3668
NOTES
The Journal of Organic Chemistry TABLE I ROH
reaction 1 ___)
ROSO-&6HsNH+
reaction 2 ___)
I Substrate Ia (mmalj
-Reaction SOaPY. mmol
(1)
I1 1Time, hrb
Geraniol (0.5)
0.75
3
Farnesol ( 0 . 5 )
0.75
3
Benzyl alcohol (1.0) Indanol (0.-5)
2.0 1.0
5 20
Indanol (0.5)
1.0
20
Concentration, 0.25 M in THF. Temperature 0-3". Also contained 1.0 mmol of AlCl8. 5
8
RH
-Reaction LiAlHd, mmol
THF, ml
3
5
2
Timeat 2 5 O , hr"
Product I1 (yield, % ) d
truns-2-6-Dimethylocta-2,6diene (98) 3 5 3 Iruns,truns-2,6,lO-Trimethyldodeca-2,6,10-triene (95) 6 12 5 Toluene (75) 3 6 5 Indan (64) Indene (11) 3" 6 5 Indan (87) Indene (6) After 1 hr at 0". Yield determined by vpc analysis using internal standard.
tered in a total synthesis of the sesquiterpene sesquicarene.' For this case the problem was solved by using the pyridine-sulfur trioxide complex in tetrahydrofuran as the reagent for hydroxyl activation according to eq 1 and carrying out the reduction step without isolation of the intermediate sulfate monoester. I n view of the effectiveness of the method in this case and its potential utility and convenience more generally, the applicahion of the technique to other substrates has been investigated. Four typical cases, geraniol, farnesol, benzyl alcohol, and 1-indanol, are reported here. I n each case the alcohol was converted to the sulfate monoester in tetrahydrofuran a,t 0-3" with a moderate excess of pyridine-sulfur trioxide reagent, using thin layer chromatographic (tlc) analysis to determine the extent of reaction as a function of time. After complete esterification an excess of lithium aluminum hydride in tetrahydrofuran was added to effect reduction, leading to the experimental results which are summarized in Table I. It is noteworthy that in the cases of the allylic alcohols geraniol and farnesol, neither cis-trans isomerization nor allylic transposition was observed to a significant extent. I n the case of 1-indanol a substantial amount of indene was obtained as a byproduct. However, by the use of the reducing agent lithium aluminum hydride-aluminum chloride (3 : 1)) the yield of this by-product was reduced to 6% and that of indari raised to 87%. It is apparent that the method described here for the replacement of alcoholic OH by hydrogen is both efficient and convenient and that it is easily competitive even when conventional procedures2 can be applied.
3
filtered, and the filtrate was used for vpc analysis. The products were identified by vpc and nmr comparison with known compounds. Reduction of truns,trans-Farnesol.-To a solution of pure trans,truns-farnesol in 2 ml of T H F was added at 0" 120 mg (0.75 mmol) of pyridine-sulfur trioxide complex, and the suspension was stirred a t 0-3" for 3 hr. Analysis by tlc a t this stage showed that the reaction was complete. A solution of 114 mg (3 mmol) of LiAlH, in 3 ml of T H F was added at 0", and then the mixture was stirred a t 0" for 1 hr and at 25" for 3 hr. After 0.11 ml of water, 0.11 ml of 15% aqueous sodium hydroxide, and 0.33 ml of water were added a t O " , ether (20 ml) was added, and the precipitate was filtered and washed with ether. The filtrate and washings were concentrated a t 60 mm, and the residue was purified by preparative tlc (silica gel, hexane, Rr 0.86) to give 98 mg (95 %) of pure trans ,trans-2,6,1O-trimethyldodeca-2,6,10triene, which was homogeneous by vpc analysis (10 ft, 10% Carbowax 20-hI, 180", 60 ml NP,retention time 2.5 min). Bulbto-bulb distillation [bp 160" (30 mm) (bath temp)] gave 90 mg (87.4%) of the triene as a colorless oil (spectra unchanged): molecular ion at m/e 206, main peak a t m/e 69; nmr (CCl,) 6 1.57 (s, 12 H, CHIC=C), 1.63 (8, 3 H, CHsCzC), 1.9-2.1 (multiplet, 8 H, =CCH&H,C=), 4.8-5.2 (multiplet, 3 H , vinyl). An authentic sample was prepared by reaction of truns,truns-famesyl bromide with lithium aluminum hydride.
