338 WATSON : 6’-AMINOQUERCETIN.XXXVl I.-6’-Aminopuercetin.By EDWIN ROY WATSON.QUERCETIN is a natural hydroxy ketonic dye of considerable valueon account of its cheapness and the fastness of the shades obtainedwith it on suitable mordants. Other members of the flavone groupare also valuable for similar reasons. The chief limitation of thegroup seems to lie in the fact that the colour never goes beyondyellow, orange, or brown shades, It seemed, therefore, of interestand value to attempt to prepare derivatives which, whilst retainingthe fastness of the original dyes, would yield deeper shades, such asred, violet, or blue.The first method which suggested itself for fulfilling this objectwas the introduction of additional auxochrome groups. Thismethod has proved very useful in the alizarin group of dyes andthere #&re many cases where the addition of one or two more auxo-chromes causes a great deepening of colour (compare brazilein andhaematein, fluorescein and gallein). There were, certainly, reasonsfor not expecting too much from the addition of more auxochromesin this group.Myricetin and quercetin have not appreciablydeeper shades than luteolin, although they contain two and oneadditional hydroxyl groups respectively. It could be argued, how-ever, that much might depend on the positions of the auxochromes,thus, anthracene-blue, W.R., 1 : 2 : 4 : 5 : 6 : 8-hexahydroxyanthra-quinone, dyes blue shades on chrome and rufigallol, 1 : 2 : 3 : 5 : 6 : 7-hexahydroxyanthraquinone, gives only brown shades on the samemordant.A t any rate, it appeared worth while t o introduce anadditional auxochrome into quercetin and observe the effect onthe colour of the dye. The preparation of a hydroxyquercetinpossessed the further interest that i t might prove identical withmyricetin, gossypetin, o r quercetagetin, and thus confirm or helpin arriving a t the constitutional f ormulz of these substances.By the complete methylation of quercetin to protect the molecule,nitration, reduction, and subsequent demethylation, 6’-aminoquer-cetin was obtained. Attempts to convert the amino-group into ahydroxyl group by diazotisation and boiling .with water or diluteacid have not as yet been successful. 6/-Aminoquercetin dyesalmost the same shades as quercetin, and its preparation did notthus effect the primary object of this investigation. The subsequentinvestigation of quercetagetin and gossypetin (A.G . Perkin, T., 1913,103, 209, 650) seems t o show that the multiplication of auxochromegroups, whatever positions they may occupy, will not appreciablydeepen the colour in this group of dyesWATSON : 6’-AMINOQUERCETIN. 339Further work on somewhat different lines has been more suc-cessful and has resulted in the preparation of derivatives of quer-cetin which dye violet and blue shades. These will be describedin a subsequent communication.I n the courso of this investrigation, several oxonium salts ofquercetin pentamethyl ether were isolated, namely, the hydrochloride,C2,H2,O7,HC1, the hydro bromide, C,,H,,O,,HBr, the sulphate,C20H2007,H2S0,, and the nitrate, C,,H,,O,,HNO, ; also di bromo-pzcercetin pentamethyl ether hydro bromide, C2,H,807Br,,HBr.These compounds are all of a bright yellow colour a.nd easily pre-pared.Their formation wras rather unexpected, as, although quer-cetin itself readily gives such bright yellow oxonium salts, pro-gressive methylation seems to reduce the tendency to form com-pounds of this character. Rhamnetin and rhamnazin yieldsulphates with difficulty, but no compounds with the halogen acids(A. G. Perkin, T., 1896, 69, 1439) and quercetin tetramethyl etheronly forms a very unstable sulphate. Their bright yellow colourindicates tihat they should be assigned a quinonoid structuresuch as:0 0-CH,These compounds seem somewhat comparable with the hydro-chloride of fluorescein dietliyl ether which is of an intense yellowcolour and decomposed by water (Nietzki and Schrijter, Ber., 1895,28, SO), and most, if not all, become bright red on the surface ifleft exposed to the air, although, f o r the present, no explanationof this behaviour can be offered.EXPERIMENTAL.Preparation of Quercetin Pentamethyl Ether.