MELVINS. NEWMANAND
5600
3,4-Dihydro-J ,$-epoxy-l,2-benzanthracene(XI).-In the best of several experiments a solution of 540 mg. (3.3 mmoles) of trisdimethylaminophosphine in 15 ml. of benzene was added dropwise to a stirred solution of 800 mg. (3.1 mmoles) of \' in 15 ml. of benzene a t 55'. -1fter 2 hr. a t 55' the benzene was removed under vacuum. T h e residue was triturated quickly with 10 ml. of cold dry ether to remove phosphorus compounds and then recrystalli ed from dry ether t o yield 555 mg. (74%) of I1 as colorless crystals, m . p . 119-120' dec. when heated a t the rate of 6" per min. The crystals turn yellow near 100" (see below); I 1 shows a band a t 11.30 and two doublets a t 5.50 and 5.59 T , J 4.5 C.P.S.for each. .4nai. Calcd. for ClsHlrO: C , 88.5; H , 4.9. Found": C , 88.2; H , 5.0. U'hen the above reaction was run a t the reflux temperature of benzene, the yield of I1 dropped t o 329;. When I1 was treated with hydrochloric acid, as in the case of I , a 43y0 yield of 3-hydroxy-1,2-ben~anthracene,~ m.p. 196-205' dec., was obtained. The acetatez2melted a t 128-129'. The infrared spectra of our 3-hydroxy-l,?,-benzanthraceneand of Prof. Fieser's were identical. \Then the material in the mother liquor of the above acid treatment was subjected to a similar treatment a further quantity of 3-hydroxy-l,2-benzanthracenewas obtained. Other tests with acidic reagents appeared less promising. We did not pursue this reaction as the I1 a t hand was needed for biological experiments. n'hen 105 mg. of pure I1 was heated a t 1 m m . the crystals turned yellox\- a t about 100". violent reaction occurred near 120" and the solid liquefied. Xt about 160" (bath temperature) the oil solidified. After heating a t 160" for 5 min., the tube \vas cooled and the pale yellow-orange solid melted a t 172-180'. The infrared spectra of this material and of authentic 3-hydroxy-l,2benzanthracene12 were almost identical. After two recrystallizations from toluene, 68 mg. (65Tc) of pure 3-hydroxy-l,2-benzanthracene, n1.p. 201-203" dec., was obtained.
3,4-Dihydro-3,4-dihydroxy-lO-methyl-l,2-benzanthracene .-
By a procedure similar to t h a t described,Y3.3 g. of pure 10methyl-1 ,?,-benzanthracene was oxidized t o the dihydroxy compound, m.p. 194-195', in G i C ; yield. .-IrinZ. Calcd. for CleH160i: C, 82.6; H , 5.8. Founds: C, 82.5; H , 5.9. l-Methyl-3-phenylnaphthalene-2,2 '-dicarboxaldehyde (VI).The oxidation of the above diol to VI, m.p. 100.0-100.5", was accomplished in 76C; yield as described.* .And. Calcd. for ClUHl4O*:C , 83.2; H , 5.1. Founda: C, 83.0; H,5.0.
3.4-Dihvdro-3.4-e~oxv-lO-methvl-l.2-benzanthracene (111). -In the best of several experiments a solution of 1.0 g. of tris.
,rELVIN
s. N E W M A N A N D JOCHANAN
BLUM
RECEIVED AUGUST5, 1964 The two reagents of the title have been prepared as reagents of use in forming charge-transfer type complexes suitable for X-ray analysis. T h e syntheses are described and many complexes listed in Tables I and 11.
The present work was stimulated by the fact t h a t 1methoxybenzo [clphenanthrene forms two 1 : l-complexes with 2,4,7-trinitrofluorenone(TNF).' One complex, m.p. 2 1 2 O , is black and the other, m.p. 157O, is brown-red? The thought arose t h a t the structure of these two complexes might be determined by X-ray analysis of the crystals. However, X-ray crystal analysis of a substance which does not have a heavy atom, such as bromine or iodine, is considerably more dificult and time-consuming than would be the case (1) T h i s work was supported b y G r a n t S o . CA-06.180~03from the U. S. Public Health Service. 12) hl O r c h i n and 0 Woolfolk, J . A m C h e m S o c . , 68, 1727 (1946). ( 3 ) A I S S e w m a n and J. B l u m . i b i d . , 86, 503 (1964).
if such an atom were present. Accordingly, we have synthesized 4-bromo-2,5,7-trinitrofluorenone( B T N F , I ) and 4-iodo-2,5,7-trinitrofluorenone( I T N F . 11) in the hope that two different complexes, suitable for X ray analysis, might be obtained with l-niethoxybenzo[clphenanthrene. However, only one complex was formed with B T N F and I T N F . \Ye have prepared complexes of B T N F and I T N F (see Table I) with a number of substances with three objects in mind: 1, to test the ability of these reagents t o form complexes; 2, to see if two different 1 : l complexes might be obtained; and 3, to make complexes available for X-ray crystallographic studies.