Registry No.-Geraniol, 106-24-1; farnesol, 106-28-5; benzyl alcohol, 100-51-6; 1-indanol, 6351-10-6. Acknowledgment.-This work was supported by the n'ational Science Foundation.
The Synthesis of 2-Amino-1-adamantanol WILLIAMV. CURRAN AND ROBERT B. ANGIER Organic Chemical Research Sectzon, Lederle Laborotories Division, American Cyanamzd Company, Pearl River, New York
Experimental Section General Method.--To a solution of the alcohol in tetrahydrofuran (THF) (freshly distilled from LiAlH,) was added a t 0" pyridine-sulfur trioxide complex (Aldrich), and the mixture was stirred at 0-3' for 3-20 hr until the reaction was complete. Then LiAlH, [or LiAlH,-AlCL (3: l)]in T H F was added at O", and the mixture wab stirred at 0" for 1 hr and then at 25' for 3-5 hr. After the excess of LiAlH, was destroyed with aqueous sodium hydroxide, ether and a measured amount of internal standard for vpc analysis were added. The precipitate was (1) E. J. Corey and X. Achiwa, Tetrahedron Lett., 1837 (1969). (2) See (a) H. 0. House, "Modern Synthetic Reactions," W. A. Ben-
jamin, Inc., New York, N. Y . , pp 34-39, 76-77; (b) L. F. Fieser and M. Fieser, "Reagents far Organic Synthesis." John Wiley & Sons, Inc., New York, N. Y., pp 582-5911, 1054; (c) R . H. Baker, L. S. Minckler, and A. 5. Hussey, J . Amer. Chem. Soc., 81, 2379 (1959); (d) E. L. Eliel, Ree. Chem. Prow., 2%, 129 (1961).
Received April 11, 1969
The chemistry of adamantane has received considerable attention in recent years. Numerous 1-substituted and some 2-substituted derivatives have been prepared. Herein we report the details of the first preparation of a simple 1,2-disubstituted adamantane. Since our preliminary communication2 two other reports on this subject have a p ~ e a r e d . ~ , ~ (1) For a recent review of the chemistry of adamantane, see R. C. Fort, Jr. and P. Von R . Schleyer, Chem. Rev., 64, 277 (1964). (2) W. V. Curranand R. B. Angier, Chem. Commun., 563 (1967). (3) M. A. McKervey, Chem. I n d . (London), 1791 (1967). (4) W. H. W. Lunn, W. D. Podmore, and 9. 9. Szinai, J . Chem. SOC.,C , 1657 (1968).
Vol. 94, No. 11, Noaember 1969
NOTES 3669
It has been abundantly established that carboniumion reactions and free-radical processes afford l-substituted adamantanes via the relatively low energy bridgehead intermediates. However, radical attack can also lead to 2 substitution.’ A possible route to the 1,2-disubstituted compounds would be an intramolecular cyclization of a suitable 1-substituted derivative onto the 2 position. For example, the insertion of a carbene, nitrene, or oxygen radical (Y in 1) onto the carbon in the 2 position to afford 2, followed by hydrolysis to 3, seemed a reasonable approach to this type of compound. The transition state leading to compounds of type 2 should be a cyclic six-membered ring and hence
SCHEMEI 0
0
II
OCNHNHz
I
B 7
6
1 0
II
0
II
L5
SH
XjC,Y
8
9
1
2
3
X=OorNH
..
Y =0; .. -N, ..
H
- A&-
I
12
H
oxycarbonyl azide to1 oxazolidinone 5.
