-lug. 20, 1956
VV'ILLGERODT REACTION WITH ilCYLklESITYLENES
4135
(b).--A batch of 3.5 g . of crude 4-ethoxy-2,3,B-triaminopq-ridine dihydrochloride was r d u x e d with 20 ml. of 98y0 formic acid for 19 hours. The volatile material was then evaporated and the residue was refluxed for 4 hours with 20 ml. of concentrated hydrochloric acid. Upon removal of the excess hydrochloric acid the residue was taken up with i: little water, then made weakly alkaline with sodium bicarbonate. The crystalline material which separated weighed 1.4 g. (57% from nitroso compound). -4 sample was charcoaled and recrystallized from water, 1n.p. 238240". .Inel. Calcd. for CsHloSIO: S , 31.1. Found: N, 31.6.
solid material mas filtered off. The yield was 0.3 g. (29.5yo). Recrystallized several times from water, once with the addition of charcoal, deazaguanine was obtained as a iiionohydrate, m.p. above 300". A ~ z a l . Calcd. for CsHGS40.H20: C, 42.86; €1, 4.80; S, 33.3. Found: C , 43.07, 43.11; H, 4.77, 4.813; N, 33.5, 33.2. (b).--,4 4.6-g. sample of 4-ethoxy-2,6-diamimo-3-nitrosopyridine mas reduced with hydrogen sulfide as above anti the reduction product condensed with formic acid. The crude cyclization product \vas refluxed with 50 i d . of 4870 hydrogen bromide for several hours. The crvstals were filtered off after cooling., dissolved in a little mater and made
\vas' neutralized with sodium bicarbonate and the separated
AMHERST,MASSACHUSETTS
[ ~ C ) X l . K I U U T I O SFKOLC
THE CIIEMICAL LADORATORY, U S I V E R S I r Y O F C A L I l W K Y I . l ]
The Willgerodt Reaction with Acylmesitylenes.
The Mechanism of the Reaction
B Y ~ - I L L I A xG. DAUBEN -4iiD JOHN B.
ROGAN'
RECEIVED JASUARY 3, 1956 The 1%-illgerodtreaction with acylmesitylenes has been investigated. I n all cases, keto-thioamides are formed. Such results suggest that carbonyl addition and reduction are not necessary steps for this reaction and a modification of previously postulated mechanisms is presented.
The mechanism of the ITillgerodt reaction has attracted considerable attention during the past decade, and to date there still is doubt as to esactly what processes are involved. To a large degree, this has been due to the failure to isolate intermediates directly2or to identify them. The technique of treating theoretically possible intermediates under conditions of the Tl'illgerodt reaction in order to determine whether or not they are actual possible intermediates has revealed the fact that the reaction is not limited to ketones, as originally supposed, but is applicable to olefins,?-5 acetylenes,"; alcohols,3--j halides,6 amines6 and even alkyl-substituted aromatic compounds. Perhaps the most satisfactory mechanism yet suggested for the conversion of ketones into amides (or thioamides) under Willgerodt conditions is that of King and Xc?rlillan' which envisions reduction of the carbonyl group to a hydroxyl (or thiol) followed by elimination to an olefin. This latter functional group is regarded as proceeding along the carbon chain by a series of reversible additions and eliminations of hydrogen sulfide until the terminal carbon is reached, a t which point irreversible oxidasimilar mechanism was suggested tions occurs. A\
(1) Recipient of t h e U 1934-1'35,j.
S Rubber Co. Irellomship in Chemistry,
(2) One notable exception ( A I . A. S a y l o r and 3 . B'. Auderson, 76, 5392 (1953)) is the isolation of what aiipears to he diisobutyl disulfide and trisulfide from the reaction of isobutylene, sulf u r and aqueous ammonia. These compounds were capable of being transformed into isobutyramide when resubjected t o Willgerodt cond i tions. (3) J. A . Kinp and F. H. .\lcMillao. THISJ O U R N A L , 68, 52; 1191G,. i t ) bl. Carmack r o d n F. DeTar. { b i d ,68, 2029 (1Q46) i L . rl. R . l'nttisno 2 o d 51 C a t m a r k . ) h i d , 6 8 , 2n3.3 (1011; T H I S JOURKAL,
r;!
