Notes - The Nuclear Magnetic Resonance Spectrum of Feist's Acid

Theory-Guided Discovery of Unique Chemical Transformations of Cyclopropenes. Robin A. Weatherhead-Kloster and E. J. Corey. Organic Letters 2006 8 (1),...
4 downloads 0 Views 3MB Size
Download PDF
OCTOBER

!956

SOTES

Nhks

1169

A department for short papers of immediate interest.

The Nuclear 3Iagnetie Resonance Spectrum of Feist’s Acid‘

solution of excess sodium deuteroxide in heavy water. Three peaks of almost equal area were obtained (see Curve A) corresponding from left to right to the two methylene hydrogens, the two ALBERTT. BOTTIXI~ AXD JOHX D. ROBERTS carboxyl hydrogens (present as water) and the 1,2Receiwd January 6 , 1966 ring hydrogens respectively. The center peak was clearly due to H2O because it increased markedly Feist’s acid3 is generally accepted to be either I on addition of a drop of ordinary water to the soluCCOzH CHCOzH tion (compare Curve B). If the acid actually had Structure 11, the C-H peaks would be in the ratio //I / c02H CH,-CH CHSzC CH,-C of one (the ring hydrogen) to three (the methyl hydrogens).6 The S M R spectrum (Cume C) of the \CHCOzH ‘\CHco,H %HCozH hydrogenation product of Feist’s acid (111) is in I I1 I11 satisfactory agreement with the assigned structure or 11. Cogent evidence in favor of I has been ob- although, in this case, the three yarieties of hydrotained by Et tlinger4 and his structural assignment gen attached to carbon absorb too closely together has been (.onfirmed by x-ray diffraction ana1ysis.j to permit clear resolution. Unequivocal evidence that Feist’s acid is I and After five days a t room temperature, the XSfR not I1 has now been obtained from its nuclear mag- absorption of the 1,2-ring hydrogens of I decreased netic resonance (XMR) spectrum (Fig. 1) in a about 75% relative to the methylene hydrogens showing that the 1,2-hydrogens were exchanging (=CHZ) ( H Z O ) (C-H)2 with the solvent. S o additional bands were noted in the spectrum after 60 days at room temperature. A Removal of the solvent under reduced pressure, dissolution of the residue in fresh deuterium oxide, and equilibration reduced the absorption of the 1,2-hydrogens to negligible proportioils without effect on the absorption of the methylene hydrogens. The occurrence of hydrogen-deuterium exchange was confirmed by examination of the infraC red spectrum (potassium bromide pellet) of the deuterated acid and regeneration of 1 by equilibration of a sample of the deuterated acid with a solution of sodium hydroxide. The strong absorption band at 910 em.-’ characteristic of the protated acid was absent from the infrared spectrum of the CYCLES PERSECOND deuterated material. The specificity of the exchange FIG. ~.--SL-CI,EARMAGNETIC RESONAXCE S P E C T ROF ~ 5-mf. SAMPIES at 40 mc. and 9-100 gauss (12-sec. sweep) (Le., no exchange of the methylene hydrogens) with S’arian Associates Model V-4300B High Resolution precludes any possibility of a facile base-induced Spectrometei,. A, 0.19 g. of I dissolved in 0.46 g. of a solu- equilibration between forms I and I1 under these tion prepared from 0.12 g. of sodium and 0.8 g. of deuterium oxide; B, same solution plus one drop of ordinary xater; conditions. The spectrum of the deuterium oxide C, 0.14 g. of I11 dissolved in 0.36 g. of a solution prepared solution of I11 did not appear to change over 00 from 0.10 g. of sodium and 0.8 g. of deuterium oxide; D, days at room temperature. solution as for B after five days at room temperature. The recent reassertion that Feist’s acid is 11‘ because of its infrared spectrum, has been coil(1) Supported in part by the Petroleum Research Fund troverted by Ettlinger and Kennedy.6 Ozonization of the American Chemical Society. (2) National Science Foundation Predoctoral Fellow, of the diethyl ester of Feist’s acid is reported t o

7”

1954-1957. (3) F. Feist, Ber., 26, i 4 7 (1893); Ann., 436, 125 (1924). (4) M.G. Ettlinger, J . Am. Chem. SOC.,74, 5805 (1952). ( 5 ) Unpublished work by J. H. Sturdivant and D. R. Petersen; D. R. Petersen, Ph.D. Thesis, California Institute of Technology, 1955. h less complete analysis has been reported by D Lloyd, T. C. Downie, and J. C. Speakman, Chemistry & Tndrtstry, 222, 492 (1954).

( 6 ) Similar conclusions have been reached from NSfR spectra of dimethyl and diethyl esters of Feist’s acid by 31. G. Ettlinger and F. Kennedy, Chemistry &: Industry, 166 (1956). K e are grateful to Dr. Ettlinger for information regarding these results in advance of publication. ( 7 ) G . R. Boreham, F. R. Goss, and G. J. Minkoff, Chemzstry BC Industry, 1354 (1955).

1170

XOTES

21

VOL.

suspended in water, whereby dibenzyl was produced from benzyl chloride, trans-stilbene (11)from benzal chloride, and trans-a,a’-dichlorostilbene (IV) from benzotrichloride. This procedure has been applied by several other investigators t o the condensation of p-( a-halogenoalkyl)anisoles2 and their d e r i ~ a t i v e s , to ~ , ~prepare hexestrol and its derivatives. Fe

2 CsHjCHCh

+

Fe

I

z

I

-

+ Hn0

--+ C~H~CHCI-CHCICBH~

+ Hn0

C6HbCH=CHCe,H,j

(A)

I1 Fe

+ Hz0

Fe

+ HnO

2 CaHjCCls --+C~H~CCL-CCIZC~H~ --+

I11 C&,CC~=CC~C~H (B) S

IV 0’

-

CHs-C,

/ \

CYo2R , CC0,R

o /

(or polymeric ozonide)

0-0

IV

yield ethyl acetoxaloacetate and no formaldehyde as expected of esters of II.*However, it should be noted that methylenecyclopropane yields but 2% of formaldehyde on o~onization.~ Furthermore, if attack of ozone is electrophilic in character, esters of I might afford acetoxaloacetates by a sequence of reasonable reactions like the following (here presented in condensed form) which involves a rather common type of ring-opening process as the key step. IV can be regarded as the normal ozonide of 11. We are indebted to Dr. Martin G. Ettlinger for helpful discussions and generous samples of I and 111. CONTRIBUTION No. 2110 GATESA N D CRELLINLABORATORIES INSTITUTE OF TECHNOLOQY CALIFORNIA PASADENA, CALIFORNIA

It is natural to consider that the reaction of benzal chloride or benzotrichloride proceeds via path (A) or (B), respectively. The present paper describes the isolation of the intermediate a-stilbene dichloride (I) and tolane tetrachloride (111) in the very early stages of the dechlorination reaction of benzal chloride and benzotrichloride, respectively, Moreover, it has been ascertained that intermediate (I) on the further dechlorination with iron gave trans-stilbene (11) and intermediate (111) gax-e t~ans-a,oi’dichlorostilbene (IV). These condensations probably proceed by way of unstable organoiron chlorides as in the case of magnesium or zinc. But the organoiron compounds are so unstable and highly reactive toward another chlorine atom that they are difficult t o detect. It is of interest to note that the organoiron intermediate hardly reacts with water, and this property differentiates it from the organomagnesium or organozinc compound. The role of the water in the reaction is not only t o keep the fluidity of the mixture but t o prevent the occurrence of Friedel-Crafts reactions leading to polymeric substances. EXPERIMEXTAL

(8) F. R. Goss, C. K. Ingold, and J. F. Thorpe, J . Chem. SOC.,123, 327 (1923). (9) J. ‘I?. Gragson, K. W. Greenlee, J. M. Derfer, and C. E. Boord. J. Am. Chem. Soc., 75, 3344 (1953).

A Note on the Dechlorination Condensation of Benzal Chloride and Benzotrichloride by Iron and Water

All melting points are corrected. The syntheses under optimum conditions are described. or-Stilbene dichloride ( I ) f r o m beneal chloride. I n a flask equipped with a reflux condenser were placed water (200 cc.) and finely divided iron (reduced iron powder) (36 g., 0.632 mole). The flask was dipped in a boiling water-bath and to it was added beneal chloride (50.9 g., 0.316 mole) with vigorous stirring in one minute. The stirring then was continued for 3 minutes. The reaction mixture was extracted with benzene and the solvent was evaporated to dryness to produce crude material. The crude crystals were washed with petroleum ether, aqueous ammonia, and then

YOSHIROOGATAAND HIROSHINAUMURA

Received March 16, 1966

Several years ago, one of the authors reported on the dechlorination condensation by iron powder

(1) Ogata, Tsuruta, and Oda, Bull. Phys. Chem. Research (Tokyo), 21, 616 (1942) [Chem.Abstr., 43,2194 (1949)j. (2) Sisido and Nozaki, J . A m . Chem. SOC.,70, 778 (1948). (3) Sisido, Noeaki, and Kuyama, J . Org. Chem., 14, 1124 (1949). (4) Buu-HOTand Hoin, J . Org. Chem., 14, 1023 (1949).

OCTOBER

1956

with water, giving a-stilbene dichloride (3.8 g , 9.6%), which on recrystallization from ethanol gave needles melting a t 191-193" (literature m.p. 191-193°).6 No depression of the m. p. was observed on admixture \vith the authentic material prepared from hydrobenzoin and phosphorus pentachloride. trans-Stilbene (11) f r o m a-stilbene dichloride ( I ) . A mixture of a-stllbene dichloride (2.0 g.), iron Powder (20 g.1) and water ( I 00 cc.) was refluxed for 24 hours. The reaction product was extracted with benzene. After evaporating the solvent, the crude material was recrystallized from alcohol, yielding 1.0 g. (70%) of trans-stilbene, m.p. 122-124" and mixture m.p. 122-124'. Tolane tetrachlorzde (111) f r o m benzotrichloride. To a mixture of R-ater (50 cc.) and reduced iron powder (12.6 g., 0.225 mole) was added beneotrichloride (29.3 g., 0.15 mole) with vigorous agitation in a boiling water-bath and the reaction product was worked up as before. There was obtained 6 g. (25%) of tolane tetrachloride, which on recrystallization from acetic acid gave 4.8 g. of the pure material, m.p. 160162" literature m.p. 163',6 162',7 and 161.5').8 Anal. Calr'd for CI4H&IP: C, 52.49; H, 3.12; C1, 44.32. Found: C, 52.48; H, 3.25; C1, 44.00. trans-a,a '-dichlorostilbene (IV) f r o m tolane tetrachloride (111). The work-up with tolane tetrachloride (0.5 g.) was the same as with stilbene trom a-stilbene dichloride, except that the reattion was stopped after 4 hours. The a,a'dichlorostilbene melted a t 143-145' (literature m.p. 153',8 150°,8 143O,* 142','0) and gave no depression on admixture with an authentic sample; the yield was 0.25 g. (64%).