-This was effectedby a modification of Valiaschko’s process (Arch.Pharm., 1904, 242,242; Ber., 1909, 42, 727). Herzig and Hofmann’s simpler process(Ber., 1909, 42, 155) was tried without success. I n preparingquercetin trimethyl ether from quercetin by treatment with potass-ium hydroxide and methyl sulphate in methyl-alcoholic solution,i t was found necessary to add the methyl sulphate and then thealcoholic potassium hydroxide as quickly as possible to the boilingsolution, instead of gradually, as recommended by Valiaschko.For the later stage of the process, namely, conversion of the tri-methyl ether into the pentamethyl ether through the potassiu340 WATSON : 6'-AMINOQUERCETIN.salt, i t was found necessary to leave a little excess of potassiumhydroxide in contact with the potassium salt.6f-Nitroyuercetin Pentamethyl Ether, C,,H,O,(OMe),*NO,.-Fivegrams of finely-powdered quercetin pentamethyl ether were addedgradually to 50 C.C.of cold nitric aGid (D 1.4 which had been boiledwith carbamide nitrate), in which it dissolved to a clear yellow-brown solution. This was poured slowly into 24 litres of coldwater, kept stirring, and the mixture was then boiled until thegelatinous precipitate changed into bright yellow, needle-shapedcrystals.After cooling, the crystals (5.15 grams) were collected,washed with water, and dried. The substance may be renderedpure by one or two recrystallisations from acetone, from whichsolvent i t is deposited in golden-yellow, needle-shaped crystalsmelting a t 202-204O. It can also be conveniently crystallisedfrom ethyl alcohol, methyl alcohol, ethyl acetate, or benzene; isvery soluble even in cold chloroform or acetic acid; and is insolublein ether, carbon disulphide, o r light petroleum. It dissolves incold sulphuric acid to a deep reddish-brown solution ,and is pre-cipitated unchanged on pouring this solution into water.0.1 134 gave 0.2405 CO, and 0.0463 H,O.0.1487 ,, 4.7 C.C.N, (moist) a t 14O and 769 mm. N=3*76.0.1516 ,, 0.4187 AgI. Me=17.61.C = 57.83 ; H = 4.53.C,,H,,09N requires C = 57.55 ; H = 4.55 ; N = 3.35 ;Me = 17.98 per cent.The above-described method of nitration is the only one by whicha good yield of the mononitro-compound has yet been obtained. I nsmall quantity (up to 10 per cent.), together with oxiaation pro-ducts, it was formed by the action of boiling dilute aqueous nitricacid on quercetin pentamethy1 ether or by nitrating this substancein the cold in glacial acetic acid solution.6'-Aminopuercetin Pentamethyl Ether, Cl,H402(OMe)5*NH2.-The preceding nitro-compound was readily reduced by tin andalcoholic hydrochloric acid, yielding the hydrochloride of 6/-amino-quercetin pentamethyl ethpr.Five grams of the finely-powderedmononitro-compound were mixed with 150 C.C. of alcohol, 15 C.C.of concentrated hydrochloric acid and 5 grams of finely granulatedtin, and the mixture kept nearly boiling. After some time, another15 C.C. of hydrochloric acid and 5 grams of tin were added. I nabout two hours, reduction was complete, and the fine, white, needle-shaped crystals produced were collected, dried, and dissolved in2 litres of boiling water to which a little dilute hydrochloric acidhad been added. The solution was filtered hot and treated at oncewith sodium carbonate solution until alkaline, which precipitatedthe free base as a sandy deposit of rhombic crystals.3.5 GramWATSON : 6'-AMINOQUERCETIN. 