Pyrolytic Formation of Arenes. III. Pyrolysis of Pyridine, Picolines and

Vernon R. Morris , Subhash C. Bhatia , Arthur W. Stelson , and John H. Hall , Jr. Energy & Fuels 1991 ... Thomas J. Houser , Michael E. McCarville , T...
0 downloads 0 Views 745KB Size
Dec. 5, 1962

PYROLYSIS OF PYRIDINE, PICOLINE AND METHYLPYRAZIKE

bon dioxide. The advantage of the sodium carbonate is solubility of sodium sulfate in the concd. sulfuric acid used to generate the carbon dioxide as contrasted with insolubility of the barium sulfate. After carbonation, the magnesium salt was hydrolyzed by 100 ml. of 6 N hydrochloric acid. After processing as for benzoic-a-14C acid,3 using 1.87 g. of ordinary Lthenoic acid as scavenger, there was obtained 4.55 g. of 2-thenoic-d4C acid. This was 75YGyield, somewhat lower than the obtained in practice runs with non-radioactive material. Reduction.-This 4.55 g. of acid was admixed with 1.24 g. of inactive acid: total, 5 . i 9 g. or 0.015 mole. I t was dissolved in 100 ml. of dry ether and added dropmise into a stirred slurry of 3.3 g . (2.5 equivalents) of LiAlH4in 100 ml. of dry ether. After 2 hours of stirring the product was hydrolyzed by cautious addition of saturated ammonium chloride solution. The sandy textured solid was filtered off, rinsed twice with ether, suspended in water and the suspension twice extracted with ether. The filtrate was extracted with four 100-ml. portions of ether. The combined and evaporated to 50 ml. ether extract was dried (i%a2S04) in a n air stream. 2-Thenyl-d4C Chloride.-To the above 50 ml. was added 6.4 g. of inactive 2-thenyl alcohol. This, plus the 5.1 g. of alcohol theoretically formed by reduction of the acid, makes 11.5 g. (0.1 mole). To i t was added 8.7 g. (0.11 mole) of pyridine. The flask, cooled by ice, was stirred and into it was added during one hour 12.5 g. (0.105 mole) of thionyl chloride $ 15 ml. of dry ether. The temperature was held below 15 . Then it was stirred for 4 hours a t 0". The ether was decanted and the residual white solid was thrice slurried with 50-1711. portions of ether. The solid was then dissolved in water, with evolution of SO*, and the solution was twice extracted with ether. The combined ether portions were washed with very dilute sodium hydroxide, dried (Fa2SO4),and concentrated to 200-250 ml. 2-(Methyl-a-14C)-thiophene.-The above ether solution was transferred with adequate rinsing into a flask and to it was added 50 g. of zinc dust and 120 ml. of glacial acetic acid. I t was stirred a t reflux for 12 hours, then the ether was decanted and the residue was repeatedly stirred with 50ml. portions of ether. The ether solution, after mashing several times with water to remove possible acetic acid, was dried (CaC12) and concentrated. To the concentrate was added 50 g. of inactive 2-methyltliiophene (from The Texas Co.) and the whole was fractionated through a 30-cm.

