Syntheses of Phenanthrene Derivatives. I. Reactions of 9

Ronnie Benshafrut, Roy E. Hoffman, Mordecai Rabinovitz, and Klaus Müllen. The Journal of Organic Chemistry 1999 64 (2), 644-647. Abstract | Full Text...
0 downloads 0 Views 523KB Size
June, 1934

REACTIONS OF 9-PHENANTHRYLMAGNESIUM BROMIDE [CONTRIBUTION FROM THE CHEMISTRY

LABORATORY OF THE

1363

UNIVERSITY OF MICHIGAN]

Syntheses of Phenanthrene Derivatives. I. Reactions of 9-Phenanthrylmagnesium Bromide BY W. E. BACHMANN The importance of compounds containing the phenanthrene group makes it desirable to have methods of synthesis leading to derivatives of definite structure. The Friedel and Crafts reaction with phenanthrene which has been employed on a number of occasions for making phenanthrene derivatives has the disadvantage of giving isomeric derivatives which are often diflicult to separate and whose structures must subsequently be determined. Specifically, we needed 9-benzoylphenanthrene in an investigation. Willgerodt and Albert reported that 9-benzoylphenanthrene (m. p. 127') is formed by the action of benzoyl chloride on phenanthrene in the presence of aluminum chloride, although they offered no proof of the structure of their product. Recently Mosettig and van de Kamp2 from their experience with the condensation of phenanthrene with acetyl chloride expressed the view that the entering of the benzoyl group into the 9-position was doubtful. We decided to prepare the ketone by a method which would leave no doubt of the structure of the final product and we turned to the Grignard reagent from 9-bromophenanthrene. This Grignard reagent has been employed previously on only three occasion^.^ We have now carried out a number of experiments with 9phenanthrylmagnesium bromide in order to determine its value for introducing the 9-phenanthryl group into molecules. Synthesis of 9-Benzoylphenanthrene.-By interaction of 9-phenanthrylmagnesium bromide (I) and benzonitrile, followed by hydrolysis of the ketone-imine that is formed, 9-benzoyl phenanthrene (11) is obtained in the form of colorless needles melting a t 89-90' instead of 127' as reported by Willgerodt and Albert. The identical product (m. p. 89-90') is obtained from 9-cyanophenanthrene and phenylmagnesium bromide and also by oxidation of phenyl-9-phenanthrylcarbinol (111) which has been prepared from 9-phenanthrylmagnesium bromide and b&zaldehyde. We are now engaged in synthesizing the isomeric (1) Willgerodt s a d Albert, J prakt Chem., [21 84, 383 (1911). (2) Mosettig and van de Kamp, THIS JOURNAL, 66, 3442 (1933). (3) Pschorr, Ber., 39, 3128 (1906); Salkiad and Lubiaskaja, r b d , 61,269 (1928); Shoppee, J . Chem. S O L ,37 (1933).

benzoylphenanthrenes for comparison with the product of Willgerodt and Albert. By reduction of their ketone Willgerodt and Albert obtained "9-benzylphenanthrene" as colorless leaflets melting a t 91-92'. We have prepared 9-benzylphenanthrene (IV) by interaction of 9phenanthrylmagnesium bromide and benzyl chloride and also by reduction oE phenyl-9-phenanthrylcarbinol ; both methods give a product crystallizing The relationin prisms which melts at 153-154'. ships are apparent from the following scheme.

I

\

a

Reaction with Alkyl Halides.-The Grignard reagent reacts with alkyl halides to give 9alkylphenanthrenes. In addition to benzyl chloride we tried the action of methyl iodide, diphenylbromomethane and 9-bromofluorene. The last three reagents give 9-methylphenanthrene, diphenyl-9-phenanthrylmethane (V) and biphenylene-9-phenanthrylmethane (VI), respectively. Triphenylchloromethane reacts rapidly with 9-phenanthrylmagnesium bromide but no crystalline product could be obtained from the reaction mixture.