341were thus obtained, being a yield of 70 per cent. on the weight ofnitro-compound taken. No further quantity could be isolated byworking up the mother liquors.The substance can be purified by recrystallisation from benzene,from which solvent i t is deposited in white, stout, heavy prisms,melting a t 200-202O. It can also be conveniently crystallisedfrom hot ethyl or methyl alcohol, acetone, o r ethyl acetate. It isvery soluble even in cold chloroform, but practically insoluble inether, light petroleum, or carbon disulphide.0.1095 gave 0.2493 CO, and 0.0534 H,O.0.1353 ,, 4.7 C.C. N, (moist) a t 30° and 758 mm. N-3.7.0.1523 ,, 0.4553 AgI. Me=19'06.C,,H,,O,N requires C = 62.01 ; H = 5.42 ; N = 3-61 ;Me = 19.37 per cent.Cl,H40,(OMe),*NH2,HC1.-This substance was obtained in fine, white, needle-shaped crys-tals in the reduction of nitroquercetin pentamethyl ether.It w-asalso prepared by boiling the free base with dilute hydrochloricacid. On cooling the solution, fine, white, needle-shaped crystalswere deposited, which were collected and dried in a vacuum oversoda-lime, when they melted and decomposed at 245-247O. Thepure dry compound dissolves in cold water, owing to decompositionof the salt and the formation of a pseudo-solution of the base, but,after a short time, crystals of the free base are deposited.C = 62.09 ; 13 =5'41.6~-Aminopuercetin Pentamethyl Ether Hydrochloride,0.4886 was neutralised by 11.2 C.C.N/lO-KOH; HC1=8*37.6'- A mino qu erce tin Pe ntam e t hyl Ether Sulpha t e,C,,H,,O,N,HCl requires HC1= 8-61 per cent.C1,H402 (OMe) rj*NH,,H,S04-This salt was prepared by dissolving the free base in boilingdilute sulphuric acid ( 1 :20). On cooling the solution, white,needleshaped crystals separated, which melted and decomposed at245-247'.0.2020 was neutralised by 8.2 C.C. nT/lO-KOH; H,SO,= 19.89.C,,H2,O7N,H2SO4 requires H,SO, = 20.20 per cent.6I-Diace tylarnino pzsercetin Pentn me thy1 Ether,C,,H4O,(0Me),*NA~.-Aminoquercetin pentamethyl ether was boiled with excess ofacetic anhydride for several hours. The solution was then filteredhot and, after the addition of alcohol, evaporated on the water-bath. The residue, which almost immediately became solid, wasrecrystallised from alcohol.If necessary, the whole process wasrepeated. The substance crystallises in white prisms, melts a t 185O342 WATSON : 6’- AMINOQUERCETIN.is readily soluble in acetic acid, bu-t less so in alcohol. It has anexceedingly bitter taate, whilst neither pentacetylquercetin, hepbacetylaminoquercetin, nor quercetin pentamethyl ether possessesthis property.0.1249 gave 0.2809 CO, and 0.0596 H20.0.1990 ,, 5.8 C.C. N, (moist) a t 28O and 759 mm. N=3*19.0.2716 ,, 11.6 C.C. N/10-acetic acid; C2H40,=25*6.C = 61.34 ; H = 5.30.C24H2,OSN requires CT = 61.1 ; H = 5.3 ; N = 2.9 ;C2H402 = 25.5 per cent.Incomplete acetylation gave also another substance, less solublein alcohol and precipitated on adding alcohol to the acetylationmixture.This melts a t 223-226O (see P., 1911, 27, 163) and isprobably monoace t ylaminoquerce tin pentam e thy1 ether,C,,H40,(0Me),*NHAc.6/-A mino puerce tin Hydriodide, C,,H,02(0H j,*NH2,HI.