1171

NOTES

product, zolon red. This new dye was found to be a useful reagent for Though its analytical application Was d h " I d in s c " detail, little n'aS known about the Constitution Of zolon red and its silver complex; this is examined in the Tyork of this report. zOloll blue and zolon red have identical empirical and therefore appear to be isomericGaspar3 indicated that more than one tautomeric form may exist' for a given dye formula in this system. These variations are well known in l-phenyl3-methyl-pyrazolone-5.4 The most likely tautomeric forms of the zolon blue-zolon red systems are as follows:

I

I

/N\

:N

f

6-

N

8-

c=o

H?-C' CH,-C-C=C-C=C-C=C-C-~-CH~ I? 0

/I

ill

'N:

'I

H I H H H H I

Acknowledgments. The authors wish to express their thanks for the aid of the Hodogaya Chemical Go., Prof. 12. Oda, and Mr. J. Ishiguro in performing these experiments. DEPARTMENT OF INDUSTRIAL CHEMISTRY FACULTY OF EKGINEERINQ KYOTO UNIVERSITY KYOTO,JAPAN (5) Zincke, Ber., 10, 996 (1877). (6) Liebermann and Homeyer, Ber., 12, 1971 (1879); Blank, Ann., 248, 1 (1888). (7) Onufrowicz, Ber., 17, 833 (1884). (8) Marckwald and Karceag, Ber., 40, 2994 (1907). (9) Zinin, Ber., 4, 289 (1871); Limpricht and Schwanert, Ber., 4 , 379 (1871). (10) Staudinger, Ber., 49, 1969 (1916).

H H H H H H

(I) and (XI) have the longest conjugated chains and greatest .rr-electron mobility. I n the work reported here, corresponding difZolon Red and the Zolon Red-Silver Complex' ferences were observed in the visible and infrared absorption spectra of these dyes. Zolon blue abGEORGE F. SVATOS A N D JEROME GOLDENSON sorbs strongly a t 620 millimicrons and zolon red a t 522 millimicrons (Fig. 1). This hypsochromic shift Received M a y 7, 1966 occurs with decreasing chain length where a lower a-electron mobility exists.6v6I n acid solution both The monomethine oxonol derived from l-phenyl- compounds show a hypsochromic shift to 430 milli2-methyl-pyrazolone-5 was described by I(norr.2 microns. When neutralized with base these acidiLater, dyes such as zolon blue, (I), which contain fied solutions again show their characteristic 522 the (-GH==CH-). group were prepared and their and 620 millimicron bands indicating that the chromophoric activity d i s c ~ s s e d .While ~ working with zolon blue Gehauf and Goldenson' found that (4) P. Karrer, Organic Chemistry, Elsevier, New Pork, treatment with an alkaline solution produced a red (1) B. Gehauf and J. Goldenson, Anal. Chem., 27, 420 (1955). (2) L. Knorr, Ann., 238, 184 (1887). (3) B. Gaspar, U. 8. Patent 2,274,782; Chem. Absh-., 36, 4042 (1942).

1946, p. 773. ( 5 ) G. W. Kheland, The Theory of Resonance, Wiley, Yew Tork, 1944, pp. 137-158. (6) R. I) Stieglitz and Leech, Ber., 4 6 , 2147 (1913); J . =lm. ('hem. Soc., 36, 272 (1914)

OCTOBER

I956

ing unstable phosphorylated product is believed to undergo a concerted change involving loss of phosphonic acid and t,he simultaneous migration of a phenyl group from carbon to nitrogen. (CBH&C--NHOH CBHS

1187

NOTES

+ CH3PO(OC3H7-i)-C1+ H

\ I C:tjH~-C--N-O-PO /- L a CsHs

€1 [ (CsHs)z-C-K-C6H~

(OC3Hy-i) + AHa

11

OPO(OC3Hy-i)

/

CH3 (CsH&C=SC6H5

+

+ HOPO(OC3H7-i) I

CHH

An initial phosphorylation on the nitrogen rather than on the oxygen of the triphenylmethylhydroxylamine molecule followed by loss of phosphonic acid could also yield the same product. Triphenylmethylhydroxylamine, when treated with diisopropyl phosphorochloridate under the same conditions as isopropyl methylphosphonochloridate, did not react. EXPERIMENTAL

Reaction of amidoximes with Sarin, D F P , and the corresponding chloro analogs. The amidoxime6 was dissolved in the minimum amount of water and the solution was adjusted to p H 7.6. An equimolar amount of the fluorophosphate or the fluorophosphonate was added to this solution with stirring and a constant pH 7.6 was maintained by titration with alkali from a Beckmann Model K autotitrimeter. When the reaction was completed, the solution was made acid and the product was isolated by filtration or by extraction of the acidic solution with chloroform or ether. Recrystallization from appropriate solvents yielded the phosphonylated products. The same products could be obtained by reacting 1 mole of amidoxime with 1 mole of chlorophosphate or chlorophosphonate in a non-aqueous solvent in the presence of triethylamine. Several of the phosphorylated products were first isolated only as oils but were obtained crystalline after passage over a column of activated alumnia prior to recryst'allization. The physical and analytical data for the phosphorylated amidoximes a,re contained in Table I. Preparation, of iV,N-dimethylbenzamidoxime. To a stirred solution of benzohydroximyl chloride (15.5 g., 0.1 mole) dissolved in 30 cc. of absolute alcohol maintained at 0" was added 50 cc. of a solution of dimethylamine, (9.9 g., 2.2 moles) in absolute alcohol. The mixture was kept at 0" for 30 minutes and then was allowed to stir a t room temperature in a stoppered filtering flask for 24 hours. Part of the alcohol was removed in vacuo. Cooling yielded a solid which was filtered and washed with cold alcohol. On recrystallizat,ion from alcohol the solid melted a t 120". Anal. Calc'd for CsHI2Y20:C, 66.0, H, 7.3. Found: C, 66.0, H, 7.4. This material was reacted 'with the chloro derivative of Sarin (Table I). Rearrangement of Sarin-phosphorylated benzamidoxime to (6) Acetamidoxime: Michaelis, Ber., 24, 3439 (1891); Benzamidoxime: Kruger, Ber., 18, 1053 (1885); Nicotinamidoxime: Nordmann, Ber., 17,2746 (1884).

phenylurea. Sarin-phosphorylated benzamidoxime (4.0 9.) was placed in 15 cc. of water and was refluxed for 1 hour. After separating the water solution from an oily material and cooling, 150 mg. of a solid was obtained melting a t 147" whose analysis corresponded to that of phenylurea. Anal. Calc'd for C,&N20: C, 61.8, H, 5.90, N, 20.6. Found: C, 62.0, H, 5.8, N,20.7. The reddish-brown oil which also formed was not identified. Rearrangement of triphenylmethylhydroxylanzine Lo benzophenone anil. Isopropyl methylphosphonochloridate (1.57 g., 0.01 mole) was added dropwise to a stirred, cooled solution of triphenylmethylhydroxylamine (2.75 g., 0.01 mole) in benzene in the presence of triethylamine (1.01 g., 0.01 mole). The mixture then was allowed to come to room temperature and was kept a t room temperature for two hours. The amine hydrochloride was filtered and the solution was concentrated to dryness. The residue was crystallized from absolute alcohol and yielded 1.2 g. of a solid of m.p. 1P1". This solid did not contain phosphorus and gave a negative Tollens test. Elemental analysis conformed with that of benzophenone anil. Anal. Calc'd for CIgHlsN: C, 88.5, H, 5.85, X, 5.43. Found: C, 88.3, H, 5.80, N, 5.10. I n order to prove that the substance of m.p. 111" was the anil, 0.2 g. was hydrolyzed with 18% hydrochloric acid to benzophenone and aniline. The former was isolated as its 2,4-dinitrophenylhydrazone,the latter as its benzenesulfonanilide. Mixture melting point determination with authentic samples of these derivatives gave no depression. Triphenylmethylhydroxylamine when treated with diisopropyl phosphorochloridate under the same conditions as isopropyl methylphosphonochloridate did not react.

Acknowledgment. The authors wish to thank the Analytical Research Branch of the Research Directorate, Chemical Warfare Laboratories for performing the analytical determinations. BIOCHEMICAL RESEARCH DIVISION CHEMICAL WARFARE LABORATORIES ARMY CHEMICAL CENTER, hxARYLAND

Unsaturated Amines. IX. Through BisEnamines to Aromatics' XELSONJ. LEONARD AND ROKALD R. SAUERS~

Received i V a y SI, 1966

The preparation of enamines3 by the reaction of ketones with piperidine and pyrrolidine suggested an application to the synthesis of substituted p phenylenediamines from 1,4-cyclohexanedione. The heating of a mixture of 1,4-cyclohexanedione (I) and pyrrolidine, with collection of the theoretical amount of water, gave a product exhibiting the in(1) Article VI11 in this series: S . J. Leonard, L. A. Miller, and P. D. Thomas, J . Am. Chem. Sor., 78, 3463 (1956). (2) Xational Science Foundation Fellow, 1964-1955. (3) C. Mannich and H. Davidsen, Ber., 69, 2106 (1936); F. W.Heyl and M. E. Herr, J . Am. Chem. SOC.,75, 1918 (1953): Ill. E.Herr and F. W. Hevl. J . Am. Chem. SOC., 75, 5927 (1953); G. Stork, R. Terreli, and J. Szmuszkovicz, J . Am. Chem. SOC.,76,2020 (1954); J. L. Johnson, M. E. Herr, J. C. Babcock, R. P. Holysz, A. E. Fonken, J. E. Stafford, and F. W. Heyl, J. A m . Chem. Soc., 78,430 (1956).