W.E. BACHMANN

1364

Reaction with Aldehydes and K&cmes.-9Phenanthrybagmsiurn bromide reacts normally with formaldehyde to give 9-phenanthryicarbinol and with acetaldehyde and benzaldehyde to give secondary alcohols of formula VI1 (in which R is CH3,C~HF,).Reaction of the Grignard reagent with fluorenone gives bipheenylene-9-phenanthrylcarbinol (VIII) ; with benzophenone diphenyl9-phenanthry1-carbinol (IX) is formed.

By hydrolysis of the ketone-imine that is produced by interaction of 9-phenanthrylmagnesium bromide and o-tolunitrile there is obtained otolyl-9 phenanthryl ketone (XI). This synthesis is of special interest inasmuch as Clar6was unable to isolate the ketone from the mixture of isomers that was obtained by condensation of o-toluyl &Imide and phenanthrene, We have foand that the pure o-tdyl-9-phenanthryl ketone iS eonverted to 1,2,3,4-dibmzoanthracerve (XII) when it is heated at its boiling point. Since Qar ubtained this diknzo&aePrre by hating his mixture ai ketones om results show that he actually had a-tdyl-4pknmthryl ketone and dso furnish proof of the stmdare of the dibenzoanthrame,

? P O 0X

Vof. 56

V

XI

v

XI1

Reaction d the Grlgnard Reagent w%h CarBiphenylene - 9 - phenanthrylcarbind (VITI) bon M d a and wi& CWoetfty€Carbonate.is reduced by hydriodic acid quantitatively to With carbon dioxide the M g m r d reagent gives biphenylerte-9-phenanthrylmethane(VI). When a good ykM cif phepra~hreaei9-carboxylicacid diphenyl-9-phenanthrylhrylcarbinol (IX) is heated and the rruethd ranks with the recent process of with hydriodic acid, two products are formed; Mosdtig and van de E m p a is a convenient one of these is the nomal reduction product di- met-Bod af preEpapiRg thi?3acid. With &orm&yl phenyl-9-phensrrathrylmethane (V) ; the other carbonate the Grignwd reagent gives the &yl product is foamed not. by reduction but by lass ester of phenanthrc?.Bte-9-carbzyEcaeid. This reof the dements of w a t a since the same com- action o f f a sa n o w method of obtaining rapidly pound is oM&d in nearly q,mtitative yield phenantEnene-9-carboxy€icacid (by hydrolysis of based on the when the carbinol is warmed with sulfuric acid in the ester) in excellent ykkl acetic acid. Triarykarbinols freqsently undergo bromophem&hrene). 9 . ( 9 ’ - P ~ ~ I ) L g h ~ ~ ~ ~ ~re-. - T h e dehydration to give 9wylfluorenes and it was comidemd that such P reaction had taken place action between an aromatic Grignard reagent in this instance. Of the two W y compounds VI RMgX znd anhydrous ewprie chloride ususlly and X that could be formed in this manner only yields the hydrotarbon RR; 9-phenantkyEmagX ( 1 , 2 , 3 , 4 - d i b e n z o - e ~ ~ ~ ~ ~ ~ren ~ ~nesium e n e ) brarnirte. like& gives the oxrespanding 9-(9’-p~enanthryl)-~henan~~~e mained when it was hand that the prsdwt was hydrocarbon different from VI that had been prepwed born 9phenanthrylmagnesium bromide and 9-bramofluorene. Structure X was confirmed when it was found that the compound was identical with 1,2,3,4-&benzo-9-phenylfluorenethat w a s synthesized recently by K0e1sc.h.~ Synthesis of o-Tolyl-9-phenanthryl Ketone (XI) and 1,2,3,4-Dibenzoanthracene (%I).(5)‘ Clar, Ber., 6% 380, 1574 (1929).

fw0

(4) Koelsch, THISJOURNAL, 86,480 (1934).