-Thissubstance w.as isolated as the chief product in demethylating6/-aminoquercetin pentamethyl ether with hydriodic acid. Thesame reagent effected the simultaneous reduction and demethyla-tion of 6/-nitroquercetin pentamethyl ether and there was noadvantage in isolating the intermediate .aminopentamethyl ether.Five grams of 6/-nitroquercetin pentamethyl ether were gently boiledfor an hour with 150 C.C. of hydriodic acid (D 1.7). Then the mix-ture was boiled vigorously until most of the hydriodic acid wasdistilled off and only 10-15 C.C.remained. Water was added t othe residue, making the bulk up to 750 c.c., sulphur dioxide waspassed in until all free iodine was destroyed, the mixture was boiledand filtered hot to remove a little insoluble matter, and the brightorange-red filtrate was concentrated. On cooling, aminoquercetinhydriodide separated in orange-coloured nodules, was collected,washed with a little aqueous sulphur dioxide, and dried. Threegrams of the deep orange hydriodide were thus obtained. Attemptsto isolate a further quantity by concentrating the mother liquorwere unsuccessful, and i t was found advisable, in later preparations,to work up the mother liquor to give the remainder of the amino-quercetin as hydrochloride. 6’-Aminoquercetin hydriodide isfairly soluble in hot dilute hydrochloric acid, but much less so inthe cold.It is partly decomposed on boiling with water.6l-Amino qzt erc e tin Hydrochloride, C,,H,O,( OH) ,-NH,,HCI .--Forthe preparation of this salt; 6/-nitroquercetin pentamethyl etherwas boiled with hydriodic acid f o r an hour as described in the pre-ceding paragraph. The product w.as then poured into nine timesits bulk of water, freed from iodine by sulphur dioxide or a slightexcess of sodium hydrogen sulphite solution, filtered to remove acertain amount of insoluble matter, and treated with exceas of soliWATSON : 6’- AMINOQUERCETIN. 343sodium acetate. The sodium acetate produced a bulky, gelatinous,yellow precipitate of a sodium salt of aminoquercetin, and this wascollected and washed with cold water as rapidly as possible.Theprecipitate must not be allowed to remain in the air or it will bepartly oxidised to dark-coloured products. It was then dissolvedin boiling dilute hydrochloric acid (1 part of hydrochloric acid to12 parts of water) and the solution concentrated. On cooling,aminoquercetin hydrochloride was deposited in buff-yellow, fine,needle-shaped crystals in radiate aggregates, which were collected,washed two or three times with small quantities of cold dilutehydrochloric acid, and dried in the steam-oven, care being takennot to break up the cake or expose a larger surface than possibleto the air while moist. F o r purification, it was recrystallised fromdilute hydrochloric acid, avoiding any unnecessary oxidation.When free from oxidation products, aminoquercetin hydrochlorideis of a buff-yellow colour, and readily soluble in hot dilute hydro-chloric acid, but much less so in the cold.It does not melt even a t330O. On boiling the dry salt with water, it is partly decomposed.0.1150 gave 4.0 C.C. N, (moist) a t 18O and 755 mm.C,,H,,O,N,HCl requires N = 3-96 per cent.The most successful preparations gave 60 per cent. of amino-quercetin hydrochloride, reckoned on the weight of the nitro-penta-methyl ether taken.6’-Aminopurercetin Sulphate, C,,H,O,(OH),=NH,,H,SO,.-Onsubstituting dilute sulphuric acid for hydrochloric acid in the pre-ceding preparation, there was obtained aminoquercetin sulphate inbuff-yellow nodules or warts.A quantity of sulphate was alsoobtained from the hydrolysis of hepta-acetylaminoquercetin bysulphuric acid in alcohol and, on adding warm water and allowingto remain, the sulphate was deposited in aggregates of fine needles.It melts and decomposes a t 222-227O:N = 3-98.0.1314 gave 3.8 C.C. N2 (moist) at 30° and 756 mm.C,5H,,07N,H,S0, requires N = 3-37 per cent.6’-Aminopuercetin, C,5H402(OH)5*NH2.