Dec. 20, 1964
5601
4-HALO-2,5,7-TRINITROFLUORENONES AS COMPLEXING AGENTS
TABLE I COMPLEXES O F 4-BROMO-2,5,7-TRINITROFLUORENONE Aromatic component
1 Benzene 2 Anthracene
3 Fluoranthrene
Prepared in
Benzened Benzeneethanol Benzeneethanol
Recrystallized from
Color
M . p . , OC."
Formula
-Bromine: Calcd.
%Found
Same
Yellow
Dec.
CleH10BrN307
17.0
17.1'
Same
Dark violet
208-209
C Z ~ H , ~ B ~ N S O 1; 4 . 0
13.8
13.7 Orange 761-198 Cl~H~~Brh'30713.4 Same l-cyclohexenyl13.3 Wine-red 135-136 C2~H18BrN307 13.3 Acetic acid Same naphthalene' 12.9 12.8 Orange-red 263-265 C31H1eBrN307 Benzene Same 5 Chrysene Olive green 13.0 278-280 C ~ I H I ~ B ~ N ~1 2O. 9~ 6 Kaphthacene' Acetic acid CCl, 7 1-Methylbenzo[c]12.7 phenanthrene' .4cetic acid Acetic acid-CC14 Red 163-164 C3~H18BrN307 1 2 . 6 8 1-Methoxybenzo[c]12.3 12.2 Benzene CCla Purple-violet 197-199 C3zH18BrN308 phenanthreneh Olive green 261-263 Ca3Hl&kNaO; 12.4 12.3 Benzene Same 9 Benz[a]pyrene 10 1-Acetylaminobenzo [c]. 11.9 phenanthrene' cc1, Same Cherry red 173-175 C33H1~BrN408 1 1 . 8 11 1,12-Dimethylbenzo[c]12.3 12.4 CC14-acetic acid Red Acetic acid 144-145 C~~H BrZN307 O phenanthrene' Purple . 9~ 1 1 . 9 243-244 C ~ S H ~ S B ~ N 1~1O 12 Dibenz[a,h]anthracene Benzene Same * Analyses by Spang Laboratory, Ann a Most melting points occur with decomp.. hence the rate of heating in general is rapid. Analysis by A . Bernhardt, Arbor, Mich. Correct values (within 0.37,) were also obtained for C, H , and N for each compound. Miilheim, Germany. Formed in and recrystallized from benzene; sample held a t 1-2 mm. over paraffin wax for 1 day for analysis. e T N F complex has 2 moles of T N F ; see Klemm ref. in footnote e, Table 11. 1 Owing to insolubility of naphthacene, best results were obtained by using excess hydrocarbon in boiling acetic acid, followed by filtration while hot to remove excess naphthacene. 0 Does not form picrate; T h ' F derivative was formed in poor yield; see M . S. Newman and W. B. Wheatly, J . Am. Chem. Sac., 70, 1913 Forms two 1 : 1 T S F derivatives; see ref. 3 in text. i No solid T N F derivative was obtained; see M . S. Newman and J . (1948). Blum, J . A m . Chem. SOG., 86, 1835 (1964). j M . S. Newman and M. Wolf, ibid.,74, 3225 (1952), report t h a t neither picrate nor T N F derivative of this hydrocarbon could be prepared. Later, the T N F and T E X F derivatives were prepared; see M . S. Newman and . . W . B. Lutz, i b z d , 78, 2469 (1956). 4
The required bromo and iodo compounds I and I1 were prepared by nitration of 4-bromo- and 4-iodofluorenone, respectively. I n order t o prove t h a t the nitro groups had the positions indicated, the bromine and iodine atoms were removed reductively b y heating I and I1 with copper powder and benzoic acid.4 Isolation of 2,4,7-trinitrofluoren~ne~ in each case proved t h a t the structures of I and I1 are as stated. On working with I and 11 it became obvious t h a t these were excellent complexing agents. A number of complexes are listed in Tables I and 11. Especially noteworthy is the fact t h a t benzene forms stable 1: 1 complexes with I and 11. These complexes do not lose benzene on holding a t 1-2 mm. pressure at room temperature for a day. However, the melting point cannot be taken since the benzene is lost on heating a t temperatures below 100°. T o our knowledge these represent the only known stable solid charge-transfer type of complex involving benzene and a polynitro compound. Benzene and toluene have been shown not to form complexes with T N F . 5 The melting points of the few comparable complexes with 2,4,5,7-tetranitrofluorenone6 are lower than those from B T N F or ITNF. All of the hydrocarbons we used form 1 : 1-complexes except prehnitene, which forms a 2 : 1-complex (see Table 11). MTe have available quantities of the complexes listed in Tables I and I1 if any person is interested in determining the crystalline structure.