(6) Mosettig and ran de Kame,T I ? ~ ~ J o u R I W, J A3858 ~ ~ , (am.

June, 1934

REACTIONS OF 9-PHENANTHRYLMAGNESIUM BROMIDE

1365

(XIII). Attempts to prepare the compound moved from the ckear sdution and the Sphenanthrylfrom 9-bromophenanthrene by reaction with magnesium bromide was then treated with a reactant. copper and with sodium have so far been un- When the reaction was complete the product was hydrolyzed. The essenthl drtrr are given in Table I. In all successful. cases the products were obtained as colorless crystals. The Chemiluminescence of the Grignard Reagent. analyses were carried out by W. S. Struve. -9-Phenanthryhagnesiurn bromide exhibits a Preparation of 9-Renzoylphenanthrene.-(a) From 9splendid chemiluminescence when it is exposed phenanthrylmagnesium bromide and benzonitrile : see to air. In a dark room the greenish-blue light Table I. (b) From Seyanophenanthrene: 9-cyanophenanthrene emitted when the solution is swirled in a flask open was prepared in excellent yield from 9-bromophenanthrene t o the air is sufficient for reading several lines of and cuprous cyanide according to the directions of Mosettig journal type. When air is passed through the and van de Kamp;6 the product after separation from insolution heat is developed and the salt of 9- organic material by extraction with chloroform was purihydroxyphenanthrene is formed. In a number of fied by distillation and recrystallization from chtoroform and alcohol; the 9-cyanophenanthrene was obtained as experiments we employed the disappearance of needles which melted a t 109.5-110'. Werner reported chemi1uminescent property of the solution as an 103" and Mosettig and van de Kamp gave 103-104° for indication of the end of the reaction between the the melting point. A mixture of 5.4 g. of 9-cyanophenanthrene and a soluGrignard reagent and a reactant. tion of phenylmagnesium bromide prepared from 6 g. of Conclusion.---In addition to the compounds bromobenzene was refluxed for five hours. The reaction mentioned the Grignard reagent reacts with a mixture was then worked up as described in note (a) under number of other substances, including sulfur. Table I ; yield of 9-benzoylphmanthrene. 3 g. (42%); We are also extending our study to the isomeric m. p. 89-90'. (c) By oxidation of phenyl-9-phenanthrylcarbinol. bromophenanthenes, some of which may prove useful in synthesizing compounds leading to The seconctary alcohol was oxidized to the ketone By making the b r a m o m ~ e s i u msalt RR'CHOMgBr and reaccholesterol and ergosterol. We plan ta deter- t i m of the salt with benzaldehyde, which abstracts €3 and mine the migration aptitude of the phenanthryl MgBr.' To a cold filtered solution of ethylmagnesium group in the pinacol rearrangement and also the bromide which had been prepared from 0.8 g. of ethyl effect of the group on the dissociation of hexa- bromide in 10 ce. of ether was added 5cc. d benzene and then phenyl-9-phenanthrylcarbiml until no further evoarylethanes and of pentaarylethanes. Experimental 9-Bromoph~mthrene.-The 9-bromophenanthrene obtained by heating phenanthrene dibromide was puriged by distillation under reduced pressure followed by recrystallization from ligroin (6G-8O0); our product melte8 a t 65-66' while the value usually reported m the literature is 63'. If the bromophenanthrene is not quite pure a dark green color and some red precipitate is formed ab the start of the reaction with magnesium; the pure bromide gives a transparent light-brown solution without intermediate color effects. heparation of the Grignard Reagent.-Rchorr used a mixture of ether and anisde as the sdvent in the p e p m a tion of the Grignard reagent because with ether alene the reagent separates as a solid. We have found that a mixture of ether and benzene keeps the reagent in solution; moreover, the benzene is more easily removed from the final product than is anisole. For the experiments listed in TaMe I a mixture of 6.45 g. of 9-bromophenanthrene, 0.65 g. of magnesium ribbon, 0.05 g. of iodine, 15 cc. of ether and 15 cc. of benzene was refluxed on a steam-bath for four to five hours; numerous quantitative experiments showed that sS-e870 of the theoretical amount of magnesium reacts in this length of time. Since the reagent reacts with oxygen, the reaction mixture was always protected from free access of air by means of a mercury trap and the reaction was often run in an atmssphere OP &rogen. The excess of magnesium (about a 0 5 9.) was re-