-Only the followingmethod was found successful for the liberation of the free base fromthe above-mentioned salts:-To one gram of the dry salt (hydro-chloride or hydriodide) were added 15 C.C. of pyridine and 30 C.C.of hot water and the mixture was kept just on the point of boiling.After a short time, crystals began to appear, chiefly a t the surfaceof the liquid and on the sides of the flask above the liquid, andthese, on keeping, increased in quantity.Forty-five C.C. of boilingabsolute alcohol were now added, the mixture once more broughtto the boiling point and allowed to cool, and the crystals whichN=3.12344 WATSON : 6'-AMINOQUERCETIN.separated were a t once collectecl, washed with absolute alcohol, anddried. From 1 gram of the hydriodide, only about 0.3 gram ofthe base was obtained. 6/-Aminoquercetin consists of buff-yellow,fine, needleshaped crystals, almost insoluble in alcohol, acetone,chloroform, ethyl acetate, phenol, or nitrobenzene. No suitablesolvent for its recrystallisation was discovered. It does not melt,but blackens and decomposes about 3 2 0 O .The following analyses were carried out with the substance driedat 150O:0.1171 gave 0.2434 CO, and 0.0401 H,O.C= 56.68 ; H = 3.80.0.1066 ,, 4.4 C.C. N, (moist) a t 24O and 760 mm. N=4.62.C,,H,,07N requires C = 56.78 ; I3 = 3.47 ; N = 4.41 per cent.Position of t h e ,4 ntiiio-group iu A r~ii~iozi'Prcetiti.-Tkis was de-termined, indirectly, by ascertaining the position of the nitro-groupin the nitroquercetin pentamethyl ether from which aminoquer-cetin is obtained. The nitro-pentamethyl ether was oxidised bypotassium permanganate and gave 6-nitroveratric acid,NO2\-/ CO,B/-\O Me,OMeas one of the products. Four grams of the nitro-pentamethyl etherwere dissolved in cold sulphuric acid and reprecipitated by pouringthe solution into a large volume of water.The substance was thusobtained in a finely-divided, amorphous condition. After washing,i t was rubbed into a paste with a cold aqueous solution of 10 gramsof potassium permanganate and allowed to remain over-night. Bythe next morning the pink colour of the permanganate had com-pletely disappeared. The precipitated manganese dioxide wasfiltered off, the filtrate acidified with hydrochloric acid, andextracted with ether. On spontaneous evaporation, the etherealextract left a yellow oil which quickly solidified. This was recrys-tallised from benzene and found to be 6-nitroveratric acid (Ber.,1876, 9, 938).Dyeing Properties of 61-9 mirzo~uerceti~z.-Full dyeings of thefollowing shades were obtained on wool with 4 per cent. of thedyestuff (reckoned on the weight of the wool): pure brown onchrome, yellowish-brown on alum, orange-red on tin, and brownish-black on iron.The shades on chrome and alum were browner, ontin redder and deeper, than those obtained by comparative testswith quercetin.Hepta-a~etyl-6~-arnii~o~uerceti,l., C,,H,O,(OAc),*NAc,. - Thischaracteristic derivative of aminoquercetin was obtained from thefree base or its salts by boiling with acetic anhydride for an hourWATSON : 6’-AMINOQUERCETIN. 345The acetyl derivative was precipitated in crystalline form byadding alcohol. For purification, it was again dissolved inacetic anhydride, boiled for half-an-hour, and precipitated byalcohol. The acetylation of the hydrochloride was effected byboiling together for one and a-half hours 1 gram of the well-driedhydrochloride, 1 gram of fused sodium acetate, and 5 C.C.of aceticanhydride. An equal volume of glacial acetic acid was then addedand the mixture poured into excess of water. The acetyl derivativewas precipitated as a white, amorphous powder, which was collected,dried, and purified by dissolving in acetic anhydride, boiling forhalf-an-hour, and precipitating by alcohol. By either method, itwas obtained in white, rhombic