1188

VOL.

KOTES

21

stability usually associated with an enamine func- A mixture of 20 g. (0.16 mole) of I,4-dichlorobutane, 5.5 g. tion. Snalysis suggested the composition C14H22ni2 (0.05 mole) of p-phenylenediamine, and 0.5 g. of anhydrous zinc chloride was heated under reflux for 4 hours. Treatment and therefore the bis-enamine structure IIa, but at- with excess 10% aqueous ammonium hydroxide followed by tempted purification a t this stage resulted in oxi- separation and evaporation of the organic layer yielded 1.0 g. dative transformation. Intentional air-oxidation (9%) of crude 1,l'-p-phenylenedipyrrolidine. Sublimation yielded the benzenoid product, 1,l'-p-phenylene- gave pure material with the same physical constants as for the product described above. The melting point of dipyrrolidine (IIIa), the structure of which was those mixtures of 1,l'-p-phenylenedipyrrolidine from the two checked by synthesis from p-phenylenediamine sources was not depressed. Mixtures of the corresponding and 1,4-dichlorobutane. It was also possible to pre- dipicrates were likewise undepressed in melting point.

I1

I11

( a , x = l ; b, X-2)

pare 1,l'-p-phenylenedipiperidine (IIIb) from 1,4cyclohexanedione and piperidine by the air-oxidation of the bis-enamine intermediate IIb. EXPERIMENTAL

Reaction of 1,4-~yclohexanedionewith pyrrolidine. A soliition of 11.2 g. (0.1 mole) of 1,4-cyclohexanedione in 250 mI. of thiophene-free benzene, to which 28.4 g. (0.4 mole) of pyrrolidine had been added, was heated under reflux in a nitrogen atmosphere for one hour, during which time the theoretical amount of water (3.6 ml.) was collected in a DeanStark trap. Evaporation of the benzene in a vacuum yielded 17.5 g. (8070) of a dark red solid, which on sublimation became colorless, m.p. ca. 137" (dec.), and remained so on recrystallization from ether a t Dry-Ice temperature, m.p. ca. 144" (dec.). The analysis was slightly low in carbon and hydrogen for C1,H2&;n, and the compound appeared to pick up oxygen very rapidly, with coloration. The ultraviolet absorption spectrum in hexane solution exhibited maxima at 272 mp (log e 4.18), 268 mfi (log e 4.15), and 340 mp (log B 3.39). The infrared spectrum showed a peak a t 1633 and two near 800 cm.-l in addition to those present in the aromatized structure (see below). Aromatization was effected by bubbling dry air through a benzene solution of the crude diene a t 25' for 18 hours. Evaporation of the solvent was followed by sublimation of the product as colorless needles, m.p. 148-150' (dec.); Ahexam mal 270 mp, log E 4.46; 267 mp, log e 4.42; and 340 mp, log E 3.62.4 The infrared spectrum in Nujol was clear above 3060 cm." and shovred maxima (selected) a t 1593 (w), 1531 (s), 1487 ( w )and 1470 (s) cm.-'. From 1.0 g. of crude bis-enamine IIa there was obtained 0.5 g. of 1,l'-pphenylenedipyrrolidine. Anal. Calc'd for C1,&ON2: C, 77.73; H, 9.32; K, 12.95. Found: C, 77.56; H, 9.17; N, 13.23. The dipicrate crystallized as yellow needles from absolute ethanol, m.p. 147-147.5' (dec.). Anal. Calc'd for C Z ~ H J T ~C,O46.29; ~ ~ : H, 3.89; N, 16.61. Found: C, 46.38; H, 3.83; N, 16.76. 1,I '-p-Phenylenedipyrrolidine from p-phenylenediamine. ( 4 ) P. Grammaticakis, Bull. SOC. chim. France, 534 (1951 ) reported " : :A 263 my, log E 4.16, and 312 mp, log E 3.16, for N,N,X ',9 '-tetramethyl-p-phenylenediamine.

Reaction of 1,4-cyclohexanedionewzth piperadme. A solution of 2.3 g. (0.02 mole) of 1,4-cyclohexanedione and 6.8 g. (0.08 mole) of piperidine in 50 ml. of benzene was heated under reflux in a nitrogen atmosphere for 5 hours. About 0.4 ml. (557,) of water was collected. Evaporation of the benzene yielded a red oil which solidified on cooling. The ether-soluble portion of the residue was sublimed, giving 1.0 g. (22%) of colorless needles, m.p. 142-144" (dec.) (analysis slightly low in carbon and hydrogen for CISH26N2). Aromatization was effected in 5770 yield from the crude bis-enamine I I b by air-oxidation. The pure l,l'-pphenylenedipiperidine was obtained by sublimation as colorless needles, m.p. 108-109". Anal. Calc'd for C16H21N2: C, 78.63; H, 9.90; N, 11.47. Found: C, 78.34; H, 9.64; N, 11.52. The dipicrate crystallized as yellow plates from absolute ethanol, m.p. 192-192.5' (dec.). ilnal. Calc'd for C28H30N801d: C, 47.86; H, 4.30. Found: C, 48.04; H, 4.38. 1,I '-p-Phenylenedipiperidine from p-phenylenedaamtne. A mixture of 2.3 g. (0.022 mole) of p-phenylenediamine, 10.3 g. (0.045 mole) of pentamethylene dibromide, 4.77 g. (0.045 mole) of anhydrous sodium carbonate, and 50 ml. of dry toluene was heated under reflux for 21 hours. Strong aqueous sodium hydroxide was added to the solid phase, and the mixture was extracted with three 50-ml. portions of toluene. The combined toluene extracts were evaporated, giving 3.0 g. of brown powder. The ether-soluble portion yielded about 0.3 g. (6%) of sublimate, m.p. 108-109°, identical with the product described above. The dipicrates were also identical by the usual criteria.

THENOYES CHEMICAL L.4BORATORY UNIVERSITY OF ILLINOIS URBANA, ILLIXOIS

Phenazine Syntheses. IX.' 1-Halogenophenazines DONALD L. VIVIAN Received M a y 31, 1966

In the course of extending ring closure through the nitro group2 to a number of representative phenazines, 1-bromo- and 1-iodophenazine and several alkoxy derivatives of these have been prepared. The syntheses were all made through the 6-halogeno-2-nitrodiphenylamines,as shown by the example :

(1) Paper VIII, J. Org. Chem., 21, 1030 (1956). (2) TVaterman and T'ivian, J. Org. Chem., 14, 289 (1949).

OCTOBER

1956

NOTES

With the inclusion of the results given herein, seven of the eight possible 1- and 2-halogenophenazines now are known;3 only 1-fluorophenazine has not as yet been described. EXPERIMENTAL

1-Bromophenazine. (a). 1-Bromo-b,S-dinitrobenzene. While this compound has been described both by Wender4 and by van Duin and van Lennep,6 it is felt that a more detailed account of the preparation should be given. A solution of 18.3 g. of 2,3-dinitroaniline6 in 200 cc. of glacial acetic acid was added dropwise to a stirred and cooled solution of 7.7 g. of NaNOz in 55 cc. of conc'd H2S04, the temperature being held a t 15-20'. The resulting diazonium solution then was added over a five-minute period to a stirred solution of 15 g. of CuzBr2,80 cc. of 487, HBr, and 80 cc. of water, which was maintained a t 70-80' during the addition. The cooled solution then was poured into about 2 1. of water, and the precipitate was washed free of acid, giving a quantitative yield of 24.7 g. of l-bromo-2,3-dinitrobenzene, shrinking a t 99", and melting a t 101-102.5°;7 lit.,* m.p. 101.5'. Recrystallized from alcohol, this product melted a t 102.5-103.5'. (b). 6-Bron~o-2-nitrodiphenylamine.A mixture of 18.5 g. of l-bromo-2,3-dinitrobenzene and 26.5 cc. of aniline was refluxed in 375 cc. of absolute ethanol for a period of 3 weeks, and all unchanged material then was removed by steamdistillation. The residue was extracted with ether, the ether was evaporated, and the product was chromatographed on alumina from a benzene solution. Extraction from the alumina by benzene gave 13.2 g. of a very hard, dark red solid melting a t 50-52 '. Two crystallizations from absolute ethanol gave cinnamon-red microcrystals, m.p. 64-65 Anal.* Calc'd for ClzHgBrNzO2:C, 49.2; H, 3.09. Found: C, 49.2; H, 3.24. (c). I-Bromophenazine. A mixture of 2.0 g. of 6-bromo-2nitrodiphenylamine, 2.6 g. of ferrous oxalate dihydrate, and 20 g. of granulated lead in an open flask was heated for 6 minutes in an oil-bath a t 267-270'. Sublimation from the broken-up reaction mixture a t 270' and about 0.5 mm. gave 1.26 g. of yellow product melting a t 126-130°, after earlier softening. Recrystallized twice from n-hexane, the compound formed small sulfur-yellow needles, m.p. 132-134". Anal. Calc'd for Cl2H7BrN2:C, 55.6; H, 2.72. Found: C, 55.9; H, 2.85 I-Zodophenazine. (a). 2,s-Dinitro-1-iodobenzene. This compound is described by \;Vender4as melting at 138"; as in the case of the bromo compound, it is felt that more details of the method of preparation will be helpful. A diazotization of 18.3 g. of 2,3-dinitroaniline was carried out as given above, and the diazonium solution was added over a period of about 5 minutes t o a mechanically-stirred solution of 17 g. of K I in 180 cc. of water, maintained a t about 70". Pouring the resulting solution into an excess of water, filtering, and washing the product free of acetic acid, followed by abdrying, gave 23.3 g. (79%) of crude l-iodo2,3-dinitrobenzene, melting a t 133-137". Recrystallization from absolute ethanol (cooling to -50") gave a product with m.p. 138-140'. (b). 6-Zodo-%nitrodiphenylamine. This compound was prepared by refluxing 11.8 g. of 2,3-dinitro-l-iodobenzene

".