lution of ethane gas occurred; 1.53 g. of earbinol was required; after a short time a large amount of the bromomagnesium salt of the carbinol precipitated. Two cc. of benzaldehyde was added and. the resulting mixture was refluxed on a steam-bath for one hour; hyclrolysis gave 1.0 g. (66%) of 9-benzoyLphenanthrene (m. p. 89-90') which was identical with the product obtained by methods (a) and (b) . The ketone was also prepared conveniently from 9bromophenanthrene without isolation of the intermediate carbinol. Thirteen grams af 9-bromophenanthrene was eonverted t o the Grignard reagent; the solution was cooled to -5' and treated gradually with a solution of 5 g. of benzaldehyde i n 25 cc. of benzene, the temperature being kept MOW Q". The mixture was kept cold for twelve hours; dhPing this time a b g e a m o w t of the carbinol salt precipitated. The lquid was decanted and to the residual solid was added 10 cc. of benzaldehyde; after the mixture had been refluxed for one hour it was hydrolyzed. The ketone was purified by distillation under reduced pressure fb. p. 235-240" (1 mm.)) and recrystaliization; yield, 6.2 g. Reduction of Pheny&9-phenanthryluarbinol to 9-Benzy1phenanthrene.-A mixture of 0.5 g. of the carbinol, 0.5 g. of iodine, 0.5 g. af red phosghorus, 1 cc. of water and 10 cc. of acetic acid was refinxed for one hour, fikered hot ;and poured inte water. B y recrystallization from acetic

(7) Gomberg a d BBcttmsnn, TWISJOURNAL, Shankland and Gomberg, i b i d . . 6%, 4873 09301.