[COSTRIBUTION FROM

THE

4519

vacum-jacketed Vigreux column. The yield of product boiling over a 0.5' range was 54.5 g. Prior to assay, a small amount of this radioactive 2methylthiophene was diluted 11.4 times with ordinary 2methylthiophene. The activity was found to be 0.111, 0.113 ,uc./mmole, from which the starting sample was 0.112 X 11.4 or 1.28 pc./nimole. Pyrolysis.--A 50-ml. sample of this radioactive 2-methylthiophene was pyrolyzed through the Vycor reaction tube at 800" with a hot contact time of 10.5 sec. T h e non-gaseous product (40.5 g . ) collected in the ice trap as a black liquid which was fractionated through a Davis column: b.p., up 6.55 g.; 84-109', 1.27 g.; 109-112", 12.17 g. The to No, residue (.4), 54% by weight, was processed b y chromatography (below). The recovered 2-methylthiophene was 24%. T h e products isolated were benzene, 1 mole %; thiophene, 19 mole 5% ( b y difference); naphthalene, 0.1 mole 70; benzothiophene, 0.1 mole "70. Activity of the Benzene.-From 3 g. of the first fraction containing benzene and thiophene was isolated 0.17 g. of mdinitrobenzene, using the procedure outlined above. It was recrystallized 3 times from methanol; m.p. 90'. This pure material had a n activity of 2.32, 2.36 (ave., 2.34) pc./mmole. Separation of Solids.-Residue A was mixed with 50 ml. of petroleum hexane, poured onto a column ( 4 X 100 cm.) containing 800 g. of chromatographic grade alumina (Merck and Co.), and eluted with hexane. Naphthalene passed through first, then benzothiophene. T h e naphthalene was purified by two crystallizations from alcohol, then sublimation at 60" (750 mm.); yield 46 mg., m.p. and mixed m.p. sharply a t 80". I t s activity was determined t o be 1.31, 1.39 (ave., 1.35) pc./mmole. T o purify the crude benzothiophene sufficiently for radioactive assay i t was mixed with 2.5 ml. of water, 10 ml. of glacial acetic acid and 2.5 ml. of 30Yo hydrogen peroxide. After warming the mixture for 30 min. at looo, 35 ml. of water was added and the mixture was left at 5" for several hours. The precipitated benzothiophene I,I-dioxide was dried at 80°, then was heated for an hour at 75' (0.5 mm.) in a sublimation apparatus. No naphthalene appeared. The dioxide was thrice recrystallized from benzene-methylcyclohexane (1:2) to yield 52 mg. of pure product, m.p. and mixed m a p . 142". Its activity was 0.93, 0.87 (ave., 0.90) pc ./mmole.

CHEMICAL LABORATORY O F NOPTHWESTERN UNIVERSITY, EVANSTON, ILL.]

Pyrolytic Formation of Arenes. 111. Pyrolysis of Pyridine, Picolines and Methylpyrazine D. HURDA N D JUSTINE I. SIMON' RECEIVEDMARCH12, 1962

B Y CHARLES

Pt educt\ from the pyrolysis at 823-S5Oo of pure picoline-free pyridine included quinoline, benzonitrile, acetonitrile, acrylonitrile and benzene. T h e picolines were much less stable than pyridine, products being pyridine, isomeric picolines, quinoline, nitriles, benzene, some ethylene but no acetylene. Analytical methods were developed for these mixtures. Methylpyrazine is considerably less stable to heat than any of the picolines. More acetonitrile and acrylonitrile were formed than from the picolines, and benzene was not found.

Since benzene, naphthalene, 3-niethylthiophene and other pyrolytic products were obtained froni 2r n e t h y l t h i ~ p h e n e ,it ~ ~was ~ of interest to test carefully purified %picoline, 3-picoline, 4-picoline and pyridine itself under comparable thermal conditions. Earlier pyrolytic work with pyridine and its homologs is not large. R ~ t h the , ~ first to study (1) University Fellow, 1933-1954; Xational Science Foundation Fellow, 1954-1955. (2) C. D. Hurd, A. R . Macon, J. Simon and R . Levetan, J. Ant. Chem. Soc., 84, 4509 (1962). (3) C. D. Hurd, R . V. Levetan and A. R. Macon, rbid., 84, 4515 (1962) (4) C. F. Roth, Ber., 19, 360 (1886).

pyridine by passing i t through a red hot glass tube, noted hydrogen cyanide and bipyridines as products. Meyer and Hofmann-Meyer5 reported 2,2'bipyridine as the chief product, but they also found the 2,3'- and 2,4'-isomers. Ruhemann6 used temperatures of 600 to 900". At the lower temperatures (600-650') he reported t h a t production of hydrogen and bipyridine was the chief reaction, whereas with increasing temperatures nuclear scission into hydrogen cyanide be( 5 ) H . Meyer and A. Hoffmann-Meyer, J . prakl. Chcm., 102, 287 (192 1). (6) S. Ruhemann, Braunkohlc, 28, 749 (1929); Erddl u. Teer, 6 , 455 (1929). T h e two articles are identical.

