Investigation on Partheniol, a Sesquiterpene Alcohol from Guayule Citation Fong, Conrad Tuck Onn (1946) Investigation on Partheniol, a Sesquiterpene Alcohol from Guayule. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/7g5v-4r45. https://resolver.caltech.edu/CaltechTHESIS:06082025-133401537 Abstract No abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Biology) Degree Grantor: California Institute of Technology Division: Biology Major Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Haagen-Smit, Arie Jan Thesis Committee: Unknown, Unknown Defense Date: 1 January 1946 Record Number: CaltechTHESIS:06082025-133401537 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:06082025-133401537 DOI: 10.7907/7g5v-4r45 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 17422 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 18 Jun 2025 17:55 Last Modified: 18 Jun 2025 18:07 Thesis Files PDF - Final Version See Usage Policy. 23MB Repository Staff Only: item control page

INVESTIGATION ON PARTBENIOL A SESQUITERPENE .ALCOHOL FROM WAYULE Thesis by Conrad T. o. Fong In Partial Fulfilment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1946 Acknowledgement I wish to express my appreciation to Professor A. J. Haagen-Smit for the guidance and valuable advice which he has given me in my work. I also wish to thank Doctors J. Bonner, .A.. c. Hildreth, and J. G. Kirchner for their cooperation. I am grateful to Dr. G. Oppenheimer and G. Swinehart for the micro-analyses carried out during the course of this investigation. Tabla of Contents Introduction 1 Experimental Section 12 Isolation of Pa.rtheniol 12 ~antite.ti ve and Qualitative Analyses of Pa.rtheniol 17 Absorption Spectrum of Pa.rtheniol 23 Preparation of Pa.rthenyl-3,5-dinitrobenzoate 24 Preparation of Patthenene 25 Partial Hydrogenation of Partheniol with Raney Nickel 26 Preparation of Azulenes from Partheniol 29 Preliminary Qualitative Ozonization of Pa.rtheniol and Tetrahydroparthenene 42 Quantitative Deterrr.ination of Methylene Group in Partheniol and in Tetrahydroparthenene by Ozonization 45 Deterrr~nation of the Position of the Hydroxyl Group in Pa.rtheniol 47 Determination of the Positions of the Double Bonds 66 Discussion 71 Summary 84 Bibliography Introduction Terpenes are the main characteristic constituents of many essential oils, the volatile components of plants. The name 11 terpen 11 is commonly considered to have been introduced by Kekule for hydrocarbons with the empirical formula, c10H16 • With the abolishing of the use of the broad term camphor to signify crystalline oxygen compounds with empirical formulae, c n16 o and c10H18 o, 10 the definition of terpene has been extended to cover this group of compounds. The name terpene as it stands today not only implies compounds which are composed of c5H8 units and oxygenated derivatives of these compounds, but also cover compounds which contain less hydrogen and those which are more saturated. There are, moreover, exceptions to this classification; for example, santene (c 9H14 ), is considered a terpene compound because of its resemblance and relation to terpenes. The natural-occurring terpenes are generally complex mixtures of numerous terpenes and their stereoisomers and are usually liquid compounds. For this reason, early investigators found it difficult to identify and classify the various terpenes which were isolated from plants. An outstanding contri- bution to this field of research was made by Wallach, who, by the preparation of crystalline derivatives, was able to identify several hllndred of these compounds. Stimulated by Wallach 1 s work, Semmler and others studied the struc- ture of terpenes and as a result confirmed Berthelot 1 s hypothesis that the hydrocarbons, c10H16 , c15H24 , and C2oH3 2, are related to isoprene (C 5Ha ) • Monoterpenes, which are often referred to as terpenes, in the restricted sense, are composed of two isoprene units and, hence, possesses the empirical 2 formula c10H16 • Hemiterpenes (C 5H8 ) are compounds of single isoprene units and sesquiterpenes (c 15¾4 ) are composed of three such units. There are al- so the di-, tri-, tetra-, and polyterpenes, which contain four, six, eight, and numerous isoprene units, respectively. A diagramatic presentation or this classification of terpenes is shown below. Hemiterpenes (C 5H8 ) c-c-c-c t C Monoterpenes (C 1oH 16 ) I I c-c-c-crc-c-c-c d : ~ Sesquiterrenes (C1 5H2 4 ) I I c-c-c-c+c-c-c-ctc-c-c-c ~ I t I t 1 C I C Diterpenes (c 20H32 ) 1 I I I I I c-r.-c-CtC-0.