crystals, melting a t 151-153O,sparingly soluble in alcohol, readily so in glacial acetic acid oracetic anhydride. This derivative serves to characterise 6l-amino-quercetin :0.1275 gave 0,2671 CO, and 0*0510 H,O. C =57*12 ; H = 41-44.0.2033 ,, 4.5 C.C. N, (moist) a t 21° and 755 mm.N=2.50.0.4967 ,, 56.9 C.C. N / 10-acetic acid = 68.7.C2,H2,OI4N requires C = 56-96 ; H = 4.09 ; N =2*29 ;C2H40, = 68.7 per cent.If acetylation has not been complete, there is also formed anothersubstance melting above 200° which is probably the hexa-acetylderivative.Quercetin Pentamethyl Ether Diazonium Chloride,C,,H,02(0Me),*N,C1.One gram of finely-powdered 6/-aminoquercetin pentamethylether was treated with 10 C.C. of water and 2.8 C.C. of concentratedhydrochloric acid, warmed and mixed thoroughly to convert thebase entirely into hydrochloride. On adding 9 C.C. of a solution ofsodium nitrite (1 : 50) to the cooled mixture, the stiff, white pasteof hydrochloride became yellow and passed almost entirely intosolution. On scratching the sides of the tube, the diazoniumchloride was precipitated as yellow needles. After two hours, thesewere collected, washed with benzene and dried on a porous tile,and afterwards over soda-lime in a desiccator.This substance isreadily soluble in cold waeer or dilute hydrochloric acid and inalcohol. From solution in the latter medium, it can be precipi-tated in the crystalline form by ether. Its solution in water ordilute hydrochloric ,acid gives a t once a crimson precipitate with analkaline solution of &naphthol. The aqueous solution is fairlystable and gives the diazo-reaction, even after a few minutes’ boiling.VOL. cv. A 346 WATSON : 6'-AMINOQUELK!ErI'IN.The solid, on keeping in the air, becomes red on the surface. Whenheated in the steam-oven, itd loses most of its nitrogen:(1) 0.1376 gave 7.1 C.C.N, (moist) a t 19O and 749 mm. N=5*84.(2) 0.1520 ,, 8.2 C.C. N, (moist) a t 30° and 755 mm. N=5*81.[Sample (2) was purified by solution in alcohol and re-precipita-Querce tin Pen tame t hyl Ether Diazo nium Sulphat e,C2,Hl,0,N2C1 requires N = 6-44 per cent.tion with ether.]C,,H4O2(OMe),*N,HSO4.-This substance was gradually precipitated in short, needle-shapedcrystals on the addition of dilute sulphuric acid, (1 :4) to a coldsaturat.ed aqueous solution of the diazonium chloride :0.1520 gave 6.9 C.C. N, (moist) a t 30° and 756 mm.C2,H,,0,,N,S requires N = 5-64 per cent.6'-P-iVaph t holazo puerce tin Pentame thy? Ether,Cl,H,0,(OMe),*N2*C,oH,*OH.-One gram of 6/-aminoquercetin pentamethyl ether was diazotisedas described in the preparation or the diazonium chloride, water(about 100 c.c.) was then added until all the diazonium chloridehad dissolved, and this liquid was added gradually to a solutionof 0.4 gram of P-naphthol and 1.9 grams of potassium hydroxidein 50 C.C.of water. The gelatinous, crimson precipitate wascollected, washed, dried, and recrystallised from glacial aceticacid, from which solvent it was deposited in needle-shaped, crimsoncrystals, melting and decomposing a t 222-225O (compare P., loc.cit. It is fairlysoluble in alcohol, insoluble in aqueous alkalis, and dissolves insulphuric acid with an indigo-blue colour :N=4*90.The sample melting a t 211O was not pure).0.1521 gave 7.1 C.C.N, (moist) .at 29O and 758 mm. N=5.09.C3,H2608N2 requires N = 5.16 per cent.Triwitroquercetin Pentamethyl Ether ( ?), C,,H,0,(OMe)5(N0,)3.When quercetin pentamethyl ether was dissolved in cold fumingnitric acid and the solution poured into water, there wzs precipi-tated a canary-yellow substance which, on boiling the mixture,became granular. This substance, after filtering and washing, wastreated with dilute aqueous ammonia, when a considerable portiondissolved, forming a brownish-red solution. The residue, onanalysis, was found to contain a percentage of nitrogen correspondingnearly with trinitroquercetin pentamethyl ether. Attempts torecrystallise this compound were not successful. It was scarcelysoluble in alcohol, acetone, benzene, or acetic acid, but was readilyso in nitrobenzene, and melted a t 190-205°WATSON : 6'-AMINOQUERCETIN. 