4520

CIIARLES I). HURUA N D

JUSTINE

1. SIMON

Vul. 84

TABLE I PYROLYSIS OF PICOLIXES 2-Picoline Temp O C

.

-

700

-' 120

--

I00 --ID I

800 835 830

Contact Recovered time, subst., sec. %

-_

,

Products, mole %" e f

a

b

C

d

0.0

0.0

.o

.o

..

.0

-

.0

..

. I

.2

"

.5 .5

0.0 0.4 1.1 1.1

11 8 12 20 14 11 11

93 89 86 70 59 46 29

0.0 .0 .0 .13 .26 .39

18 20

85 66 70 23 33 21 19

0.0

-.a1

.I

1. 3 1.3

.4

0 .0

0.0

.o

.o

..

..

tr. 8.4 9 2 21 21 25 25

..

3 3 14 13 20 16 24 21

tr. 2.1 3.5 3.4 4.2 3.2 2.6

..

.. ..

.. ..

g

h

i

j

0.0 0.0 1.5 4.9 2.4 1.2 1.3

0.0

..

.0

.. ..

64 53 31 41 42 47 41

.o .3 .5

.,

. I

.33

0.0

.. .3 .8

3-Picoline i00

--750 ii

t a

800 SO0 825 850

12 20 12 18 13

.o .o .19 .10 .2,5 .48

-

.0 . I

.J 1.0 1.3

tr. 0.3 .2

.c l

.4

.. *. 1.0

..

1.3

.. 1.5

4-Picoline 0.0 .. .0 .. tr. .. 0 .6 . . .-l 0.6

.. ..

..

.. ..

3.9

.. ..

..

3.0

, .

..

..

..

..

---

-

.7

..

..

1.2

.D

9 94 0.0 0 .0 .. 3.7 0.0 0.0 13 83 .0 .o .. 8.0 .3 1.0 i ia 16 69 .o tr. 2 .0 .. 14 .4 so0 12 .31 0.6 .. 63 20 .I -3. _3 825 20 27 .-lo 1.2 20 1.4 .. .9 850 21 6 .4s .. 0.8 .-3 .. 7.4 0.5 .3 a a, benzene; b, acetonitrile; c, acrylonitrile; d, henzonitrile; e, pyridine; f , 2-picoline; g, 3-picoline; i, quinoline; j, residue in wt. %.

700 750

..

..

.. ..

.. .3

.. h,

61 56 47 32 50 48 34 16 30 58 41 41 26 4-picoline;