I '1
Ii4ll~ ,JIJII1;
I1
> l L 3 l l i l , ~ l ~I/,>,/68.
I
by Carmack and DeTar4 except an acetylenic intermediate was postulated. I n any event: both mechanisms require initial attack a t the carbonyl group. I n an attempt to determine whether or not such a step is in reality a necessary requirement, acetylmesitylene was nllowed to react under the conditions of the m'illgerodt reaction. The decreased reactivity toward carbonyl addition of such a hindered ketonc is well known.8 From the reaction of acetylniesitylene, sulfur a i i t l inorpholine there was isolated an orange-yellow crystalline solid (I) in 347; yield which contained both nitrogen and sulfur. .llthough no recognizable product could be obtained from acid hydrolysis, prolonged saponification afforded a mixture of an acid (11) and a neutral nitrogen-containing material (111). These latter two compounds were shown to be inesitylglyoxylic acid and its morpholide, respectively, by comparison with authentic samples. n'hen the original reaction product I was allowed to react with Ratiey nickel, both desulfurization and reduction of the carbonyl group occurred and an atnino alcohol 11: was obtained. The structure of this material was establishecl as cymesityl-/?-(N-niorpho1ino)-ethanolby coiiiparison with a sample prepared by the lithium aluniiriuiii hydride reduction of w-(S-morpho1ino)-acetylmesitylene, which in turn was prepared by allowing w bromoacetylmesitylene to react with morpholinc. On the basis of the foregoing degradations, the acetylmesitylene n'illgerodt product can be assigned the structure I. 'rhe rompound I i l i a ? 1)r i - q a r d d pit her A\ :i tiur ( 8 1 R r. K e d - w h i h i r l , 6 6 , 12107 f l Q t i J ,1, 11 5 t r x - R r t r m a t i , r h r ' l 76,7811',151!. I J I' I ' , I d 1 , ' I 1, G , P V I , , / , ! , I ,7 7 , 12'14 ( I
WILLIAMG. DAUBEN AND JOHN B. ROGAN
4136
CH3
Vol. 78
CH3 I1
mal Willgerodt product, except that the carbonyl group has been retained, or as a product resulting from the direct oxidation of an active methyl group by sulfur and thus bear no relationship to the Willgerodt reaction. To distinguish between these possibilities, propionylmesitylene (V) was allowed to react with sulfur and morpholine under the same conditions as acetylmesitylene. The product of the reaction was a pale yellow solid containing both nitrogen and sulfur and gave an intense violet fer-
ric chloride enol test and which was soluble in dilute aqueous base. On treatment with acid, the material was both hydrolyzed and decarboxylated to yield acetylmesitylene which, in turn, was identified as the dinitro derivative. On the basis of this degradation sequence and on the analytical data, the product of the Willgerodt reaction with propionylmesitylene was assigned structure VI, N- (amesitoylthioacety1)-morpholine. The compound VI appears to exist largely in the enolic form VIa and/or VIb. The ultraviolet spectrum which is
I11
this is to be compared with the spectrum of N - ( a phenylthioacetyl) -morpholine, also shown in Fig. 1, which shows a sharp band a t 279 mp ( E 14,100) and a shoulder a t 365 inp ( E 230). For further comparison, the spectrum of N-(mesitoylthioforniy1)morpholine is shown in Fig. 2. This latter compound displays three maxima which are a t 258 mp ( E 8000)) 314 mp ( E 5900) and 407 mp ( E 540). The bathochromic shift of 11 mp found with VI can be accounted for on the basis of either structure VIa or VIb. The infrared spectrum of VI, however, gives evidence that perhaps VIa is the preferred form. There are weak absorptions in the 4 p region which are characteristic of S-H stretchingg while in the 3 p regilon there are no bands characteristic of 0-H stretching. A strong band appears a t 6.29 p and could be due to chelated unsaturated ketone structureg as is present in VIa. I n the 6.7-6.9 p . region thiere are two bands which are characteristic of a thioamide groupingg and which are found in the spectra of all of the thioamides in this study. If this grouping was conjugated with a double bond, as in VIb, these bands would be expected to be shifted to longer wave length. Although remote, the possibility exists that the formation of VI could have been due to enolization of propionylmesitylene followed by oxidation a t the a- or allylic carbon. Thus, n-butyrylmesitylene (VII) also was subjected to the Willgerodt reaction and the expected N-(P - mesitoylthiopropiony1)morpholine (VIII) was isolated.