6% 4967 (1930),

Vol. 56

W. E. BACHMANN

mrn

m

y i m

-

3

M

*s VI

. I

3

Y

. I

c

3 Lo

m

.-

DIMESITYL TETRAKETONE

June, 1934

1367

acid the product (0.4 9.) was obtained in the form of color- amount of colorless cubes precipitated; after five minutes less prisms which melted a t 153-154' alone or when mixed of boiling all of the cubes had gone into solution and with 9-benzylphenanthrene obtained from S-phcnanthryl- shortly after the clear hot solution deposited the hydrocarbon (X) in the form of fine needles; w t . 0.4 g.; m. p. magnesium bromide and benzyl chloride. The nature of the intermediate cubes has not Reaction of Diphenyl-9-phenanthrylcarbinolwith H y - 208-209'. driodic Acid.-The carbinol (0.5 9.) was dissolved in 10 been determined; the crystals melt at 156' and are concc. of warm acetic sicid, 1 cc. of hydriodic acid (d. 1.7) verted a t that temperature to the dibenzoplienylfluorene was added and the mixture was refluxed for one-half hour; (XI. the mixture of hydrocarbons which precipitated was filtered Reduction of Biphenylene-9-phenanthrylcarbinol (VIII) caroff; wt. 0.4 g. (no color with concentrated sulfuric acid). to Biphenylene-9-phenanthrylmethane (VI).-The By recrystallization from toluene and from acetic acid the binol (VIII) (0.5 9.) was heated with 10 cc. of acetic acid mixture was separated into approximately equal amounts containing 1 cc. of hydriodic acid (d. 1.7) for one hour; of diphenyl - 9 phenanthrylrnethane (m. p. 175-176') the product (0.45 g.) which precipitated was biphenylene-9and the less soluble 1,2,3,4-dibenzo-9-phenylfluorene (m. p. phenanthrylrnethane (VI); m. p. 196-197" alone or when 210-21 1"). The dibenzophenylfluorene crystallized from mixed with the product obtained from S-phenanthryltoluene in fine colorless needles; the melting point was magnesium bromide and 9-bromofluorene. unchanged when the compound was mixed with a sample 1,2,3,4-Dibenzoanthracene (XII) from o-Tolyl-9-phenof dibenzophenylfluorene which was synthesized by anthryl Ketone (XI).-The ketone (0.6 g.) was heated at Koelsch* from 9-(a-hydroxybenzyl)-lO-phenyl-phenan- its boiling point for forty-five minutes and then distilled threne and obtained by him from 1-biphenylene-3-phenyl- a t ordinary pressure; the solid distillate after recrystallizaindene. The identity of the two hydrocarbons was fur- tion from xylene and from acetic acid yielded 0.2 g. of long ther established by making the 9-benzoyl derivative by straw-colored needles; m. p. 200-202"; Clar6 reported I n agreement with Clar the dibenzoanthracene reaction of the hydrocarbon with sodium amalgam and 196-197 benzoyl chloride; the benzoyl derivative of our hydrocar- gives a pale violet-red color with concentrated sulfuric acid ; bon melted a t 225-228" alone and when mixed with the it yields a red picrate (m. p. 205") and on oxidation with chromic acid gives 1,2,3,4-dibenzoanthraquinone. 1,2,3,4-dibenz0-9-benzoyl-9-phenyl-fluorene of Koelsch. Dehydration of Diphenyl-9-phenanthrylcarbinol(IX) to Summary 1,2,3,4-dibenzo-9-phenylfluorene(X).-To a solution of A study has been made of the reaction of 90.5 g. of the carbinol (IX) in 10 cc. of acetic acid was added 5 drops of concentrated sulfuric acid and the mix- phenanthrylmagnesium bromide. ture was boiled. I n a few seconds the intense green color Nine new 9-substituted phenanthrene derivaof the solution changed to a brilliant red color and a large

-

'.

(8) Dr. Koelsch kindly prepared t h e benzoyl derivative and compared i t and t h e hydrocarbon with samples of t h e corresponding compounds t h a t be had 'Synthesized.

[CONTRIBUTION FROM THE CHEMICAL

tives have been synthesized by means of the Grignard reagent.

FEBRUARY 28, 1934 ANN ARBOR,MICHIGAN RECEIVED

LABORATORY OF THE

UNIVERSITY O F

ILLINOIS]

The Highly Activated Carbonyl Group. Dimesityl Tetraketone BY ARZY R. GRAYAND REYNOLDC. FTJSON The search for compounds containing highly activated carbonyl groups whose properties can be studied' conveniently has been extended to the polyketones. The great reactivity of these compounds shows the remarkable cumulative effect of several carbonyl groups when placed in adjoining positions. It has now been possible to prepare dimesityl tetraketone-a compound which presents this same enhanced reactivity combined with a presumptive inertness of the terminal carbonyl groups. The synthesis of dimesityl tetraketone was made possible by the discovery that mesitylglyoxal (I) undergoes a condensation of the benzoin type to give the corresponding acyloin (11). (1) Gray and Fuson, 'THIS JOURNAL, 56, 739 (1934).

The acyloin dissolves in aqueous alkali but is reprecipitated upon acidification of the solution. With ferric chloride solution it gives a coloration. The acyloin is easily oxidized by nitric acid, which converts it into the tetraketone (111). The latter is a bright orange-red solid melting a t 133-134'. CHs

c9H11cor .1

CH~C-_)-COCHO + CHs I

I11 CsHiiCH(0H)COOH IV

CpHllCOCHOH

I1 C~H~ICOCOCOCOC~HI~

+ CoHiiCOCOOH4 + COz V