(3) Vivian and Hartwell, J . Org. Chem., 18, 1065 (1953). (4) Wender, Gazz. chim. ital., 19, 230 (1889). (5) van Duin and van Lennep, Rec. trav. chim., 38, 364 (1919). (6) Vivian, Hartwell, and Waterman, J . Org. Chem., 20, 800 (1955). (7) All melting points given by the author are corrected. (8) Microanalyses by the Microanalytical Laboratories of the National Institutes of Health, under the direction of Dr. W. C. Alford.

1189

and 10.5 cc. of aniline in 215 cc. of absolute ethanol for 8 days. Steam-distillation left a residue which was taken up in hot alcohol, treated with carbon (Darco), and filtered, giving 13.5 g. of a crude red solid. This on recrystallization from ethanol gave orange plates, m.p. 73-75'. Anal. Calc'd for Cl2HsIN2O2:C, 42.4; H, 2.67. Found: C, 42.4; H, 2.90. (c). 1-Iodophenazine. A mixture of 1.0 g. of 6-iodo-2-nitrodiphenylamine, 1.3g. of ferrous oxalate dihydrate, and 10 g. of granulated lead was heated in an open flask in an oil-bath a t 268-270" for 7 minutes. On sublimation from the ground mixture at 270' and about 0.5 mm., there was obtained 0.48 g. of crude product. Recrystallized from 3 cc. of chloroform, 0.23 g. of the compound was obtained as short, medium yellow rods, m.p. 142-144'. Anal. Calc'd for CIIH;IN2: C, 47.1; H, 2.30. Found: C, 47.0; H, 2.54. 1-Bromo-7-methoxyphenazine. (a). 6-Brclmo-4'-methoxy-bnitrodiphenylamine. A mixture of 12.4 g. of l-bromo-2,3-dinitrobenzene, 18.5 g. of p-anisidine, and 250 cc. of absolute ethanol was refluxed for four weeks. The mixture then was poured, with stirring, into a solution of 100 cc. of HC1 in 400 cc. of HzO, the resultant mixture extracted twice with ether, and the ether washed to neutrality. The black, semiliquid residue from evaporation of the ether was dissolved in benzene and chromatographed on alumina. Elution with benzene and evaporation of the solvent gave a liquid which soon solidified (on scratching) to 13.7 g. of a dull red solid. Two crystallizations from ethanol gave dull red microcrystals, m.p. 65-67'. Anal. Calc'd.for C13HllBrX203:C, 48.3; H, 3.43. Found: C. 48.2: H. 3.70. ' ( b ) . l-B;omo-7-methozyphenazine.A mixture of 4.0 g. of 6-bromo-4'-methoxy-2-nitrodiphenylamine,5.2 g. of ferrous oxalate dihydrate, and 40 g. of granulated lead heated for 9 minutes in an oil-bath a t 270" (internal temperature reached a maximum of 281") gave 2.6 g. of yellow product on vacuum-sublimation a t about 0.5 mm. from a bath a t 270'. Recrystallized twice from benzene, this formed small, pale yellow needles, m.p. 170.5-171'. Anal. Calc'd for ClsH9BrN20: C, 54.0; H, 3.14. Found: C, 54.0; H, 3.20. 1-Zodo-7-methoxyphenazine. (a). 6-Zodo-~'-methoxy-b-nitrodiphenylamine. This substance was prepared similarly to the bromo compound, from 14.7 g. of 1,2-dinitro-3-iodobenzene and 18.5 g. of p-anisidine, refluxed 4 weeks in 250 cc. of absolute ethanol. The yield was 12.0 g. of a dark red solid, after chromatography. Three recrystallizations from absolute ethanol gave dull red microcrystals. m.p. 79.581.5'. Anal. Calc'd for Cl~HllINzOI: I, 34.3. Found: I, 34.0. (6). 1-Zodo-7-methoxyphenazine. The size of the reaction flask made considerable difference in the yield obtained from ring closure on the above intermediate. When 2.0 g. of 6iodo-l'-methoxy-2-nitrodiphenylamine was mixed with 2.6 g. of ferrous oxalate dihydrate and 20 g. of granulated lead in a IO-cc. Erlenmeyer flask, and heated for 8 minutes in a bath at 275-278', vacuum-sublimation from the mixture gave 0.93 g. of yellow product. Repeated, but in a 25-cc. Erlenmeyer, the yield was only 0.55 g., and the product was much more strongly colored. Recrystallized twice from benzene, the compound forms light yellow small needles or prisms, m.p. 178-180.5'. Anal. Calc'd for ClsHaINzO: C, 46.4; H, 2.70. Found: C, 46.5; H, 3.05. 1-Bromo-7-ethoxyphemzine. (a). 6-Brom0-4~-ethosy-2-nitrodiphenylamine. This compound was prepared similarly to the methoxy homolog. From 9.9 g. of l-bromo-2,3-dinitrobenzene, there was obtained 9.6 g. of dark red material, which was purified by further chromatography on alumina from benzene solution. In this manner were obtained dark purplish-red microcrystals, m.p. 81-82'. Anal. Calc'd for C14H13BrNz03: C, 49.9; H, 3.88. Found: C,50.5; H, 3.96.

1190

VOL.

SOTICS

( b ) . 1-Bromo-7-ethoxyphenazine. When 1.0 g. of 6-bromo4'-ethoxy-2-nitrodiphenylamine was mixed with 1.3 g. of ferrous oxalate dihydrate and 10 g. of granulated lead, heating the mixture for 12 minutes in a bath at 255-260" gave 0.63 g. product on vacuum-sublimation from the whole ground sublimation mixture a t 270" and about 0.5mm. This product had a good deal of red material in it, however, so it was again subjected to treatment with ferrous oxalate dihydrate and granulated lead, as before, giving 0.50 g. of lighbyellow microcrystals, m.p. 156-157", on purification by vacuum-sublimation at about 0.5 mm., from a bath at 270". Anal. Calc'd for C14H1,BrN20:C, 55.5;H, 3.66. Found: C, 55.9;H, 3.80. LABORATORY OF CHEMICAL PHARMACOLOGY INSTITUTE NATIONAL CANCER BETHESDA 14, ~ ~ A R Y L A N D

Ethyl 4-Formylphenoxyacetate DALE1;.ROBERTSON

Received M a y S1, 1956

The preparation of ethyl 4-formylphenoxyacetate (I) from 4-formylphenoxyacetic acid and ethyl alcohol has been reported.' The product was impure, disintegrating at 100" and becoming completely liquid at 155". We wish to report the preparation of I from phydroxybenzaldehyde and ethyl bromoacetate using the potassium carbonate method employed by A. Robertson2 for the ortho isomer. Our product, after recrystallization, melted a t 42-42.5" and the analyses for it and the oxime derivative were consistent with the values calculated for I and its oxime. The infrared spectrum of the product indicated the presence of ester, arylaldehyde, and phenoxy groupings and para substitution. We submit t,his evidence in correction of the prior literature on the compound I. EXPERIMENTAL

Ethyl 4-fomylphenoxyacetate ( I ) . Ethyl bromoacetate (167 g.), p-hydroxybenzaldehyde (122 g.), potassium carbonate (138 g.), and dry acetone (500 ml.) were stirred vigorously under reflux conditions for two hours. The mixture then was poured, with stirring, into two liters of icewater. The resulting solid was filtered, washed well with ice-water and dried (1) by pressing with a rubber dam and (2)over calcium chloride in a vacuum desiccator. The light yellow product melted a t 35-38" and amounted to 175 g. (84.1%). One crystallization from ethanol raised the m.p. to a constant 42-42.5" (uncorr.). Anal. Calc'd for Cl1H12O4:C, 63.45:H, 5.81. Found: C, 63.53;H, 5.59. Oxime. The oxime was prepared according to method A in Shriner and Fuson, Identifiation of Organic Compounds, Second Edition, John Wiley and Sons, New York, X. Y., 1940, page 167. The oxime (white needles from alcoholwater) melted a t 61-62" (uncorr.).

(1) Elknn, Ber., 19, 3042 (1886). (2) Robertson, J . Chem. SOC.,489 (1933).

21

AmE. Calc'd for Cl1Hl3NO4:C, 59.19; H, 5.87. Found: C, 58.81;H, 5.60. BIOCHEMICAL RESEARCH DEPARTMENT THEDow CHEMICAL COUPANY MIDLAND, MICHIGAN

The Chlorination of Phenyldimethylsilane' GLENA. RUSSELL

Received J u n e 4, 1966

It has been shown that a t 80" the relative reactivities of the a- and P-hydrogen atoms of cumene towards a chlorine atom are 12.8: 1.2 It would be of interest if similar data on the reactivities of the hydrogen atoms of the silicon analog of cumene, phenyldimethylsilane, were available. Since silicon-hydrogen bonds usually are weaker than carbon-hydrogen bonds13 it is expected that generally a chlorine atom will react with a siliconhydrogen bond in preference to a carbon-hydrogen bond. I n the case of cumene and its silicon analog this generality may not apply since the 2-phenyl2-propyl radical undoubtedly possesses considerably more resonance stabilization than the phenyldimethylsiIy1 r a d i ~ a l In . ~ fact, towards the highly selective peroxy radical tthe a-hydrogen atom of cumene is about 200 times as reactive as the ahydrogen atom of phenyldimethylsilane, presumably because of this resonance e f f e ~ t . ~ When phenyldimethylsilane was photochemically chlorinated in the liquid phase at 80" we observed that substitution a t the a-position occurred 30 times as readily as substitution at the 0-position. However, further investigation demonstrated that the substitution reaction occurred as readily in the dark as in the presence of intense illumination. Apparently an ionic substitution reaction occurred so readily that the free radical reaction was completely overshadowed. The cleavage of silicon-hydrogen bonds by bromine in the dark has been observed previously.6 Presumably ionic substitution by chlorine occurs in a similar manner involving (1) Directive Effects in Aliphatic Substitutions. VIII. (2) G. A. Russell and H. C. Brown, J . Am. Chem. SOC.,

77, 4031 (1955). (3) Unfortunately, bond dissociation energies for various silicon-hydrogen bonds are not available. The silicon-hydrogen bond strength of silane (80.8 kcal. mole-I) is 10 kcal. mole-' less than the carbon-hydrogen bond energy of methane (90.8kcal. mole-') [M L. Buggins, J . Am. Chem. Soc., 75, 4123 (1953)l. (4) H. Gilman and G. E. Dunn, Chem. Revs., 52, T7

(1953). (5) G. A. Russell, J . Am. Chem. SGC.,78, 1047 (1956). (6) A. Stock and C. Somieski, Ber., 50, 1739 (1917);C. Eaborn, J. Chem. Soc., 2755 (1949);TiT. H.Nebergall and 0. H. Johnson, J . Am. Chem. Soe., 71, 4022 (1949);W.H. Nebergall, J . Am. Chem. SOC.,72, 4702 (1950).

