.This solution corresponding in composition with the neutral or slightly basic salt was analysed and the experimental solutions prepared from it. The sulphuric acid used in making the solutions was free froni chlorides and nitrates and contained only minute traces of arsenic. Experimental Condition,s.-Fifty C.C. of the solution were placed in a small flask the weighed amount of the metal added and the flask shaken frequently until the metal had all dissolved. The The surface area is determined chiefly by the thickness and is not appreciably altered by small variations in the length and breadth Small pieces were used to facilitate stirring. * Average value for 200 pieces chosen a t random. 7 Average value for 200 pieces chosen at random ACID SOLUTIONS. PART I.235 time required for complete solution was noted andl the ferrous salt formed estimated by titration with N / 10-permanganate solution. All the experiments were carried out in duplicate a t the ordinary temperature (18-210). With most of the solutions the results were found to be very consistent the difference between the values found in #duplicate experiments being less than 1 per cent. In some of the experiments with zinc which lasted for several hours (solutions 5 6 S) somewhat larger differences were found and a small amount of oxidation by air may have occurred. The mean FIG. 1. 7 SO gram-rnolmdes pjr Eitre. Magnesium. values found for each solution are recorded in table I and plotted in Figs. 1 and! 2. Discussion of Results. Each curve in the diagrams represents a series of solutions con-taining a constant amount of Fe,O and varying amounts of SO,.The ordinates represent the amount of reduction observed ex-pressed as a percentage of that theoretically possible if no hydrogen were evolved. To save space this number is referred to in the table and in the text as the “percentage reduction,” as the use of I* 236 SIJGDEN O N REDTJCTIOX EY METALS IN this term does not imply aiiy hypothesis regarding the mechanism of the reaction. The left-hand end of each curve therefore repre-sents a neutral or a feebly acid solution in which the metal dis-solves slowly. The curves were extended t o the right the limits corresponding with solutions froin which crystals of the acid sulphate were deposited or for the more dilute solutions in which FIG.2. 9 1 Y 3 4 5 6 7 SO gram-molecules per litre. Zinc. R concentratioii of 6 grain-molecules of free acid per litre was reached,. With both zinc and magnesium;the addition of acid to a nearly neutral solution produces a decrease in the percentage reduction. With magnesium this continues throughout the whole range of concentrations but with zinc a minimum is reached and then th ACID SOLUTIONS. PART I. 2.37 percentage reduction increases rapidly. A new type of reaction appears to set in; the zinc becomes dull grey in colour and dis-solves much more slowly with a scarcely perceptible evolution of gas. This peculiar behaviour of zinc is most marked with the solutions containing higher concentrations of ferric oxide but even the most dilute solutions give curves which differ in shape from the corresponding curves for magnesium.The gradual replace-ment of the rapid reaction by the slow one is well seen in solutions 21 and 22. The zinc dissolves rapidly a t first changes in appear-ance after a few minutes and then dissolves slowly with very little evolution of gas. I n this connexion i t should be mentioned that Pring and Tainton (T. 1914 105 710) in studying the electro-deposition of zinc from solutions of zinc sulphate containing about 1 gram-molecule of free sulphuric acid per litre found that the ratio of zinc t o hydrogen liberated at the cathode increased with increasing concentrations of acid. The time required for solution gives an approximate estimate of the relative rates of reaction of the metals in the different solu-tions.Remarkable variations in this rate were observed. Thus the zinc which dissolved in an hour in a solution containing 0.80 gram-molecule of ferric oxide per litre and rather less acid than was required for the neutral salt was not completely dissolved in twenty-four hours in a solutioii containing the same amount of ferric oxide and 2 gram-molecules of free acid per litre. This behaviour cannot be ascribed to the presence of so much free acid, because solutions containing much more acid but less ferric salt, dissolve the same amount of zinc in a few minutes. With magnesium the time required for solution falls off con-tinuously as the concentration of free acid increases and finally reaches a limiting value of just less than a minute with all con-centrations of ferric salt.It is difficult t o reconcile these results with the theory that nascent hydrogen owes its activity to the presence of free atoms, as on this view the activity should decrease continuously with increasing acidity for both metals. The theory of direct reduc-tion by the metal is also inadequate for even if the assumption is made that zinc can directly attack the ferric sulphate in weakly acid solutions at a speed comparable with its rate of solution in the acid the existence of a perturbing condition has to be recog-nised in the case of zinc in the more concentrated solutions which not only modifies the relative velocities of the two reactions but also lowers markedly the velocity of solution, It is hoped t o extend these experiments to other metals an 238 ON REDUCTION BY METALS I N ACID SOLUTIONS.PART I. oxidising agents and to investigate the effect of varying the temperature and pressure. TABLE I. Fe203 803 gram-mols. gram-mols. Soln. per litre. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 2 0 21 22 23 2'4 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 I per litre. 2.25 3.40 c ) . 65 2.90 3.15 340 3.90 4.40 5.40 1.93 2.30 2.60 2.80 2.93 3.05 3-18 3.30 3.55 3.80 4.30 4-55 4.80 1.43 1-73 2.98 2.20 3.73 3.20 3.70 4.12 4.34 4.78 5-66 6.65 0.72 1-10 1.60 2.60 4.60 6-00 0.80 1.30 2.30 4.30 6.30 BIRRBECK COLLEGE, BREAM'S BUILDINGS, FETTER LANE E.C.4. Magnesium. Zinc. 1 Percentage Time Percentage reduction. mins. reductio 11. 60.6 58.5 55*(i 50.6 44.0 39.0 31.2 2 1.8 11.4 -------------60.7 42.0 36.5 3 1.0 25.3 20.0 16.6 12.4 9.0 4.9 2.1 0.4 44-0 30.7 20.7 13.2 2.5 1.5 17.0 11.1 6.9 1-5 0.8 - -6 0 88.1 90 s5.9 400 8 7 . 9 @a. 1000 88.7 ca. 2000 Y I . ( i ca. 3000 92.3 - -- -13 75.2 6 63.8 6 57.9 4 51.3 3 47.3 9 45.6 I9 43.8 9 9 42.1 J 9 41.1 9 7 43.4 48.5 2;; 58.5 9 67.5 30 66.1 6 54.3 5 46-2 3.5 40.0 3 32.7 2.5 27.2 1 25.2 - -2 25.6 3 25.6 60 28.4 60 61.6 32 42.7 27 35.0 14 96.0 9 18.3 20 17.6 GO 30.0 40 22.0 32 15.9 16 11.4 I1 9-2 [Received January 17th 1921.

ISSN:0368-1645    

DOI:10.1039/CT9211900233

出版商:RSC    

年代:1921

数据来源: RSC