3470.0652 gave 4.9 C.C.N, (moist) a t 14O and 761 mm. N=8*86.C,,H,,Ol3N, requires N = 8.28 per cent.Dibromoquercetin Pentamethyl Ether Hydro b ro rnide,C15H302(OMe)5Br2,HBr.*This oxonium salt was obtained by brominating quercetin penta-methyl ether either in glacial acetic acid or in carbon disulphidesolution, for example, 2.5 grams of the pentamethyl ether weredissolved in 20 C.C. of glacial acetic acid and to this solution, whencold, were added, gradually, 2.2 grams of bromine (2 mols.) alsodissolved in about 20 C.C. of glacial acetic acid. A t once a deepcrimson liquid and precipitate were formed, which soon, however,became light yellow.When the whole of the bromine had beenadded, the mixture soon became a semi-solid, bright yellow massof fine, needle-shaped crystals. It w-as thoroughly stirred a t inter-vals and after keeping over-night the precipitate was collected andwashed with acetic acid.Dibroinoquercetin Pentamethyl Ether, C,,H,O,(OMe),Br,,EtOE.-The oxonium hydrobromide. prepared as above was, withoutdrying, treated with about 50 C.C. of alcohol which turned thesurface of the mass crimson. On warming, the whole dissolved,and, on cooling, the solution deposited about 1.5 grams of the newsubstaiice in almost colourlws crystals. It was puriGed by recrys-tallisation from benzene, which did not eliminate the molecule ofethyl alcohol, It is readily soluble in hot alcohol, still more soin hot benzene, and crystallism in prisms which, when depositedfrom benzene, are short, and arranged in sheaves.It melts a t0.1255 gave 0.2119 CO, and 0.0467 H,O ; C = 46.04 ; H = 4-13.0.2507 ,, 0.1640 AgBr. Br=27*84.C22H2408Br2 requires C = 45.84 ; H = 4-16 ; Br = 27.77 per cent.As quercetin was produced by the demethylation of this substancewith hydriodic acid, it could not be dibromoethoxyquercetin penta-methyl ether.Monobromopuercetin Pentarnethyl Ether, C15H402(OMe)5Br.-This was obtained by boiling dibromoquercetin pentamethyl etherhydrobromide with water. From each molecule of the oxoniumhydrobromide approximately two molecules of monobasic acid wereliberated, perhaps in accordance with the equation :C,,H30,(0Me),Br2,HBr + H20 = HBr + HBrO + C,,H,O,(OMe),Br.* This and the following bromo-derivatives were Ibreparecl with the idea thatthey might be more readily nitrated than quercetin pentamethyl ether ; thesuccessful preparation of nitroquercetin pentamethyl ether rendered their use,however, unnecessary.173-175' :A A 348 WATSON : 6'-AMINOQUERCETIN.A sample of the oxonium hydrobromide, oarefully washed withacetic acid, was dried on a porous tile and finally, for several days,over soda-lime in a vacuum desiccator, being removed from timeto time, finely powdered and replaced in the desiccator.Afterboiling with water, 0.3887 gram gave 11.2 C.C. N/lO-acid, whilstthe above equation requires 12.7 C.C.The insoluble product was recrystallised from benzene or alcoholand obtained in needles melting a t 215O.It was of a light creamcolour and was not readily soluble in either of the above-mentionedsolvents.The same substance was obtained directly by the bromination ofquercetin pentamethyl ether in acetic acid solution in the presenceof fused sodium acetate: 1 gram of the pentamethyl ether wasbrominated as in the