OCTOBER

1956

NOTES

1191

Anal. Calc'd for CaHlzSi:C, 70.5; H, 8.8; Si, 20.6. Found: C, 70.5; H, 9.1; Si, 20.6. The molar refraction, 45.62, was in excellent agreement with the value of 45.73 calculated from the data of Warrick.12 Chlorination procedure. The chlorination procedure was similar to that used in the chlorination of cumene.2 PhenylC B H ~ S ~ ( C H J , H2 C12 -+ dimethylsilane was placed in a 200-ml. three-necked flask I containing a True-Bore stirrer, a gas inlet capillary tube, CsH5 CeHiSi(CHI)2C1 HC1 C12 and a Dry Ice condenser. The amount of chlorine used was estimated from measurement of the volume of liquid chlorine I n both the photochemical and dark reactions of a t -34'. The chlorine from a calibrated tube was allowed to vaporize into a stream of dry nitrogen and after passing chlorine with phenyldimethylsilane it was observed through a 500-ml. gas mixing bulb was introduced in the that only substitution occurred; one mole of hy- phenyldimethylsilane through the capillary tube. The drogen chloride and one mole of alkyl and silyl flask was maintained a t 80 f 1" by a water-bath and %as chlorides were formed for each mole of chlorine illuminated by a 300-watt bulb approximately 2 inches from added. This demonstrates that addition of chlorine the flask. A large excess of nitrogen was used and after passing through the Dry Ice condenser, hydrogen chloride to the aromatic ring did not occur and that cleavage was removed from the nitrogen by a sodium hydroxide trap. of the silicon-phenyl bond by chlorine,* or by hy- Chlorine could not be detected in the exit gases from the drogen ~ h l o r i d e ,or ~ the cleavage of the silicon- chlorination flask. After the desired amount of chlorine had hydrogen bond by hydrogen chloride,'O did not been introduced over a one-hour period, the solution was degassed for several hours with dry nitrogen to remove the compete with the cleavage of a silicon-hydrogen last trace of hydrogen chloride. The sodium hydroxide trap bond by chlorine. If any of these reactions had was analyzed for chlorine content as was the chlorination occurred less than one mole of hydrogen chloride product. The chlorination product was hydrolyzed by shaking with would have been formed per mole of alkyl and silyl mater and the aqueous layer analyzed for chloride ion. chlorides. Unreacted phenyldimethylsilane had to be completely reThe fact that the chlorination product formed in moved from the aqueous layer since it reduced silver ion. the absence of illumination could be quantitatively The same amount of chloride ion was found when the hyand easily hydrolyzed indicates that the reaction drolysis was conducted for l/2, 2, or 24 hours. Chlorination results. Photochlorination of 0.24 mole of product was phenyldimethylchlorosilane and that a t 80" with 0.060 f ,006 mole of aromatic substitution did not occur. The small phenyldimethylsilane chlorine yielded 0.0538 mole of hydrogen chloride. The amount of unhydrolyzable chloride formed in the chlorination product contained 0.0539 mole of alkyl and photochemical chlorination has thus been attributed silyl chlorides from which 0.0523 mole of chlorine ion was to free radical attack on the methyl groups of immediately liberated upon hydrolysis. The remaining 0.0016 mole of chlorine was not ionized by prolonged conphenyldimethylsilane. tact with water. The hydrolyzable chloride is undoubtedly phenyldimethylchlorosilane while the unhydrolyzable chloride is most likely phenyl( chloromethy1)methylsilane. EXPERIMENTAL When the above experiment was repeated in the absence of Preparation of phenyldimethylsilane. Phenyldimethyl- illumination the amounts of hydrogen chloride, total alkyl chlorosilane was prepared by Dr. J. R. Ladd by the reaction and silyl chlorides and hydrolyzable chloride vere identical within experimental error. This result indicates that the of methylmagnesium chloride and phenyltrichlorosilane. Material boiling a t 85.5" a t 21 mm. was reduced by lithium 0 0016 mole of unhydrolyzable chloride formed in the photoaluminum hydride. To 10 g. of lithium aluminum hydride in chemical reaction is not a product of ionic aromatic sub200 ml. of ethyl ether 128 g. of phenyldimethylchlorosilane stitution, but most likely results from the attack of a was added over a three-hour period. The solution was re- chlorine atom on a methyl group of phenyldimethylsilane. fluxed for three hours before the excess lithium aluminum GENERALELECTRIC RESEARCH LABORATORY hydride was destroyed by ethyl acetate. The reaction mixSCHEKECTADY, NEW YORK ture was poured over ice, the ether layer separated from the aqueous layer, and the aqueous layer extracted twice with (12) E. L. Warrick, J . Am. Chem. SOC.,68,2455 (1946). 200 ml. of ether. The etheral solution was dried over sodium sulfate and distilled until a pot temperature of 85" was reached. The residue did not give a qualitative test for chlorine. The residue upon distillation through a small packed column (ca. 5 plates) under nitrogen gave 99 g. Polynitrogen Systems from the Hydrazino(977,) of phenyldimethylsilane, b.p. 57" a t 20 mm., n': carbonic Acids. Part VII. Some Reactions of n'; 1.4995, dy 0.889. 1.4985, di" 0.876;

electrophilic attack on hydrogen and nucleophilic attack on silicon.'

+

L

+

I

+

1-Phenyl-5-methylmercaptotetrazole

(7) D. R. Deans and C. Eaborn, Research, 5 , 592 (1952); F. L. SCOTT,'F. C. BRITTEN,AND J. REILLY J . Chem. SOC.,3169 (1954). (8) B. 0. Pray, L. H. Sommer, G. M. Goldberg, K. T. Received June 6, 1966 Kerr, P. A. 1 3 Giorgio and F. C. Whitmore, J. Am. Chem. SOC.,70, 433 (1948). (9) H. Gilman and F. J. Marshall, J . A m . Chem. SOC.,71, In connection with other work, it became of in2066 (1949). terest to examine some properties of l-phenyld(10) H. E. Opita, J. S. Peake, and W. H. Nebergall, J . (1) To whom inquiries concerning reprints are to be sent. Am. Chem. Soc., 78, 292 (1956). (11) R. A. Benkeser and D. J. Foster, J. Am. Chem. SOC., Present address, Chemistry Department, University of California, LOSAngeles 24, California. 74, 5314 (1952).

1192

NOTES

methylmercaptotetrazole (IB). The alkylmercapto group of IB, as with the parent compound IAI2 was resistant to displacement by ammonia or amines under a variety of c o n d i t i o ~ i sCompound .~~~ I B underwent nitration in a mixture of concentrated nitric arid sulfuric acids to give the p-nitro derivative (IC). The position of the nitro group was proven by cleavage of I C with hot concentrated hydrochloric acid to yield p-nitroaniline. The influence of the tetrazole group5 in directing substitution t o the para position of the phenyl ring has been reported earlier, again under strongly acidic conditions, by Garbrecht and Herbst.* The nitro group of I C was reduced with aluminum amalgam, and the resulting amino group underwent diazotization in the normal manner. The diazonium compound was allowed to couple with ,&naphthol, dimethylaniline, and phenol. p--N CHaS-C

I/

\N-N

I

R IA, R = H IB, R = CsH6 IC, R p-NO*-CeHa ID, R = P - N H ~ C ~ H ~

(2) F. L. Scott, D. G. O'Donovan, and J. Reilly, J . d p p l . Chem., 2, 368 (1952). (3) It is felt that controlled pyrolysis, or acidolysis a t elevated temperatures, may offer the best techniques for such dethiolations. Such experiments are under way. (4) Compare the unreactivity of the 5-methyl group as commented on by C. R. Jacobson, and E. D. Amstutz, J . Org. Chem., 18, 1183 (1953). (5) Under the nitration conditions used both in the present work and those of ref. 6, the tetrazole ring is undoubtedly protonated [cf. e.g., A. Hantzsch, Ber., 63, 1782 (1930)1, and the orientation encountered is that of the protonated species. In each case, the nitrated compound is largely the pderivative. Therefore, in both instances, it would seem that protonation occurs a t a position in the substituted tetrazole ring other than the 1-position. In our reaction, the 4-position appears a likely alternative [compare P. A. S. Smith, J. Am. Chem. SOC., 76, 436 (1954)j and in Garbrecht and Herbst's work, positions 5 and/or 4 appear possible as protonation sites. An analogous effect is encountered in the nitration of many other phenyl-substituted polynitrogen heterocyclic compounds and the general effect in many instances appears the same, uiz. protonation at a site other than that next to the phenyl ring with subseqrient o,p-nitration in the latter. Compare K. Schofield, Quart. Reus., 4, 382 (1950). (6) W. 1,. Garbrecht and R. M. Herbst, J . Org. Chem., 18, 1014 (1953).

VOL.