came important. At 900', he stated that practi- found i t to melt a t 30". This low melting point cally cotnplete cleavage of the ring occurred, forming explains why i t might have failed t o crystallize gaseous cleavage products. I n addition to HCK from the oil found by 32.2eyer and Hofmannhe reported this gas analysis for the temperature LIeyer. range 700-900": H2 84-8137~, CHI l2-l4Yc, Just as toluene breaks initially into the benzyl olefins 1-27;, with no more than a trace of acet>- radical, so also the picolines break initially'O a t lene. S O O O and 2 to 13 mm. pressure into the picolyl Krumholz' passed (i kg. of com~iiercialpyridine, radicals. Bond energies for 2-, 3- and 4-picoline b . p . 113.2-114.5', through a quartz tube a t S.50' of 'Tti.3, 'Tci.3 and '7'7.5 kcal.,.hiole, respectively, during three days. The products, with yields proved to be remarkably close to toluene's ( i i . 5 ) . based on unrecovered pyridine, were 2,2'-bipyri- Gaseous products were hydrogen and methane in dine lOyL, the 2,,3'-isomer 10%) the 8,4'-isomer 4% a 3 : 2 ratio from 2- or 3-picoline, and '7:3 from 4and quinoline 27c. Also isolated were other bi- picoline. T h a t pyridine was formed (2lY0 yield) pyridines, terpyridyls, indole and indolizine. Some from 2-picoline a t 800' was reported by ,kneniya evidence was indicated also for pyrrole, 2-picolino- and Suzaki. l 1 nitrile, benzonitrile and dihydroskatole. Since Pyridine was purified for the present work t o there was no assurance t h a t his pyridine was free remove traces of picoline. This material yielded of picolines which may have been responsible for quinoline and benzonitrile on pyrolysis a t 825the quinoline, i t became a matter of importance to S30', thus confirming Krumholz's report. lo Also study pure pyridine from this point of view. found were other nitriles and benzene. T h a t There is still less reported work on the picolines benzene was found is remarkable. I n an experiand most of this is on %picoline. Meyer and ment a t 823" and 9 seconds contact time, one-third Hofmann-lleyerj exposed the latter to "red heat" of the pyridine was recovered and from the reand believed t h a t G,G'-dimethyl-2,2'-bipyridine mainder these yields were formed : benzene 0.18 was in the oily product boiling between 210-295", mole yCc, acetonitrile 0.2, acrylonitrile 0.4, benzob u t this conclusion was challenged by Wibaut8 nitrile 1.1,quinoline 1.3, and a considerable higher who prepared the latter and reported t h a t i t had boiling residue. Yields of the nitriles were about different properties from the pyrolytic material. doubled a t 850' (5 sec.) and were much less a t SOO" AIeyer's product may have contained 2,2'-ethylene- (12 sec.). dipyridine (NC5H4CH2-)?,later synthesizedg and The several picolines were pyrolyzed a t temperatures between iOO-S30°. Table I summarizes the (7) P. Krumholz, S e l e ( l a chitit., data. Gaseous products and some carbonaceous (8) H. \Tillink, J r , and J. P. e r . lini'. chiin , 5 4 , 97.; ( 19:3:j [!I) 1'. C, Caniplwll and ( 1 ! l i I j.

T'. C.

Te;igiic, J. A W I .Cheiii. Sor , 7 6 , 1371

1101 J. S. Roberts and SI. S z w a r c , J . Chuvi. P h j s , 16, !I81 (1918). ( I 1) T. Xinrniyn :in11 ST. Su).:tki, Japan Patetit t i 7 8 ( l ! ) 47, 1 IL'fiO (I!l-A).

PYROLYSIS OF PYRIDINE, PICOLINE A N D METHYLPYRAZINE

Dec. 5 , 1962

deposit appeared in all of the runs. The gas, chiefly hydrogen and methane, was analyzed for unsaturated hydrocarbons in several of the runs. Acetylene was absent, but about 1 mole of ethylene, based on unrecovered picolines, appeared in the runs a t SOO" and about 2 mole yc a t 830". Hydrogen cyanide was formed in every run above r - - 0 113 , b u t no attempt was made to estimate it quantitatively. I n the experinient with pyridine a t S23" (9 sec.) 647, was recovered, and a t 850" (3 sec.) .5iycwas recovered. The much smaller recoveries of the picolines a t these temperatures, with even shorter periods of heating, demonstrate t h a t they are much less stable than pyridine. Decomposition rates of the three picolines vary but little aniong themselves between 70U to 7 i 3 " . This would be expected in view of Szwarc's finding that bond dissociation energies of the three picolyl radicals were so nearly identical. From SOU to S50" sotne differences were observed : 2-picoline was the most stable isomer and 4-picoline the least. It is above -0 ii a that benzene and nitriles first appeared, indicative of rupture of the ring. The slightly larger bond dissociation energy reported by Szwarc for 4-picoline ( i i . 5 kcal ) over 2-picoline ('75.3) is not incompatible with the greater instability of 4-picoline found in the present study. Szwarc was dealing with the formation of the picolyl radicals as the slower, ratedetermining step and suggested that the small differences might follow if dipole radicals were present in the resonance hybrid