-C-CtC-C-C-C~C-G-C-C l l I t I I C : C I C Triterpenes (C3oli4a) : I C 1 I I 1 I I I It; I c-q-c-c~c-c-c-cfc-c-c-crc-c-c-c+c-c-c-c+c-c-c-c I I , I t I t I I C I C 1 C Tetraterpenes (C40H54) I I I C 1 1 1 I I I C I 1 I I c-c-c-cLc-c-c-c+c-c-c-c~c-c-c-c+c-c-c-ctc-c-c-crc-c-c-c,c-c-c-c C I C I I C I I d Polyterpenes (c 5H8 )n r c-~-c-cfc-¥-c-cl c-----------c C Jn I I C I I ~ I I 6 I I C 3 The most common empirical f'ormula f'or monoterpenes, Clolils,, shows a lack of six atoms of' hydrogen to satisf'y the empirical formula f'or a saturated aliphatic hydroparbon (Cn¾n+-2 ) with the same number of ca,rbon atoms. This dif- ference in hydrogen content may be accounted for by the presence of double bonds and/or by a cyclic . . structure. Structures which 'are in agreement with the empirical formula c H are tabulated below. 10 16 structural type Number of rings Number of double bonds acyclic monocyclic bicyclio tricyclic none 3 1 2 2 l 3 none All four of these structures have been found in terpenes; examples for each structural type are cited below. Myrcene (acyclic) Limonene (monocyclic) 17\-Pinene (bicyclic) Tricyclene (tricyclic) The natural-occurring acyclic, monocyclic, and bicyclic monoterpenes are presented in Figures l, 2, and 3, respectively. Ocimene Myrcene Citronellol (Rhodinol) Geraniol ( cis) Nerol (trans) Linalool dl-Ci tronella l Rhodinal Citral a {cis) Citral b (trans) / Tagetone 2,6-dimethyl-J octen-4-one Figure 1. Artemesia ketone Natural-occurring acyclic monoterpenes Lavandulol ¢ 2 L) • Terpinolene o<-Terpinene Limonene l 6 A () ) ) l ~ :"8(9 M;n hadieK; ~ p-Cymene ) II y ) () A Y-Terpinene Dl-Phe llandrene Cl-< CL( ~-Phe landrene ¢-OH Sylvestrene 1 3 5 8 9 6 , , , ( )p-Menthatetraene / . Carveol OH OH OH - OH P~ lyl alcohol Piperitol o<-Terpineol Dihydroc~veol Dihy ro-0-terpineo ~ X:" ) d-Terpinen-4-ol CHO 1,8- ineole qo 1,4-Cineole I I 0 ~o , Cryptal 1-Ca rvomenthone ¢ Phellandral 1-Dihydrocarvone a ~ / d-Pulegone o Buchucamphor dC-Santolinenone (Diosphenol) Isopulegone / ,_,,,,c,c~ i_j )~o Irone (provisional) Figure 2. / ¢,o ll Menthone ¢ Ascaridole A Carvone CHO Peri la aehyde I / C.t-10 Natural-occurring monocyclio monoterpenes Piperitone ~ -Pinena CIX 1-, -Pinene d-A3 -carene d- i -carene CIX CK3 Cl(~ Santene Sabinene dro" Borneol @OH Pinocampheol l -',-Tluljene Camphene HOG cirOH dl-Pinocarveol Santenol CHO 01; 6oH ~ Tluljyl alcohol d-Sabinol (IX dJO Fenchone 1-Santenone 6= 0 ti) Gr d-.Myrtenal 1-Pinocamphone (tyo Camphor ~o ) Umbellulone Figure 3. Thujone Natural-occurring bicyclic monoterpenes ~=o d-Verbenone 4 It was an accepted concept, for a long period of time, that the isoprene units in all terpenes are connected in a regular 11 head-to-tail 11 pattern; this arrangement of isoprene units is evident from an examination of the terpene structures shown on page 3. This concept was disturbed, however, when it was discovered that the structures of fenchone and artemesia ketone are not of this pattern; a confirmation of this irregularity was later furnished by the actual synthesis of these compounds. Later it was found that lavandulol also does not conform to the regular isoprene rule. Thus, it was realized that the 11 head-to-tail 11 arrangement is not valid for all terpenes. This invalidity of the old concept became evident as studies in sesquiterpenes, triterpenes, etc. proceeded. ctx Fenchone Artemesia ketone Lava.ndulol The empirical forUUlla for sesquiterpenes, c ~ 4 , was first suggested by 15 Wallach. Later, Selllilller and others independently advanced a classification of the sesquiterpenes which has greatly aided subsequent research on these compounds. A simple calculation shovra that the sesquiterpenes may be divided into four groups, which are as shovm. 5 Structura l type Number of double bonds acyclic mono cyclic bicyclic tricyclic 4 3 2 1 15° d 15° o.84 .89-.87 .92-.90 .935-:91 nu (Mit )n 1.53 1.51-1.49 1.51-1.50 1.50-1.49 69.5 67.8 66.l 64.4 Examples of these structural variations are sho~m in Figure 4, which gives the structures of the natural-occurring sesquiterpenes. Up to the early part of this century the constituti on of only one sesquiterpene alcohol, farnesol, had been determined. The progress of research in this field was slow because ordinary oxidation methods are not usually successful in elucidating the structures of these compounds; the oxidation products obtained were, in general, too simple or too complex to furnish any clue on the structure of the parent compound. A great advancement in sesquiterpene research was made by Ruzicka ~ ~ ( 1 .,2). These investigators found that, on dehydrogenation with sulfur, cadinene gave the na phtha lene hydrocarbon, cadalene (l,6-dimethyl-4-isopropylnaphthalene). By d~hydr,ogenating eudesmol and selinene, another naphthalene hydrocarbon, eudalene (l-methyl-7-isopropylnaphthalene), was obtained. Assuming that all terpenes are composed of whole ltnlltiples of isoprene units (c 5n8), it is evident that in the formation of eu- dalene from eudesmol or selinene, a loss of one methyl group has occurred. The missing methyl group is probably an angular methyl group which i s relatively unstable. The structure of eudalene has been confirmed by actual synthesis. Farnesol Bisabolene c:,(-Cadinene Nerolidol Zingiberene Selinene Eudesmol Guaiol Vetivone re A 0 0 Copaene * Santonine Elemo1* o(-Cyperone Caryophy llene * ©<f<CH~OH d.