21

The reaction of I B with bromine in acetic acid resulted in cleavage of the tetrazole ring. The products i~olat~edmere 2,4,6-tribromoaiiiline and its acetyl derivative. EXPERIMENTAL'

I-Phenyl-6-methylmercaptotetrazole (IB) was prepared by the method of Stolle and Strittmatt.er8pQand was obtained as cream-colored plates of m.p. 80-82' (reported8 m.p. 80"). Anal. Calc'd for CsH8N4S: N, 29.2; S, 16.6. Found: N, 29.2; S, 16.5. Attempted displacement of the alkylmercapto group. Only faint traces of desulfurized material were encountered when I B was refluxed with 1to 3 equivalents of aniline in ethanolic solution for from 3 to 8 hours. A similar stability was displayed towards ammonia, diethylamine, and various arylamines, IB being recovered from such abortive aminolyses in ca. 90% yields. Refluxing with diethylamine for 40 hours was analogously unsuccessful, while treatment of IB with this amine in a sealed tube a t 110' for a similar time resulted in extensive decomposition, the only product being a black tarry mass.s Kitration of IB. To 10 ml. of concentrated sulfuric acid containing 5.0 g. of I-phenyl-5-methylmercaptotetrazole was added 6 ml. of concentrated nitric acid, the temperature being maintained a t 0" throughout. T o the light yellow mixture was added a further 20 ml. of concentrated sulfuric acid, and the resulting liquor then was allowed to stand a t room temperature overnight. It was heated on a steam-bath ca. 4 hours until nitrous fumes ceased to evolve, and then was poured cautiously onto an excess of crushed ice. It was then either neutralized with sodium hydroxide (which gave greater yields) or extracted with ether (which afforded purer products). The former technique, in the present instance, yielded 5.7 g. (93% yield) of a grey-brown substance, of m.p. 138-144'. After 3 recrystallizations from ethanol, compound IC was obtained as small light-grey needles of m.p. 147-148'. Anal. Calc'd for CsH7N,S02: C, 40.5; H, 2.95; N, 29.5; SI 13.5. Found: C, 40.7; H, 3.1; X, 29.4; S, 13.4. IC could be obtained in the following forms: amorphous, plates, needles or granules, depending on the volume of solvent used in the crystallization and the temperature range, and velocity of cooling utilized. Orientation of the mononitroderivative (IC). To 50 ml. of concentrated hydrochloric acid were added 2.0 g. of IC. The mixture was refluxed for 3 hours with the addition of a further 20 ml. of acid after 90 minutes. A yellow color developed during the reaction and a strong odor of methyl mercaptan was noted. The solution was cooled to O', poured onto ice, and neutralized with 10% sodium hydroxide solution. A yellow amorphous solid separated out (m.p. 140147', weight 0.72 g., 62% yield) which after 4 recrystallizations from aqueous ethanol was obtained as yellow needles

(7) All melting points are uncorrected. All microanalyses are by Drs. Wieler and Strauss, Oxford, England. (8) R. Stolle and A. Strittmatter, J . prakt. Chem., 133, 60 (1932); oide R. Stolle and F. Henke-Stark, J . prakt. Chem., 124, 261 (1930). (9) The infrared spectra of this, and related substances with their concomitant (sic) anomalies (e.g. the apparent absence of azide absorption a t 2160-2120 em.-' in the thiocarbamyl azides we have examined), etc. will be reported on later. (10) I. Heilbron and H. M. Bunbury, Dictionary of Organic Compounds, 4th Edition, Eyre and Spottiswoode, London, 1953; (a) Vol. 3, p. 629; (b) Vol. 4, p. 544.

OCTOBER

1956

of m.p. 148'. A mixture m.p. determination with pure pnitroaniline (reportedloa m.p. 148'), indicated no depression; a mixture m.p. with I C showed the mixture to melt a t 103-113'. Thus I C was 5-methylmercapto-I-p-nitrophenyltetrazole. Reduction of IC. T o 1.0 g. of I C dissolved in 75 ml. of 90% ethanol was added 1 g. of aluminum foil (which had been amalgamated by immersion for 10 minutes in an aqueous, 6770, solution of mercuric chloride). The mixture was allowed to stand a t room temperature for three hours, during which time a slow evolution of gas was detectable. The resulting sludge then n-as filtered off and on cooling to 0" for 24 hours, the filtrate yielded 0.69 g. (81% yield) of light-brown colored needles of m.p. 156-159O. After 4 recrystallizations from absolute ethanol, the 5-methylmercapto-1-p-aminophenyltetrazole (ID) was obtained as cream needles of m.p. 160161'. Anal. Calc'd for C8H9NSS:C, 46.4; H, 4.3; N, 33.8; S, 15.5. Found: C, 46.9; H, 4.2; N, 33.2; S, 15.3. Diazotization and coupling of ID. To a suspension of 0.5 g. of I D in 30 ml. of water was added 0.45 ml. of hydrochloric acid. T o the well stirred mixture, whose temperature x a s maintained below 10' throughout, was added, dropwise, a solution of 0.5 g. of sodium nitrite in 5 ml. of water. After a further 10 minutes, during which most of the suspended I D Rent into solution, ca. 1 g. of urea was added to remove excess nitrite. The mixture was stirred, its temperature being still kept below lo", for a final 15 minutes, then it was filtered and the filtrate was added dropwise to a solution of 0.7 g. of 6-naphthol in 30 ml. of ca. 0.1 N sodium hydroxide solution. Immediate deposition of IE as a deep red solid began. After 2 hours, it was filtered and washed with warm R-ater, dried, and obtained as 0.77 g., 87.5% yield, of material of m.p 230-232'. This, after recrystallization from chloroform, separated as fine orange-red needles of m.p. 233.5-234.0 '. Anal. Calc d for Ci8Hi4N60S:c, 59.7; H, 3.9; N, 23.2; S, 8.8. Found: C, 59.9; H, 4.0; N, 22.8; S, 8.8. The coupling product with N,N-dimethylauiline was formed analogously, except that the amine substrate was dissolved in ethanolic hydrochloric acid solution prior to coupling. The product (IF) was formed in 88% yield and was crystallized as fine red needles, of m.p. 190-191', from chloroform. Anal. Calc'd for C16HI,N7S: C, 56.6; H, 5.0; N, 28.9; S, 9.4. Found: C, 56.4; H, 4.8; N, 28.8; S, 9.5. With phenol, an analogous compound was formed in similar yield, m.p. 208-209". Anal. Calc'd for C14HleNsOS: C, 53.8; H, 3.8; N, 26.9; 8, 10.3. Found: C, 53.2; H, 3.7; N, 27.0; S, 10.3. Reaction of the above 5-methylmercapto-1-p-diazoniumphenyltetrazole salt with benzal guanyl- or phenylhydrazones, or with benzal diaminoguanidine or anisal guanyl hydrazone resulted in failure to isolate a coupling product under a variety of conditions. Bromination of IB. To a solution of 3 g. of I B in 25 ml. of glacial acetic acid was added 2 ml. of bromine, dissolved in 25 ml. of the game solvent. The addition was effected dropwise and the mixture was continuously agitated. After 60 minutes stirring at room temperature, the mixture was heated, a t IOO", for 30 minutes, during which time the odor of methyl mercaptan was noted. On cooling and concentrating in a stream of air, 0.43 g. (9% yield) of an amorphous solid of m.p. 118-120", was obtained. This on recrystallization from aqueous ethanol was obtained as tiny platelets, m.p. 120-122", which did not depress the m.p. of authentic 2,4,6-tribromoaniline (reported"J(b) m.p. 122'). Anal. Calc'd for CsH,Br3N: C, 21.8; H, 1.2; N, 4.2; Br, 72.7. Found: C, 22.6; H, 1.2; N, 4.4: Br, 72.5. On working up the filtrate a further portion (0.22 g., 4% yield) of this material was obtained, together with 0.51 g., (9yoyield) of 2,4,6tribromoacetanilide which was obtained as white needles m.p. 232' (reportedlO(b) m.p. 232') which did not depress the m.p. of an authentic sample.

1193

NOTE8

Anal. Calc'd for C8H6BrJNO:C, 25.8; H, 1.6; N, 3.8; Br, 64.5. Found: C, 26.3; H, 1.3; N, 4.0; Br, 64.7. Comparable results were obtained when the bromination was effected in refluxing anhydrous chloroform solution with, or without, magnesium oxide as base. CHEMISTRY DEPARTMENT UNIVERSITY COLLEGE, CORK,IRELAND. CHEMISTRY DEPARTMENT, OF CALIFORXIA, UNIVERSITY Los AKGELES24, CALIFORXIA

The Ultraviolet Absorption Spectra of the Pyridine Analogs of Chalcone LESTERE. COLEMAN, JR.~

Receioed June 6, 1966

I n connection with work on the synthesis of new monomers for free radical polymerization studies, a series of nine pyridine analogs of chalcone were prepared by the base-catalyzed condensation of the appropriate aldehyde and acetophenone or acetylpyridine. The ultraviolet spectra of these transa,p-unsaturated ketones are of interest since they provide an opportunity for comparison with chalcone and its 2-fury1 and 2-thienyl analogs. Szmant and co-workers3 correlated the absorption spectra of substituted chalcones with the electronic nature of the substituents by considering the whole molecule of the chalcone as one conjugated system. The principal absorption band is assumed t o originate from electronic oscillations represented by I. (+)

0

~ -c H - c H = ~

Ot-) -

IA

3 (-J+,cH=cH-c

0

if)

$1

-

IB

I

Samant found from experimental data that in +unsaturated ketones represented by 11, electron-attracting groups on ring A cause large hypsochromic shifts, but when present on ring B they tend to give bathochromic effects.

I1 (1) Present Development Ohio. (2) Marvel, (1955). (3) Szmant (1952).

address: Materials Laboratory, W ~ g h tAir Center, Wright-Patterson Air Force Base, Coleman, and Scott, J . Org. Chem., 20, 1785 and Basso, J . Am. Chem. Soc., 74, 4397

1194

NOTES

In other papers, Szmant4Jstudied the absorption spectra of the 2-thienyl and 2-fury1 analogs of chalcone. Results of these investigations indicated that replacement of the phenyl group by the 2-fury1 or 2-thienyl group gave a consistent bathochromic effect. This effect was greater when replacement occurred in position A. The spectra of the unsubstituted pyridine analogs of' chalcone exhibit a distinct maximum between 280 mp and 318 mp with an E value of approximately 20,000. This band is analogous to the 312 nip band of trans-chalcone and the observed shifts of this band in the pyridine analogs when compared with chalcone can be explained in terms of the greater electron-withdrawing effect of the pyridine ring as compared with benzene. In Table I, it can be seen that replacement of phenyl by a 2-, 3-, or 4-pyridyl group in position A causes a hypsochromic shift of 9-32 mp while replacement of the phenyl group by pyridine in position B causes a small bathochromic effect. All of the pyridine analogs reported are trans and have a planar configuration. However, 2-pyridalacetophenone (111) which should be similar t o 4pyridalacetophenone is an exception. The behavior of I11 is probably caused by some unusual electronic effect due to the position of the nitrogen in the ring. One possible explanation would be the resonance stabilization of the IA structure by formation of a three-membered ring containing

I11

4

IV TABLE I ULTILAVIOLET ABSORPTION SPECTRA or

VOL.

nitrogen (IV) similar to that proposed by Cram6 for the intermediate in the Neber rearrangement. EXPERIMENTAL

All of the pyridine analogs of chalcone have been recently reported and characterized.2 The absorption spectra were determined in 95% ethanol using a Cary recording spectrophotometer and the values for the two 4-pyridyl derivatives represent corrections of reference 2. The spectral characteristics discussed here are summarized in Table I.