Experimental 4-Bromo-2,5,7-trinitrofluorenone(BTNF, I).-In
t h e best of several runs 2.54 g. of powdered 4-bromofluorenone7 was added (4) This method is based on a comparable process for removal of aromatic chlorine in niti-o compounds; see W . T . Smith, Jr., J . A m . Chem. SOC., 71, 2855 (1949). (5) D. E. Laskowski and W. C. McCrone, A n a l . Chem., a6, 1497 (1954). ( 6 ) M . S. Newman and W. B. Lutz, J . A m . Chem. SOL,7 8 , 2469 (1956).
during 5 min. t o a stirred mixture of 36 ml. of fuming nitric acid ( d . 1.59-1.60), 25 ml. of 967, sulfuric acid, and 3 ml. of acetic acid a t room temperature. T h e mixture was brought t o reflyx during 15 min. and held there for 60 min. T h e clear brown solution was cooled and poured on ice. T h e yellow precipitste was collected, washed with water, dried, and recrystallized twice from benzene t o yield yellow crystals of the benzene complex (see Table I ) of B T N F . After removing t h e benzene by heating t o 70" under vacuum, a yellow powder remained. Recrystallization from ethanol afforded 3.08 g. (78y0)of I as shiny yellow plates, m . p . 191-192'. Anal. Calcd. for ClaHnBrh'aO;: C , 39.6; H , 1.0; Br, 20.3; N , 10.7. Found: C , 3 9 . 7 ; H , 1.1; Br, 20.5; N, 10.7. 4-Iodofluorenone .-Moist 4-aminofluorenone hydrogen sulfate, prepared from 49 g. of 4-aminofl~orenone,~ was suspended i n a solution a t 0" of 8 5 ml. of concentrated sulfuric acid in 1.5 1. of water and diazotized with 18 g. of sodium nitrite. T h e mixture was stirred a t 5" for 30 min., then filtered. The clear filtrate was treated with urea t o remove excess nitrous acid and cooled t o 0'. On adding slowly a solution of 60 g . of potassium iodide in 150 ml. of water a heavy red precipitate was formed. After a series of color changes on heating t o 100' and holding a t reflux for 30 min. the mixture was cooled and filtered. T h e solid was washed well with 10% potassium hydroxide t o remove phenolic impurity and eventually was distilled a t 1.5 m m . ( b . p . 184-185"). Recrystallization from ethanol afforded 57 g. (74?,) of bright yellow elongated prisms of 4-iodofluorenone, 1n.p. 121-122". Anal. Calcd. for C13H710: C , 51.0; H , 2.3; I , 41.5. Found: C , 51.1; H , 2.2; I, 41.4. 4-Iodo-2,5,7-trinitrofluorenone (ITNF, II).-As in the case of B T S F , 3.06 g. of 4-iodofluorenone was nitrated with 36 ml. of fuming nitric acid, 25 ml. of concentrated sulfuric acid, and 7 ml. of acetic acid. T h e reflux period was 45 min. T h e crude nitration products melted over a wide range, b u t one crystallization from benzene yielded t h e benzene complex (see Table 11) in almost pure state. Recrystallization from ethanol afforded 2.10 g. (487,) of I1 as yellow fluffy crystals, m.p. 193-194". (7) 4-Bromofluorenone was prepared essentially as described by E. H . Huntress, K . Pfister, 3rd, and K . H . T. Pfister, J . A m . C h r m . S o r . , 64, 2845 (1942), but on a much larger scale. The yields were somewhat higher. I n the last step, use of a 307' excess of cuprous bromide led to a 547" yield of pure 4-bromofluorenone, m . p . 127-128O.
MELVINS. NEWMAN AND
5602
JOCHANAN
BLUM
Vol. 86
TABLE I1 COMPLEXES O F 4-IODO-2,5,~-TRINITROFLUORENONE Prepared in
Aromatic component
1 Benzene 2 Toluene 3 2-Methylnaphthalenea 4 Acenaphthene Phenanthrene
5
Recrystallized from
Color
M . P . , 'C."
-Iodine! Calcd.