~ ) C O C H Z C H Z C H--f ~ -77
CH3/v\CH3
S
VI11
VIb
The ultraviolet spectrum of VI11 is shown in Fig. 2, and i t is seen that the 279 mp (E 13,500) band, characteristic of the thiomorpholide grouping, is present. Therefore, i t appears that the products resulting from these three acylmesitylenes are, indeed, due to a Willgerodt type reaction. The fact that acetyl-, propionyl- and n-butyrylmesitylene upon reaction with sulfur and morpholine under Willgeroclt conditions yield a thioamide which retains the original carbonyl group instead of the usual des-keto thioamide can only be interpreted to mean that attack a t the carbonyl group is not a necessary step in the reaction. Instead, the carbonyl group would appear to serve only to activate the a-carbon, and it is on this carbon atom the reaction begins. Other evidence consistent with this view is available from the results obtained by Por-
shown in Fig. 1 consists of a single broad intense band with a maximum a t 290 mp (E 17,000) and
(9) L. J. Bellamy, “The Infrared Spectra of Complex Moleniles.” John Wiley and Sons, Inc.,New York, N,V.,1454.
SH
xnu0
Aug. 20, 1956
WILLGERODT REACTION WITH ACYLMESITYLENES
4137
5 0
4 0
4 0 \
\ W
w‘
a
0 3 0
3 0
‘3
0
-1
2 0
2.0
\
\ I O
200
3 00
400
500
mP\ Fig.
\I
I O
1.-X-( a-Plienylthioacety1)-morpholine, -;. mesitoylthioacetyl j-morpholine, ------.
N-( C Y -
200
3 00
“P.
500
400
Fig. 2.-X-(Mesitoylthioformyl)-morpholiiie,
-;
mesitoylthiopropiony1)-morpholine,-----.
K-(p
ter,I0 who studied the reaction of alkylpyridines rect oxidation of the methyl group followed by rewith sulfur and amines. It is known that an 01duction of the carbonyl group. It is to be realized, or y-alkylpyridine behaves similarly to an alkyl ke- however, that as the number of different types of tone in that the pyridine nucleus activates the adjacent methylene group. Reaction of a- and yalkylpyridines with an amine and sulfur yields Wthioamides or compounds derived from them, 0 whereas P-alkylpyridines are recovered unchanged. ~i Whether this initial attack on the methylene RCCHCHR’ I alpha to the carbonyl is common to the LVillgerodt P ~ (or R ~SH) reaction with both hindered and unhindered ketones is not known, but a t the present time there appears compounds which undergo reaction with sulfur and to be no need for separate mechanisms for the two an amine increases, the concept of a single mechclasses of compounds. At any time subsequent to anism for the Willgerodt reaction becomes meaningthe initial attack a t the a-carbon, the carbonyl less. group of the unhindered ketone could be reduced to Experimental a methylene group. That such a reduction is posWillgerodt Reaction with Acetylmesitylene .--A mixture sible is to be seen in the fact that the formation of a 2.00 g. (12.3 mmoles, b.p. 121.5-123.0°, T Z . ~ 1.5153) ~D of saturated hydrocarbon sometimes is a side reaction of acetylmesitylene, 0.40 g. (12.3 mmoles) of sulfur and 1.08 g. It is unlikely that (12.3 mmoles) of xnorpholirie was heated under reflux for 8 in the Willgerodt H2S alone can be the reducing agent, since it is hr. T h e dark reaction mixture was cooled, poured into known to react with ketones to form gem-dithiol~l~water and extracted with benzene. The benzene extract washed with water and the