Acknowledgment. The author is indebted to Dr. C. S. Marvel and Dr. Tu'. A. Nelson for their suggestions. NOYESCHEMICAL LABORATORY UNIVERSITY OF ILLINOIS URBANA, ILLINOIS (6) Cram and Hatch, J . Am. Chem. So:., 75, 33 (1953).

The Preparation of 4-p-Tolylpyridine CLACDE J. SCHMIDLE, JOHN E. LOCKE,AYD RICHARD C. MANSFIELD

Received June 6 , 1956

The preparation of 4phenylpyridine from the starting materials a-methylstyrene, formaldehyde, and ammonium chloride or methylamine hydrochloride has been reported in a previous communication.] The purpose of this paper is to report the synthesis of 4-p-tolylpyridine (111)using the same sequence of reactions, but starting with p , a-dimethylstyrene. During the preparation of 2-p-tolylpyridine picrate, Meek, Memow, and Cristol? also obtained two other picrates, presumably those of 3- and 4-ptolylpyridines, but 4-p-tolylpyridine itself has not been reported.

A--CH=CH-CO-B A 2-Pyridyl 3-Pyridyl 4-Pyrid yl Phenyl Phenyl Phenyl 2-Pyridyl Phenyl 2-Thienyl 2-Fury1 Phenyl Phenyl

B Phenyl Phenyl Phenyl 2-Pyrid yl 2-Methyl-5-pyridyl 4-Pyridyl 2-Pyrid yl Phenyl Phenyl Phenyl 2-Thienyl 2-Fur,vl

A,

303 298 280 318 315 317 304 312" 345" 344" 320" 324"

E

x

10-1

2.10 2.47 2.83 1.86 2.24 2.00 1.60 2.67 1.92 2.68 1.93 1.07

Reported in ref. 5 ,

(4)Szmant and Basso, J . Am. Chem. Soc., 73,4521 (1951). (5) Szmant and Planinsek, J. Am. Chem. SOC.,76, 1193 (1954).

21

H2C

M\

/N

-

I;R = M e 1I:R-H

R

15: 6 111

The reaction of p,a-dimethylstyrene, formaldehyde, and methylamine hydrochloride to give 3,6dimethyl-6-p-tolyltetrahydro-1,3-oxazine and 1methyl-4-p-toly1-4-piperidino1, and the rearrangement of this oxazine to l-methyl-4-p-tolyl-l,2,3,6(1) Schmidle and Mansfield, J . Anz. Chem. SOC., 78, 1702 (1956). (2) Meek, Merrow, and Cristol, J . Am. C h e m Soc., 74,

2667 (1952).

OCTOBER

1956

1195

NOTES

tetrahydropyridine (I) have already been described.3,1 4-p-Tolyl-l,2,3,6-tetrahydropyridine(11) was prepared hy reacting p , a-dimethylstyrene, formaldehyde, and ammonium chloride and then treating the reaction mixture with excess hydrochloric acid to effect rearrangement and dehydration. 4-p-Tolylpyridine (111) mas obtained by dehydrogenation of either I or I1 with palladium and nitrobenzene. EXPERIMENTAL'

I-Methyl-4-p-tolyl-l,2,S,B-telrahydropyridine (I).This compound, b.p. 110-115° (0.9 mm.), m.p. 76-78', was prepared in 61 70 yield from methylamine hydrochloride, formaldehyde, and p,a-dimethylstyrene using the procedure (Method C) described previously4 for t,he preparation of l-methyl-4phenyl-I, 2,3.6-tetrahydropyridine. Anal. Calc'd for Cl3H1iN: C, 83.37; H, 9.15; S , 7.48. Found: C, 82.95; H, 9.08; N, i.36. ~-p-Tolyl-l,2,5,8-tetrahydropyridine (11). This compound, b.p. 110-125" (0.75 mm.), m.p. 55-5i0, was prepared in 3370 yield from ammonium chloride, formaldehyde, and p p d i met'hylstyrene using the direct preparation procedure described previously' for the preparation of 4-pheny1-1,2,3,6tetrahydrop!.ridine. Snal. Calc'd for CIPHISN:C, 83.19; H, 8.73; S , 8.09. Found: C, 83.06; H, 8.71; N,8.30. The hydrochloiide melted at 193-195' after recrystallization from a 1 0 : l mixture of acetone and isopropyl alcohol. Anal. Calc'd for C12HI8ClN:C, 68.72; H, 7.69; S , 6.68; C1, 16.9. Found: C, 68.65; H, 7.54; S , 6.77; C1, 16.8. 4-p-Tolylpyridine (111). ;I. From l-methyI-4-p-folyl1,2,5,6-tetrahydrop?/ridine (I). This compound (111) was prepared from 76 g. (0.41 mole) of 1-methpl-4-p-tolyl1,2,3,6-tetrahydropyridine(I), 5 g. of 5% palladium on alumina catalyst, and 150 g. (1.22 moles) of nitrobenzene a t 150-60' during 2Ij2 hours using the procedure described previously1 for the preparation of 4-phenylpyridine from 1methyl-4-phenyl-l,2,3,6-tet,rahydropyridine. There was ohtained 3 i g. (54%) of 4-p-tolylpyridine (III), b.p. 122-127" (0.7 mm.). This solidified and me1tc:d at 90-91' after recrystallization from heptane. Anal. Calc'd for Cl,Hl,?;: C: 85.17; H, 6.55; N , 8.28. Found: C, 84.89; H, 6.45; 3, 8.28. The picrafe melted at 199-201 ' after recrystallization from ethanol containing a small amoiint of acetone. Anal. Calc'd for ClyH14X40i: C, 54.2i; H, 3.54; N, 14.07. Found: C, 54.49; H, 3.87; S, 13.80. B. From &-p-tolyl-1,2,5,6-t~trahydropyridine (11). .4 mixture of 5 g. of 5% palladium on alumina catalyst, 190 g. (1.54 moles) of nitrobenzene, and 46 g. (0.27 mole) of 4-ptolyl-1,2,3,6-tetrahydropyridine (11)was stirred in a nit,rogen atmosphere at 130-160" during two honrs while wat,er was removed b y a wat.er separator. The mixt.ure was cooled, filtered, and distilled to give 30 g. (87%) of 4-p-tolylpyridine (111), b.p. 90-110" (0.1 mm.). This solidified and melted at, 90-91 after recrystallization from heptane. The m.p. of a mixture with I11 prepared from I was 90-91". Anal. Calc'd for C12H11?V*:C, 85.17: H, 6.35; S , 8.28. Found: C, 85.14; H, 6.53; E,8.43. The picrate melted a t 199-201 after recrystallizat,ion from ethanol containing a small amount of acetone. The O

O

m.p. of a mixture with the picrate of I11 prepared from I was 199-201'. $nul. Calc'd for C18Hl4N407: C, 54.27; H, 3.54; S, 14.07. Found: C, 54.52; H, 3.78; r\i, 14.15.

Acknowledgment. We wish t o thank Mr. C. TY. Nash and his staff for analytical data reported. RESEARCH LABORATORIES ROHMA m HAASCOMPAXY I'HILADELPHIA 37, I'ENSSYLVASI.4

The Aminomethylation of p-Isopropyl-amethylstyrene CLAUDE J. SCHMIDLE, JOHN E. LOCKE,A N D RICH.4RD c. ~ ~ A N S F I E L D ReceiDed June bi, 19,58

It is the purpose of this paper to report the aminomethylation of p-isopropyl-a-methylstyrene. The reaction of dimethylamine, formaldehyde, and p-isopropyl-a-methylstyrene gave S,S-dimethyl3-p-isopropylphenyl-3-butenylamine(I). The reaction of p-isopropyl-a-methylstyrene, formaldehyde, and methylamine hydrochloride gave 3,6-dimethyl-6-p-isopropylphenyltetrahydro1,3-oxazine (11)and 1-methyl-4-p-isopropylphenyl4-piperidinol (111). Rearrangement and dehydration of the crude reaction mixture in the presence of excess sulfuric acid gave 1-methyl-4-p-isopropylphenyl- 1,2,3,6-tetrahydrop yridine (IT'). This substance was dehydrogenated and demethylated, using nitrobenzene and palladium on alumina, to 4-p-isopropylpheiiylpyridiiie (VI). Treatment of the reaction product of p-isopropyla-methylstyrene, formaldehyde, and ammonium chloride with excess hydrochloric acid gave 4-p-

ooH I

CH3

I11 (3) Schmidle and Mansfield. J . Am. Chem. SOC..77. 5698 (195k). (4) Schmidle and Mansfield, J . Am. Chem. Soc., 78, 425 (1956). ( 5 ) All melting points are uncorrected. I

,

VI

r*:r \

R IV; R=GH3 V; R = H

1196

NOTES

isopropylphenyl-l,2,3,6-tetrahydropyridine (V) . This compound also was dehydrogenated to 4-pisopropylphenylpyridine (VI). This is an extension of our work on the aminomethylation of olefins in which an analogous series of reactions was carried out using a-methylstyrene and p-tu-dimethylstyrene. l4

VOL.