%-Found
Benzened Toluened Acetic acid Benzeneethanol Benzeneethanol
Same Same Same
Light orange Light orange Orange
Dec. Dec. 117-121
24 5 23 8 21 8
24 23 9 21 9
cc1,
Red-brown
174-176
21.4
21.4
Same
Orange
179-180
20 5
20.6
Benzeneethanol Benzeneethanol Benzene
Golden brown
247-248
19.0
18.9
6
Benz[a]anthracene
Benzene
m
Benzo [ c ] phenanthrene
Benzene
Violet-black 185- 186 19.0 19 1 Acetic acid Orange 259-260 19.0 18 9 8 Triphenylene Benzene9 1-Methylbenz [ a ]ethanol Same Olive-brown 213-214 18.6 18.4 anthracene' Benzene10 1-Methoxybenzo[ c ] Brown 190- 192 ethanol Same 18 2 18.2 phenanthrene Orange 113-115 Prehnitened 17.9 18.1 Same 11 Prehnitene Analyses by Spang LaboraMost melting points occur with decomposition; hence the rates of heating in the cases cited are rapid. Analysis by A . tory, Ann Arbor, Mich. Correct values (within 0.3%) were also obtained for C, H, and N for each compound. Bernhardt, Miilheim, Germany. d Formed in and recrystallized from hydrocarbon component; samples held a t 1-2 mm. over paraffin wax for 1 d a y for analysis. e According t o M . Orchin, L. Reggel, and E. 0. Woolfolk, J . A m . Chem. SOC.,69,1225 (1947), the T N F complex has 2 moles of T N F . See, however, L . H . Klemm and J , W. Sprague, J . Org. Chem., 19, 1464 (1954). 1 Forms only a dipicrate; see M . Orchin, ibid., 16, 1165 (1951). For T N F derivative see K. H . Takemura, H . D. Cameron, a n d M . S. h-ewman, J . A m . Chem. Sac., 75, 3280 (1953).
TABLE I11 L'LTRAVIOLET
S P E C T R A O F 4-BROMO-2,5,~-TRISITROFLUORESOSE (I) AND 4-IODO-2,~,7-TRINITROFLUORESONE (11)"
I in chloroform, 18.6 mg.11. I in 95% ethanol, 14.0 mg.ll. I in benzene, 36.4 mg.11. I1 in chloroform, 18.0 mg./l. I1 in 95% ethanol, 14.8 mg 11.
I1 in benzene, 36.0 mg./l. a
mp log 6 Amax, m p log L Amax, m p log e Amax, m p log c Amex, m r log c Aman, m p log A,,
244 0 249 5 255 5 4 34 4 36 4 37 243 0 248 5 254 5 4 27 4 25 4 29
239.0 244.0 249 5 4.45 4.48 4 53 2 2 6 . 5 2 4 3 . 0 248 5 4.46 4.40 4 39
Spectra taken on a Cary 14 spectrophotometer.
255 5 4 54 254 5 4 38
283 0 3 3 4 . 0 353b 37Bh 3 68 3 51 4 45 3.98 352* 376' 283 0 333 5 3 83 3 60 3.97 4 42 272 0 282 0 4 30 4 31 286 5 347.5 261 5 4 39 3.98 4 47 284 5 340.0 261 0 3.99 4 35 4 35 272 5 281 5 4 26 4 26
261 5 4 32 261 0 4 24
38Bh 3 62 39oh 3 66 39oh 3 73 396b 3 54
Inflection point (approx.).
Anal. Calcd. for Ci3H41S307: C , 35.4; H , 0.9; I , 28.8; S , 9 . 5 . Found: C , 35.5; H , 1.0; I , 29.1; N, 9.8. Proof of Structure of I and 11.-In the best of several trials a mixture of 0.5 g. of I1 and 0.80 g. of benzoic acid was heated in a test tube to about 170" (internal thermometer). On addition of 0 . 5 g. of copper powders during 5 min. an exothermic reaction caused the temperature t o rise t o 180'. After cooling, the unused reaction mixture was extracted with sodium bicarbonate solution. Chromatography of the neutral product over FlorisilQ using benzene as eluent yielded a first fraction which, on recrystalliza(8) "Extra Fine Copper" KO.5143, 0. Hommel Co., Pittsburgh, Pa., was used Venus Copper 44F led to poor results. (9) A magnesia-silica gel catalyst, Floridin Co.
tion from benzene, afforded 0.28 g. (78y0)of 2,4,7-trinitrofluorenone.* Comparison with an authentic sample was by m.p. and infrared spectra. T h e temperature range of the reaction, 170-180", is crucial if a good yield of T S F is t o be obtained. I n the case of the debromination of I by a similar process, the best yield of T S F obtained was 22y0, since higher melting m i x tures and tars were obtained. We feel certain t h a t no other trinitrofluorenones were present, a n d , hence, the structure of I is established. The ultraviolet spectra of I and I1 are given in Table 111. Since the spectra in benzene are so different from the spectra in chloroform and in ethanol, we believe t h a t the spectra in benzene represent the spectra of the benzene complexes.