solvent removed on the and with 0-keto-esters to yield /3-thioketo-esters. l 4 was steam-bath under reduced pressure. The residue was disIn light of the present results, it is reasonable to solved in a mixture of 70-30 hexane-benzene and chropropose that the initial intermediate in the reaction matographed on 20 g. of Merck alumina. Elution with is not an olefin, as postulated by King and McMil- 70-30 hexane-benzene afforded 0.87 g. (43.5%) of unreacted Continued elution m ith the same solvent lan,? but rather an a,punsaturated ketone. Such acetylmesitylene. yielded, after recrystallization from hexane, 0.71 g. an intermediate could be formed by attack on the mixture (20.8%) of X-(mesitoylthioformy1)-morpholine( I ) as ora-carbon by sulfur followed by loss of H2S. This ange-yellow needles, m.p. 99.2-101.1°. An analytical sample was obtained by rechromatography olefinic linkage could then migrate down the chain, as suggested by King and M ~ M i I l a n by , ~ reversible on alumina and recrystallization from methanol-water, additions and eliminations of an amine or H2Sto the m.p. 101.0-101.5°. Anal. Calcd. for C I ~ H , ~ O ? S SC,: 64.95; H,6.90; N , olefinic bond. When a terminal double bond is 5.05; S, 11.,56. Found: C , 64.69; H, 7.08; h’, 5.12; s, formed, it is irreversibly oxidized to an acid deriva- 11.61. tive. If these steps be involved, the Willgerodt re\\’hen the above reaction xias performed except that 3 action on a methyl ketone could occur only by di- equivalents of sulfur were employed, the yield of thiomorH. D. Porter, THISJOURNAL, 76, 137 (1964). L. F. Fieser and G. W. Kilmer, i b i d . , 62, 1354 (1940). C . Willgerodt, Ber., 21, 534 (1888). T. L. Cairns, G. L. Evans, A. W. Larchar and B. C. McKusick, ’THIS JOURXAL, 74, 3982 (1952). (14) S. K . Mitra, .I. rndian Chcm. Soc.. 16, 31 (1938). (10) (11) (12) (13)
pholide was increased to 33.5y0, but the purification of the product was more difficult due to the presence of unreacted sulfur. (15) Analyses were performed by t h e Microanalytical Laboratory, Department of Chemistry and Chemical Engineering. University nf California. All melting points are corrected.
WILLIAMG. DAUBEN AND
4138
Alkaline Hydrolysis of N-(Mesitoylthioformy1)-morpholine (I).-A solution of 200 ml. of 95% ethanol, 10 ml. of 1 N NaOH solution and 2.50 g. (9.03 mmoles) of N-(mesitoylthioformyl)-morpholine was heated under reflux for 2 days, after which a n additional 10 ml. of 1 N XaOH solution was added a n d the refluxing continued for one d a y longer. T h e condenser was removed and the ethanol distilled. When the residue was acidified with dilute HCl, HzS was evolved. T h e acidic mixture was extracted with ether, the ethereal solution washed with water and then extracted with dilute NaOH. T h e alkaline aqueous layer was acidified, extracted with ether, the ethereal solution washed with water, dried over MgSOc and the ether removed on a steam-bath. T h e residue (1.31 9.) was a dark oil which solidified on standing. Recrystallization from hexane-benzene yielded I .05 g. (60.57,) of mesitylglyoxylic acid (11) as pale yellow, hard, dense crystals, m.p. 118.8-119.4" (lit.'6 116"), mixed m.p. with authentic mesitylglyoxylic acid (m .p. 116.7-1 17.9"), was 117.9-119.3'. T h e infrared spectra