-

The 2-picolyl dipole radical, with the least separation of charge, would be expected to be the most easily formed and the most stable, hence most resista n t to further decomposition. The appearance of some pyridine a t the relatively low temperature of TOO" is what one would expect if the primary, rate-determining reaction is formation of the picolyl radicals, followed by fast secondary reactions. C5H4NCH3+.C5H4NCH2. H., and CjH4NCHi H either CjH&CH>. Hr,or C5HJ C H I . That pyridine production should decrease above S30" would be expected since pyridine itself decomposes fairly rapidly a t this temperature Rearrangement from one picoline t o another was noticed a t 725 and i S O o , but there was no production of benzene or nitriles until higher temperatures were used The fact that pyridine formation occurs at a somewhat lower temperature than picoline isomerization suggests a higher activation energy for the latter process than the former. The low boiling products were of particular importance for several reasons Except for the small size of the fractions (below 90") they were relatively easy to isolate dnd characterize Even so, they had been completely neglected by- earlier irivestiq,itors Ben7ene h a d ne\-cr heen stispcctcd Ben-

+

+

-

+

+

452 1

zene and the nitriles obviously were formed as the result of ring breakdown. Benzene production from all three picolines increased with increasing temperatures up to SSO", with somewhat larger yields coining from 2-picoline. A smaller yield of benzene was observed from pyridine than from any picoline. The pyrolysis of 2-niethylthi~phene~ gave rise to much hydrogen sulfide, pointing to rupture of both carbon-sulfur bonds. Comparable rupture of all carbon-nitrogen bonds in picoline would have yielded ammonia, but no ammonia was noticed. T h a t one of the nuclear C-N bonds breaks on heating, however, is evident because of nitrile formation. Creation of the cyanide group evidently is a large driving force in this reaction. Since all carbon-nitrogen bonds were not broken this speaks also against the transient existence of : CH-CH=CH-CH=C-CH,. (followed by hydrogen shift t o : CH-CH=CH-CH=CHC H : and cyclization to benzene), comparable to the formation' of cyclopentadiene slia ,CH=CHCH=C)-CH?. from the 2-thienyl radical. X1though such a postulate would have explained benzene from picoline, it would not have explained benzene from pyridine. .llso, the data speak against any forniation of benzene sjia acetylene since no acetylene was formed. Nor could there have been any significant production of benzene via gaseous olefins. The highest quantity of such olefins ever found was 2.2 niole yo and in this run (8-picoline at 830") the amount of simple arenes isolated (benzene, benzonitrile, quinoline) totaled nearly 4 niole yc. The unidentified arenes in the large tarry residue would increase this to a much larger figure. IYlien one considers that 2 to 3 moles of propylene or ethylene would be consumed in forming each new benzene ring the whole process becomes quite unreasonable. No prior work has been reported on the pyrolysis of methylpyrazine (I), but it should be capable of pyrolytic change into the pyrazinylniethyl radical I1 if it follows the pattern set by toluene, methylthiophene and picoline. Conceivably, I1 might break 011 heating into the C3Ha fragment2 froin

I

11

I11

which benzene would be a predictable product. Offsetting this, however, is the fact that nitrile TABLE I1 ~IETHYLPYRAZISE Temp., "C. 8011 823 Contact time, sec. 10 15 Subst. taken, g. 10.5 7.6 Subst. recovd.," % 25 (5) Products,b mole (;. Benzene 0.0 0.0 .> 3 .icetonitrile 13 --~cr~~Iimitrile 2 10 High boiling residue, w t . y i 37 53 Contained mine pyraziiie, as judged by new peaks in the infrared spectruni. Gases (H., hydrocarbons, H C S I ~ i n t lw i n e carlmiinceniir m;i t ter d.;o were f ~ ~ r i i i e d .

452%

CHARLES D. HURDAND JUSTINE I. SIMON

Vol. s4

TABLE 111 INFRARED A ~ R S O R P T I O N BANDSFOR THE PICOI,ISES

2-

3-

2.09 2 . ,4(J

2.09 2.23 2.40

2.HC

2 , !)I)

..

..

3 .Mi 3.81

:: .Mi

:3 .(i.5 3.84 4.1;; 4.30 4.42 4.52

-. C)