NJ
OH
NH?
4
+o ):=o
11
Ho-t-
A
0
\
-c:
energetically favorable. We decided t o use the alkoxy nitrene route, since 1 (X = 0 and Y = - N : ) appeared to be readily accessible. When this problem was initiated we were not aware of any precedent for this type of ring closure. However, very shortly after embarking on this problem, Smolinsky and Feuer5reported the synthesis of 2-amino2-methyl-1-butanol (4), and this was followed shortly by two reports6,’ on the photolytic cyclization of t-butyl0 OAS,
I
& b
-t-H
0
singlet state of the nitrene 9.9 Acid hydrolysis of the oxazolidinone 10 with 2 N hydrochloric acid gave a good yield of the 2-amino-l-adamantano1(12)isolated as the hydrochloride salt. When more concentrated acid was used, 2-amino-1-chloroadamantane (13) was also obtained. A plausible mechanism for the formation of the chloro derivative is shown below.
5
1-Adamantyl p-nitrophenyl carbonate (6) was prepared and treated with methanolic hydrazine hydrate to give carbazate 78 which was in turn treated with nitrous acid to afford 1-adamantyloxycarbonyl azide (8) (Scheme I). Photolysis of compound 8 in a quartz reaction vessel a t 2537 A gave the desired adamanto[2,l-d]oxazolidin-2-one (10) in 45y0 yield via the intermediate nitrene 9. The carbamate 11 formed by insertion of the niitrene 9 into the solvent was also isolated from the reaction mixture. Both of these compounds, 10 and 11, are presumably formed from the 15) G.Smolinsky and B . I . Feuer, J . AmeT. Chem. Soc., 86, 3085 (1964). ( 6 ) R. Dreher and D. Kuhling, Angew. Chem. Intern. Ed. Engl., 3, 309 (1964). (7) R.Puttner and K . Hal‘ner, Tetrahedron Lett., 3119 (1964). (8) W. L. Haas, E. V. Krumkalns, and H. Gerzon, J . A m e r . Chem. Soc., 88,1988 (1966).
13 (9) W.Lwomki and F. P. Woerner, ibid., 87,5491 (1965).
3670
The Journal of Organic Chemistry
NOTES E:xperimental Section
All photolyses were carried out in quartz reaction tubes in a Rayonet photochemical chamber using low-pressure mercury lamps which provide 84% of their emission at 2537 A. 1-Adamantyl p-Nitrophenyl Carbonate (6).-To a solution of 1-adamantanol (3.0 g, 0.02 mol) and quinoline (2.4 ml, 0.02 mol) in 30 ml of methylene chloride was added a solution of p-nitrophenyl chloroformate (4.0 g, 0.02 mol) in 10 ml of methylene chloride over a 40-min period with stirring. The reaction mixture was allowed to stand a t room temperature over the weekend, and extracted with two 100-ml aliquots of water and two 100-ml portions of 0.25 iV hydrochloric acid. The organic phase was dried over magnesium sulfate and evaporated to give a crystalline solid, yield 6.1 g. This product was dissolved in 15 ml of hot methylene chloride, treated with Darco, and filtered. The filtrate was diluted with 25 ml of petroleum ether to afford crystals, yield 1.0 g, mp 132-136'. The infrared spectrum of this crop revealed it to be bis-p-nitrophenyl carbonate. The filtrate was evaporated to about 10 ml, diluted with 50 ml of petroleum ether, chilled, and filtered to give 2.8 g of product, mp 87-106'. The filtrate to this product was evaporated to 15 ml to give more crystals, yield 1.5 g, mp 101-106'. These two crops were combined, dissolved in 20 ml of methylene chloride, and applied to a column of acid-washed alumina (160 9). The column was eluted with 550 ml of methylene cliloride and the eluate was evaporated to give a crystalline solid which was recrystallized from methylene chloride-petroleum ether: yield 3.3 g (52%), mp 115-118.5' (lit.