of the two samples in CHCll were identical. rlnal. Calcd. for CllH1203: neut. equiv., 192.2. Found: neut. equiv., 191.5. T h e ether solution containing the neutral fraction was washed with water and dried over MgSOl. T h e ether was removed on the steam-bath and the residual solid recrystallized from hexane, yield 508 mg. (21.5:5), 1n.p. 127.8128.5", mixed m.p. with authentic N-(mesitoylformy1)morpholine (111) (128.6-129.7') was 128.8-129.6'. The infrared spectra of the two samples in C& were identical. An analytical sample was obtained by chromatography over alumina followed b y recrystallization from hexane, m.p. 129.1-129.7'. T h e ultraviolet spectrum in 95% ethanol exhibited a maximum a t 276 mp (E 8180) with rising end absorption toward 220 rnp. Anal. Calcd. for CljH1903N: C, 68.94; H, 7.33; E, 5.36. Found: C, 68.95; H , 7.01; N, 5.14. Mesitylglyoxylic Acid.'6--With stirring, 2.00 g. (12.6 mmoles) of KMnOl dissolved in 40 ml. of water was added dropwise over a period of 30 minutes to a mixture of 1.00 g. (6.17 mmoles) of acetylmesitylene and 10 ml. of 2% NaOH solution. On continued stirring, the temperature of the reaction mixture rose to 40' and a t that time the reaction was cooled in a cold water-bath. After a total of 2.5 hours of stirring, the reaction mixture was filtered through SuperCel and the filtrate extracted with ether. Acidification of the alkaline filtrate gave an oil that was dissolved in ether, the ether layer washed with water, dried over MgSOa and the ether evaporated. Most of the oily residue solidified on standing in the cold. T h e entire residue was triturated with hot hexane-benzene and a small amount of white insoluble solid would not dissolve. T h e extract, on cooling, deposited 492 mg. (41.5yG) of mesitylglyoxylic acid as a yellow solid, m.p. 115.0-117.5°. Two recrystallizations from benzene-hexane gave 320 mg. of material, m.p. 116.7117.9' (lit.'6 116"). N-(Mesitoylformy1)-morpholine (III).-A mixture of 100 mg. (0.52 mmole) of mesitylglpoxylic acid, 1 drop of pyridine and 0.5 ml. (6.9 mmoles) of purified thionyl chloride was heated a t 50' for 45 minutes. T h e excess thionyl chloride was removed under reduced pressure and the cooled residue treated with 0.50 g. (5.8 mmoles) of morpholine. After the exothermic reaction had subsided, t h e reaction mixture was heated on the steam-bath for ten minutes, cooled and water added. T h e solid was filtered and recrystallized from hexane to yield 111 mg. (81.6%) of the acylated morpholine, m.p. 128.6-129.7'. Desulfurization of N-(Mesitoylthioformy1)-morpholine (I).-A mixture of 2.50 g. (9.02 mmoles) of N-mesitoylthioformyl)-morpholine, 250 ml. of ethanol and approximately 80 ml. of washed commercial Raney nickel was heated under reflux with vigorous stirring for 7 hr. T h e pale green solution was decanted and then filtered through Super-Cel. T h e residual Raney nickel was washed three times with hot ethanol and the washings filtered. T h e solvent from the combined ethanolic solutions was removed on the steambath and the residue dissolved in hexane. Attempts t3 extract this solution with dilute HC1 solution resulted in the formation of considerable white solid which was filtered, yield 0.68 K. ( 2 6 . 4 7 ) . Hwryatallization o f l h c .
. -.