21

21%220' after recrystallization from acetone containing about 370 of ethanol. Anal. Calc'd for Cl~HZ2C1?;:C, 71.54; H, 8.81; N, 5.56; C1, 14.1. Found: C, 71.02; HI 8.85; N, 5.47; C1, 14.0. 1-Meth yl-4-p-isopropylphenyl-l,.2,8,6-tetrahydropyridine (IV). A mixture of 130 g. (1.93 moles) of methylamine hydrochloride, 370 g. (4.56 moles) of 37$4 aqueous formaldehyde, and 300 g. (1.88 moles) of p-isopropyl-a-methylstyrene was stirred a t 95-100" for 1 hour and then was cooled. There was slowly added 150 g. (1.47 moles) of concentrated sulfuric acid and the mixture was stirred a t 95EXPERIMENTAL^ 100" for 4 hours, cooled, extracted with heptane, and made iV,S-Dimethyl-S-p-isopropylphenyl-S-butenylamine (I). A basic with excess 50% sodium hydroxide. The amine was taken up in toluene, dried, and distilled to give 159 g. (39%) 15-g. (0.125 mole) portion of S,S,X',N'-tetramethyldiaminomethane was added slowly with stirring and cooling of l-methyl-4-p-isopropylphenyl-1,2,3,6tetrahydropyridine to a mixture of 4 g. (0.125 mole) of 9570 paraformaldehyde (IT), b.p. 120-130' (0.5 mm.). This compound solidified and and 150 g. of glacial acetic acid. Then 40 g. (0.25 mole) of melted a t 64-66' after recrystallization from heptane. Anal. Calc'd for C1&H21N:C, 83.66; HI 9.83; N, 6.51. p-isopropyl-a-methylstyrene was added and the mixture was refluxed for 15 hours. After cooling, the mixture was poured Found: C, 83.47; HI 10.02; N, 6.45. The hydrochloride sintered about 200' and melted at into 500 ml. of water and p a s extracted with toluene. The aqueous layer was made basic with excess sodium hydroxide 21&220' after recrystallization from acetone containing a solution. The amine was taken up in toluene, dried, and small amount of ethanol. Anal. Calc'd for C,H&lN: C, 71.54; H, 8.81; N, 5.56; distilled to give 32.3 g. (60%) of N,N-dimethyl-3-p-isopropylphenyl-3-butenylamine (I), b.p. 100-110' (1.2 mm.), C1, 14.1. Found: C, 70.99; HI 9.01; N, 5.37; C1, 13.9. 4-p-Isopropylphenyl-1,6,9,6-tetrahydropyndine(V). A n y 1.5162. Anal. Calc'd for C I ~ H ~ J C, : 82.89; HI 10.66; N, 6.45. stirred mixture of 108 g. (2.02 moles) of ammonium chloride and 334 g. (4.12 moles) of 37% aqueous formaldehyde was Found: C, 82.86; H, 10.69; N, 6.69. S,6-Dimeth yl-6-p-isopropylphenyltetrah ydro-1 ,S-oxazine warmed to 60'. There was added 160 g. (1.00 mole) of p (11) and l-methyl-4-p-zsopropylphenyl-~-piperid~nol (111j. isopropyl-a-methylstyrene during 15 minutes while the A mixture of 70 g. (1.04 moles) of methylamine hydrochlo- temperature was controlled a t 60-65' by external cooling. ride, 200 g. (2.46 moles) of 3iY0 aqueous formaldehyde, and After the exotherm had ceased the mixture was stirred for 2 160 g. (1.00 mole) of p-isopropyl-0-methylstyrene was stirred hours while the temperature fell to 45'. There was added a t 95-100' for 1 hour, cooled, diluted with 1 1. of water, 225 g. (7.0 moles) of methanol and the mixture was stirred extracted with toluene, and made basic with excess 50% for one hour and then was allowed to stand overnight. The sodium hydroxide. The amine was taken up in toluene, methanol was removed by heating to 85' a t about 200 mm. dried, and distilled. The fraction boiling at 120-130' (0.7 pressure. After cooling, 300 g. (3.08 mole) of concentrated mm.) was redistilled to give 76 g. (33%) of 3,6-dimethyl-6- hydrochloric acid was added and the mixture was stirred for p-isopropylphenyltetrahydro-1,3-oxazine (11), b.p. 115- four hours at 96-98', cooled, poured into 700 ml. of water, made basic with excesR 50% sodium hydroxide, and ex117' (0.7 mm.). Anal. Calc'd for C11Hs3NO: C, 77.21; H, 9.93; N, 6.00. tracted with toluene. The toluene extract wa3 dried and distilled to give 96 g., b.p. 128-179" (0.75 mm.). This product Found: C, 77.77; H, 9.95; N, 6.02. The fraction boiling a t 130-140" (0.7 mm.) solidified, was was dissolved in excess 8% hydrochloric acid, extracted with combined with the residue, and was recrystallized from heptane, and made basic with excess 50% sodium hydroxide. The amine was taken up in toluene, dried, and distilled t o toluene twice to give 25 g. (11%) of l-methyl-b-p-isopropylgive 48 g. (24%) of 4-p-isopropylphenyl-l12,3,6-tetrahydrophenyl-4-piperidinol (111)) m.p. 143-144'. Anal. Calc'd for C16HZ3NO: C, 77.21; H, 9.93; N, 6.00. pyridine (V), b.p. 110-125' (0.3 mm.). The fraction boiling a t 120-125' (0.3 mm.), which consisted of 25 g., was used Found: C, 76.98; H, 9.91; N, 6.05. The foreruns, residues, and mother liquors from the puri- for analysis. Anal. Calc'd for C14H19N:C, 83.53; H, 9.51; N, 6.96. fication of I1 and 111 were combined, stripped free of toluene, and stirred a t 95-100" for 6 hours with a mixture of Found: C, 83.15; H, 9.52; N, 6.98. The hydTochZwide melted at 184-186' after one recrys10 g. of water and 100 g. of concentrated hydrochloric acid. The mixture was cooled, diluted with 500 ml. of water, ex- tallization from acetone containing a small amount of isotracted n-ith heptane, and made basic with excess 50% propyl alcohol. Anal. Calc'd for C14H&lN: C, 70.72; H, 8.48; N, 5.89; sodium hydroxide. The amine was taken up in toluene, dried, and distilled t o give 52 g. (247,) of 1-methyl-4-p-isopropyl- C1, 14.9. Found: C, 70.53; H, 8.63; N, 5.84; C1, 15.0. 4-p-Isopropylphenylpyridine (VI). A . From l-methyl-4phenyl-1,2,3,6-tetrahydropyridine(IV), b.p. 125-135' (0.8 mm.). This material solidified and melted at 63-65' after p-isopropylphenyl-l,I,S,6-tetrahydropyridine(IV). A mixture of 37 g. (0.17 mole) of 1-methyl-4-p-isopropylphenylrecrystallization from heptane. Anal. Calc'd for CljH21hT:C, 83.66; HI 9.83; PI', 6 3 1 . 1,2,3,6-tetrahydropyridine (IV), 150 g. (1.22 moles) of nitrobenzene, and 4 g. of 5y0 palladium on alumina catalyst Found: C, 82.89; H, 10.01; N, 6.49. The hydrochloride sintered about 200" and melted a t was stirred in an atmosphere of nitrogen at 140-150" for 21/zhours with a water-removal trap attached to the reaction flask. The mixture was cooled, diluted with excess dilute (1) Schmidle and Mansfield, J . Am. Chem. SOC.,77, 4636 hydrochloric acid and toluene, and filtered. The organic (1955). layer was washed with dilute hydrochloric acid and then ( 2 ) Schmidle and Mansfield, J . Am. Chem. Soc., 77, 5698 with water. The combined aqueous phases were extracted (1955). twice with a toluene-heptane mixture and then were made (3) Schmidle and Mansfield, J. Am. Chem. Soc., 78, 425 basic with excess 50% sodium hydroxide. The amine was (1956). taken up in toluene, dried, and distilled to give 19 g. (56%) (4) Schmidle and Mansfield, J . Am. Chem. Soc., 78, 1702 of 4-p-isopropylphenylpyridine (VI), b.p. 130-137" (0.7 (1956). mm.). This substance solidified and an analytical sample, (5) Schmidle, Locke, and Mansfield, J . Org. Chem., 21, recrystallized from heptane, melted at 70-72'. 1194 (1956). Anal. Calc'd for 4,HlsN: C, 85.23; H, 7.67; N, 7.10. (6) All melting points are uncorrected. Found: C, 85.21; H, 7.72; N, 7.10.

.----,-

OCTOBER

1956

NOTES

1197

Anal. Calc'd for C14HljN: C, 85.23; H, 7.67; N, 7.10. The picrate melted a t 184-186' after recrystallization Found: C,84.96;H, 7.82;S , 7.18. from ethanol containing a small amount of acetone. The picrate melted a t 184-186" after recrystallization Anal. Calc'd for CSHl8N401:C, 56.34; H, 4.25; N,13.14. from ethanol containing a small amount of acetone. The Found: C,56.38;H, 4.37;N, 13.02. B. From Q-p-isopropylphenyl-l,B,S,6-tetrahydropyridine melting point of a mixture with the picrate of VI prepared (V). A mixture of 4 g. of 5% palladium on alumina catalyst, from IV was 184-186'. C, 56.34;H, 4.25;N, 13.14. 61 g. (0.30 mole) of 4-p-isopropylphenyl-1,2,3,6-tetrahydro- Anal. Calc'd for C20H18N107: pyridine (V), and 250 g. (2.03moles) of nitrobenzene was Found: C,56.43;H, 4.22;N, 13.02. stirred a t 140-15O0 for two hours in an atmosphere of nitroAcknowledgment. We wish t o thank Alr. C. W. gen with a u-ater-removal trap attached to the reaction flask. The mixture was cooled, filtered, and distilled to give 45 g. Sash and his staff for analytical data reported. (75%) of 4-p-isopropylphenylpyridine (VI), b.p. 112-125" RESEARCH LABORATORIES (0.1mm.). This product solidified and an analytical sample, ROHMAXD HAASCOMPANY recrystallized from heptane, melted a t 70-72'. The melting PHILADELPHIA 37, PENNSYLVANIA point of a mixture with VI prepared from I V was 70-72'.