* 106-108"). For analysis a small portion of this product was recrvstallized from petroleum ether (melting raised to _ point 118.%-120"). Anal. Calcd for C:nHl,N05: C, 64.3; H, 6.0; N, 4.4. Found: C. 64.6: H. 6.4: N. 4.7. Adamantyl-1-carbazate (7).-l-Adamantyl p-nitrophenyl carbonate (3.8 g, 12 mmol) was added to 100 ml of methanol containing 1.0 ml (20 mmol) of hydrazine hydrate and refluxed for 3 hr on a steam bath. The reaction was evaporated to a yellow residue a t reduced pressure and 100 ml of ether was added to give a yellow solid which was removed by filtration and discarded. Tbe filtrate was warmed to effect solution and was extracted with six 75-ml aliquots of 10% sodium hydroxide solution and three 75-m1 portions of water. After having been dried over magnesium sulfate, the ether was evaporated to afford a white, crystalline solid, yield 2.5 g (95(z), mp 138-140' (1it.s 141-142'). For analysis a small portion of this crop was recrystallized from ethyl acetate (melting point unchanged). Anal. Calcd for CliHIBN202: C, 62.8; H , 8.6; N, 13.3. Found: C, 62.6; H, 8.6; X, 13.7. Photolysis of 1-Adamantyloxycarbonylazide ( 8 ) in Cyclohexane.-Adamantyl-1-carbazate (4.3 g, 20.4 mmol) was dissolved in 60 ml of acetic acid followed by the addition of 200 ml of water and 13 ml of concentrated hydrochloric acid. The solution was chilled in an ice bath and a cold solution of sodium nitrite (1.5 g, 21.8 mmol) in 15 ml of water was added. The resulting azide was extracted into cyclohexane (500 ml total), washed with two 250-ml portions of cold 5% sodium bicarbonate followed by two 250-ml portions of water, and dried over magnesium sulfate. The magnesium sulfate was filtered off and the filtrate photolyzed for 15.5 hr. The solution was evaporated to dryness a t reduced pressure and the oily crystalline residue was slurried in 50 ml of cyclohexane, chilled, and filtered to give 1.7 g (43y0) of the oxazolidin-2-one 10, mp 130-133.5". A small portion of this product was rerrystallized from ethyl acetate for analyses (mp raised to 134-136'): ir (KBr) 3.1, 5.7, and 5.8 p ; nmr (CDCla) 6 6.2 (NH), 3.7 (1 51 adjacent to NH); mass spectrum m/e 193 (molecular ion). Anal. Calcd for ChH15N02: C, 68.4; H, 7.8; N, 7.3. Found: C, 68.3; H, 5.7; N, 7.2. The filtrate from above was concentrated to 20 ml to give more crystals, yield 0.95 g, mp 86-93". Thin layer chromatography revealed this to be a mixture of compound 10 and another faster moving product. This compound was separated on Woelm alumina and crystallized from aqueous ethanol to afford 150 mg of carbamate 11: mp 34-87"; ir (KBr) 3.0, 5.94 p . Anal. Calcd for (217H27N02: C, 73.6; H, 9.8; N, 5.1. Found: C, 73.2; H, 91.9; N, 5.2. 2-Amino-1-adamantanol Hydrochloride (12).-One gram (5.2 mmol) of the oxazolidinone 10 was added t o 100 ml of 2 N hydrochloric acid and heated on a steam bath for 1.0 hr. The solution was filtered t o remove a small amount of insoluble ma-