I ^ _ _
I
1 f i 1 31
I
\'.)m
: ~ ~ l i v r p o n i hi,. ~ ~ l ,I r u i ~8 him ,
19, 3 i 9
[
ll,jlYJ),
JOHN
B. ROCAN
VOl. 78
solid from ethanol-benzene yielded two crops of material. T h e first crop amounted to 268 mg., 1n.p. 201-210' dec., and the second crop to 356 mg., m.p. 201.0-201.6° dec. Recrystallization of the second crop gave 306 mg. of material, m.p. 202.1-202.4' dec., mixed m . p . with authentic a-mesityl-&( S-morpho1ino)-ethanol hydrochloride ( m .p . 200.5-201.7° dec.), 201.0-201.8" dec. Anal. Calcd. for Cl6Hg2O2S.HC1~H~O: neut. equiv., 30-1. Found: neut. equiv., 306. T h e aqueous acidic layer from the filtration of the extraction was basified with ?;dOH solution and the curdy precipitate which formed was extracted with hexane. The hexane was removed on the steam-bath and the residual yellow oil (0.62 g., 87.Bc&) which solidified i x i standing was recrystallized from aqueous methanol to yield 430 mg. of material, m . p . 85.8-86.5'. Recrystallization from aqueous methanol gave 344 mg. of a-mesityl-&( N-morpho1ino)ethanol (IV), m.p. $5.7-86.5', mixed n1.p. with authentic material, 85.5-86.3 . T h e infrared spectra of the two samples in CSZwere identical. w-(N-Morpho1ino)-acetylmesitylene.-While cooling in a n ice-bath, 5.0 g . (31 mmoles) of bromine was added t o a stirred solution of 5.0 g. (30.9 mmoles) of acetylmesitylene in 20 ml. of glacial acetic acid over a period of 20 minutes. The reaction mixture (no bromine color) was poured into 50 1111. of cold water and extracted with hexane. T h e hexane extract was washed three times with water, dried ovcr magnesium sulfate and the solvent removed on the stcarnbath to yield 8.0 g. (of crude w-bromoacetylmesitylene). A solution of the crude bromo-ketone, 10.0 g. (115 mmoles) of morpholine and 50 ml. of anhydrous benzene was heated under reflux for 10 hr. T h e reaction mixture, containing solid material, was poured into water and the aqueous layer separated. An attempt to extract the organic layer with dilute acid resulted in the formation of a solid which was filtered. T h e crude hydrochloride was recrystallized from aqueous ethanol to yield 3.13 g. (33.5%) of w-( AT-morpho1ino)-acetylmesitylene h>-drochloride monohydrate, m.p. 252-254' (sintering 225') dec. T h e analytical sample was dried a t 115' and 0.5 mm. for 24 hr. Anal. Calcd. for C l s H ~ l O . N ~ H C 1 ~ H ~C,O :59.69; H . 8.01; N, 4.64; neut. equiv., 302. Found: C, 59.86; H , 8.16; N, 4.74; neut. equiv., 301. T h e aqueous layer from the two phase filtrate mas basified with NaOH a n d the oil extracted with ether. After drying over MgSO,, the ether was evaporated. T h e residual oil (1.40 9.) could not be induced to crystallize and so it was converted into the hydrochloride, m.p. 248-251" dec. a-Mesityl-p-( N-morpholino )-ethanol (IV) .--A mixture of 470 mg. (1.55 mmoles) of w-( S-morpholinoacetylmesitylene) hydrochloride monohydrate, dilute KaOH snlution and henzenc was shaken until all the solid had disappeared. T h c benzene layer was dried over h.lgS04 and the solvent evapnrated on the steam-bath. T h e residue was dissolved in 25 nil. of anhydrous ether and 0.25 g. (6.7 mmoles) of cruslictl LiXlH4 was added in small portions, with stirring. Stirring was continued for 2 hr., and the excess hydride was decomposed with a mixture of ethanol-benzene. \rater was added and the organic layer separated from the pasty solid. T h e solid was washed with a portion of ether and the ethereal solutions combined. .After washing the ethereal solution with water, the solvent was removed on the steam-bath. T h e residual oil crvstallized on standine and was recrrstallized from aqueous methanol, yield 242-1ng. (62.3(>),-m.p. 85.5-86.2 '. A n d . Calcd. for C15H2302N:C, 72.25; H , 9.30; 9, 5.62. Found: C, 72.31; H , 9.39; S , 5.69. T h e hydrochloride prepared from this material had a m.p. 200.5-201.7' (dec.). Willgerodt Reaction with Propionylmesitylene .--X mixture of 3.0 g. (17 mmoles, b.p. 132-134' (19 m m . ) , nzSD 1.5095) of propionylmesitylene,'7 1.09 g. (34 mmoles) of sulfur and 1.74 g. (20 mmoles) of morpholine was heated under reflux for 9 hr. T h e dark reaction mixture was diluted with water and extracted with ether. T h e ethereal layer was washed with saturated sodium chloride solution and the ether removed on the steam-bath. T h e residue was dissolved in 50-50 hexane-benzene and chromatographed on 50 g. of alumina. Elution with 200 ml. of .5O-T)0 hesaiir-I)enwnc yielded 1 .n7 p. ( 3 . 5 . 7 3 ) o f ~ ~ r ~ ~ ~ i i ~ t i ~ l t ~-
-
(171 A . l i J o c c s , Ucr ., 86, ilJ'l> 1 1 2 0 ) .