preparation of the dibromo-oxonium hydro-bromide, excegt that 1.5 grams of fused sodium acetate were dis-solved with the pentamethyl ether in acetic acid before the brominewas added. A cream-coloured precipitate of apparently uniformneedles was a t once obtained. After remaining over-night, theprecipitate (0.65 gram) was collected, washed with acetic acid, andrecrystallised from benzene.0'1217 gave 0.2345 CO, and 0.0438 H20.C = 52.88 ; H = 4-08.0.2522 ,, 0.1054 AgBr. Br=17*80.C,,H,,O,Br requires C = 53.21 ; H = 4.21 ; Br = 17.75 per cent.Dibromonitroquercetin Yentamethyl Ether ( ?),C,,H,O2(01Me),Br2.NO,.One gram of dibromoquercetin pentamethyl ether was dissolvedin cold nitric acid (D 1.4) and the solution poured into water. Theprecipitate was collected, recrystallised from alcohol, and againfrom acetone, and was thus obtained in rhombic crystals of a paleyellow colour melting a t 173-175O :0.1072 gave 2.4 C.C. N, (moist) a t 14O and 773 mm.C,,H170,NBr2 requires N = 2.43 per cent.The peculiar behaviour of dibromoquercetin pentamethyl ethermakes it unwise, however, to definitely assign the above formulato this compound without further analysis.N=2*67.Quercetin Pen(tamethty1 Ether Oxonium Salts, C20H,,07,HCl ;C2,H,,07,HBr ; C,,H~O,H,SO, ; and C2,H2,07,HN0,.The first three of these oxonium salts were prepared by dissolvingthe pentamethyl ether in glacial acetic acid and adding the corre-sponding acids in a concentrated form, and separated in yellow,needle-shaped crystals. They were collected, washed with glaciaWATSON : 6’-AMINOQUERCETIN.349acetic acid, dried on a porous tile, and finally over soda-lime in avacuum, being once or twice finely powdered to expose a freshsurface. The sulphate could be dried in the steam-oven withoutdecomposition. These substances were analysed by boiling withwater, filtering, and titrating the filtrate :0.3471 hydrochloride gave 8.5 C.C.N / 10-acid ; HCl = 8.9.0.4240 hydrobromide (still containing a trace of acetic acid) gaveC,,H,,O,,HCl requires HCl = 8.9 per cent.10.6 C.C. N/lO-acid, and 0.1672 AgBr; HBr=17-41.C,,H,,O,,HBr requires HBr = 17.88 per cent.0.2904 sulphate gave 12.2 C.C. .Y/ 10-acid ; H,SO, =20*6.C,,H,,0,,H2S04 requires H,SO, = 20.85 per cent.All these salts showed a tendency to develop a red colour; thehydrochloride and hydrobromide became red on the surface whenleft on the tile, and the sulphate became salmon-coloured whenheated to looo.The nitrate was prepared by adding concentrated nitric acid inthe cold to a solution of quercetin pentamethyl ether in glacialacetic acid in the presence of carbamide nitrate, but, in this way,could not be freed from carbamide nitrate. It was also obtainedby adding concentrated nitric acid to a cold solution of quercetinpentamethyl ether in absolute alcohol and separated as a brightyellow precipitate of needle-shaped crystals. It was collected andE t once dried on a porous tile and over soda-lime in a vacuum.It could not be washed with alcphol without decomposition, norwith acetic acid as, in this medium, in the absence of carbamidenitrate, it is a t once attacked by the nitric acid with the liberationof nitrous fumes:0.2433 (dried without washing) gave 7.6 C.C. NIlO-acid;0.2541 (washed with alcohol) gave 5.1 C.C. N / 10-acid ; HNO, = 12.6.HNO, = 19 6.C~oH2007,HN03 requires HNO, = 14.4 per cent.I n conclusion, I wish to express my very great indebtedness toRfr. A. G. Perkin, F.R.S., who pointed out to me the advantagesof choosing quercetin as a starting point for researches such as Ihad in view, and who has, throughout the work here described,assisted me by continual advice and guidance.CLOTH WORKRRB’ RESEARCH LABORATORY,LEEDS UNIYERBITY