terial and evaporated to a crystalline solid at reduced pressure with the aid of absolute ethanol. The compound was recrystallized from ethanol-ether to give a crystalline solid. The addition of ether to the filtrate gave more crystals. The infrared spectra and thin layer chromatograms of these two crops were identical. The total yield was 670 mg (63%): ir (KBr) 3.1, 5.05, 6.23, and 6.55 p; nmr (DeMSO) 6 8.1 ("a+), 5.2 (OH), and 3.0 (H adjacent to NH3+); mass spectrum m/e 167 (molecular ion). Anal. Calcd for CloH1,NO.HCl: C, 58.9; H, 8.9; N, 6.9; C1, 17.4. Found: C, 59.3; H, 8.7; N, 7.1; C1, 17.2. 2-Amino-1-chloroadamantaneHydrochloride (13).-Adamanto[2,1-d]oxazolidin-2-one (10, 300 mg) was heated on a steam bath for 2.5 hr in 30 ml of concentrated hydrochloric acid and evaporated to dryness a t reduced pressure. The residue was dissolved in 50 ml of absolute ethanol, evaporated to dryness, and redissolved in 15 ml of hot absolute ethanol. The solution was treated with Norit and.filtered. The filtrate was diluted with 10 ml of ether and chilled, and the product was collected: yield 135 mg. Elemental analyses and ir and nmr spectra identified this product as 2-amino-1-adamantanol hydrochloride. The filtrate was evaporated to a volume of several milliliters and ether was addedto afford more crystals, yield 47 mg, mp 308' dec. Chlorine analyses, along with ir, nmr, and mass spectra, identified this ir product as 2-amino-1-chloroadamantane hydrochloride: (KBr) 3.45, 14.35 p ; nmr (DsMSO) 6 8.6 (NH3+), 3.5 (H adjacent to ",+); mass spectrum m/e 185 (molecular ion) 187 (M 2 peak). Anal. Calcd for CloHleNCl.HC1: total C1, 31.9; ionic C1, 16.0%. Found: total C1, 31.7; ionic C1, 16.4.
+
Registry No.-10, 15252-86-5; 12,21347-36-4; 13,21371-75-5.
11, 15215-43-7;
Acknowledgment.-We wish to thank Mr. L. Brancone and staff for the microanalyses, and Mr. W. Fulmor and staff for the spectral data reported herein.
Reaction of Cyanoacetamide and Some 2-Acylcyclanones FILLMORE FREEMAN A N D THOMAS I. ITO Department of Chemistry, California State College, Long Beach, California 90801 Received March 13, 1969
Although the literature abounds in examples of basecatalyzed condensations of cyanoacetamide (I)with 1,3dicarbonyl compounds to give substituted pyridines, quinolines, isoquinolines, and 5H-l-pyrindines, 1-6 i t appears that this facile reaction has not been used to prepare the difficultly accessible 5H-2-pyrindine and cy~loalka[c]pyridine~~~ ring systems. We wish to report that I reacts with 2-acylcyclanone~,~ in the presence of diethylamine, to give substituted 5H-2-pyrindines (11-IV) and substituted cycloalka [c ]pyridines (1) F. Freeman, D. K. Farquhar, and R. L. Walker, J. Ow. Chem., 8 8 , 3648 (1968). (2) F. Brody and P. R. Ruby in "Heterocyclic Compounds," Part I , E. Kingsberg, Ed., Interscience Publishers, New York, N. Y., 1960, p 400. (3) H. 8.Mosher in "Heterocyolic Compounds," Vol. 1, R. C. Elderfield, Ed., John Wiley t Sons, Inc., New York. N . Y.,1950, p 469. (4) G . N. Walker and B. N. Weaver, J . Org. Chem., 26, 484 (1960); 26, 4441 (1961). (6) H . Junek, Monaleh. Cham., S I , 1201 (1984). (6) H. K . Sen and U. Bose, J . Indian Chem. SOC.,4, 51 (1927). (7) G . G . Hyist and K . Schofield, J . Chem. Soc., 3445 (1860). (8) A. T. Balahan and C. D. Nenitzescu, ibid., 3561 (1961). (9) 2-Acetylcyclohexanone has been reported1 to give the same substituted 3-isoquinolinol with malononitrile or I. I n contrast, Sen and Bose' reported that 2-acetylcyclohexanone and substituted 2-acetylcyclohexanones reacted with I to yield quinoline derivatives.