i i ~ ~ i t ~ ~ l ~ ~ i
XILDPERMAKGANATE OXIDATION OF XTISINE
Bug. 20, 1 9 x
-4 1 x 1
Continued elution with the same solvent mixture gave 1.48 diluted with ice-water. The white precipitate w a b filterctl of crude S-(a-mesitoylthioacet>-1)-morpholine and recrystallized from aqueous ethanol t o yield 72 mg. (1.1). Recrystallization, first from hexane and then from (457,) of material, m.p. 136.5138.5". X mixed 1ri.p. with aqueous methanol, yielded 1.21 g., m . p . 128.8-130.4". a n authentic sample of dinitvoacetylmesitglene ( i 1 i . p . 138.5-140.2') was 137.5-138.8'. T h e infrared spectra o f the twr) A n a l . Calcd. for Cl6H?,O2SS: C, 05.94; solid derivatives in CHC13 solution were identicd. 1.81; S, 11.01. Found: C, 66.06; H , 7.17; Willgerodt Reaction with n-Butyrylmesity1ene.--.\ mix10.79. ture of 2.0 g. (10.5 mmoles, b.p. 145-147' (20 niiii.), n2% The material from another experiment has a m . p . 121 .O 1.5019) of n-butyrylmesitylerie,'7 0.67 g. (21 mnioles) of 123.6' after recrystallization from hexane. T h e two difsulfur and 1.80 g. (20.7 mmoles) of morpholine was heater1 ferent melting forms are dimorphic crystals since their under reflux for 7 hr. The reaction mixture was diluted infrared spectra in CHCl., were identical. The coinpllund gave a n intense red color with aqueous inethanolic FcC13 F\-ith benzene, washed twice with water and then with and was soluble in 5'; aqueous sodiuni hydriixirle aiitl rc- saturated NaCl solution. The benzene was removed on the steam-bath and the residual oil dissolved in a mixture of covered upon acidification. 75-25 hexane-benzene and chromatographed on 25 g. of Acid Hydrolysis of N-(a-Mesitoylthioacety1)morpholine Elution with 50-50 hexane-benzene first yieldctl (VI).-.\ mixture of 185 mg. (0.64 mmole) of S-(a-mesitoyl- alumina. 1.13 g. of starting ketone and then 700 mg. of material thioacetyl)-morpholine, 5 ml. of 6 A' hydrochloric acid and 5 mhich solidified on standing. T h e crude S-(B-niesit(iyltlii[inil. of acetic acid was heated under reflux for 6 hr., during propioiiyl)-inorpIioline was recrystallized three times from lq-liich time hydrogen sulfide was evolved. T h e reaction n1.p. 99.7-100.4". mixture was cooled, diluted 17-ith water and extr;tctcti nit11 benzene-hexane, yield 285 mg. (8.857), ether. The extract was waslied with 5' sodium h>droxide Anal. Calcd. for C17H2JOPSS:C , 66.80; H , 7.59; S , 4.59; S,10.50. Found: C , 66.78; H , 7.32; S,4.87; S, (acidification of which yielded no material 1, saturated YaCl 10.11. wlution and dried over MgSO1. The ether Ira I I I I i t stearn-bath and the rwitluc was a n o i l . Til T h e low recover\- yield from tlic recrystallizat iciiis was \v:is nitrated ;it 0' with fuming FI S O , . ' 8 'l'lic. r c a c t i ~ ) i inivi tluc t o thc ertrrmi\ difiicultj i i i oht:iiiiing colorlc\\ r:itlier tiirc was swirled iii t h e ice-hath ~ ( I I - :30 wcriiids ; i i i ( I tlicii t h i i slightly tlisc~ili~retl cr!.ct:ili. . .. g . (29.87;)
(18) I