-Sa.ntalol -©)_ Cedrene* * These structures have not been definitely established Figure 4. Natural-occurring sesquiterpenes 6 Cadalene Eudalene It has been observed that a number of bicyclic sesquiterpenes and their derivatives give rise to a.zulenes on dehydrogenation. Azulene• f ormation was found possible only in a system which possesses either a cyclopentacycloheptane rrucleus or one which could give rise to this bicyclic structure. The formation of S-guaiazulene and vetivazulene from guaiol and vetivone~ respectively, shows that azulenes are bicyclic hydrocarbons with probably a cyclopentacycloheptane nucleus containing five conjugated double bonds. This S-Guaia zulene configuration was confirmed by the preparation of characteristic polynitro derivatives of azulenes and by actual syntheses (For a detailed discussion 7 on the chemistry of azulenes see the section on azulenes below). It has been pointed out that the substituted groups in a natural-occurring azulene are so situated in the nucleus that an aliphatic sesquiterpene, as farnesol, could assume the azulene configuration by a rearrangement of its acyclic chain. The investigation of azulenes, along with that of cadalene and eudalene, has been of great value in clarifying the structures of bicyclic sesquiterpenes. As with the sesquiterpenes, the diterpenes have only becently been investigated and, hence, not much is lmown. (c Many of the dtterpenes form retene H J on sulfur or selenium dehydrogenation; this characteristic reaction 18 18 Retene serves as a basis for the structural determination of the diterpenes. st~~ctures of some natural-ocourring diterpenes are shown. Abietic acid d-Pimaric acid The 8 Sclareol o<-Campho rene An acyclic representative of the triterpenes, which are composed o:f' six branched c5 units, is squalene (c 30H50 ), found in shark liver oil and .rrany other oils, as pearrut and olive oils. It is interesting to note that squalene may be synthesized from farnesyl bromide. The terpenoid saponins, many with - a ~ -amyrin skeleton, comprise a large portion of the known, natur~l-occurring triterpenes. A characteristic degradation product from the dehydrogenation of the terpenoid saponins is 1,8-dimethylpicene (Cz4 H18 ). (3 -Amyrin skeleton 1,8-Dimethylpicene 9 The tetraterpenes, consisting of eight branched c5 units, are represented by the carotenoid.s, the plant and animal pigments which are found in various colors, namely, yellow., orange, red, and violet. carotenoids is A well-knovrn example of the 6-carotene. h~~~X) ()-Carotene We have observed that terpenes nay be divided into isoprenoid units. N.any investigators have speculated that such a uni·b is the precursor of terpenes in vivo, but there is no definite experimental evidence to support this hypothesis. The first po&&ible precursor proposed for terpenes is isoprene 1 which is postulated to be formed from the condensation of acetone and acetaldehyde. It was suggested that possibly by the polymerization of these iso- prene units terpenes may be formed. In the laboratory dipentene and bisabo- lene have been obtained from polymerized products of isoprene. Although it is possible to prepare terpenes by the polymerization of isoprene in vitro, there is no direct evidence that isoprene is involved in the formation of terpenes in plants. The hypothetical precursor, 3-methyl-butenal., has also been postulated for terpenes. From 3-methyl-butenal., Fischer succeeded in synthesizing de- hydrocitral with which the synthesis of other terpenes could be easily 10 CH3 H CH 3 Ic == CI - C1/' 0 .-0 . .______ H Dehydrocitral 3-Methyl butenal carried out. It has been suggested that 3-methylbutenal is probably :formed in nature from the condensation of acetone with pyruvic acid with subsequent decarboxylation. other precursors which have been considered are isovaler- aldehyde, acetoacetic acid-acetone, etc. The posaible precursors of terpenes which have been discussed so far are based on the assumption that the c5 branched chain is the fundamental building unit of terpenes. It is also hypothetically possible that terpenes may have their origin in larger units. Accordingly, it has been postulated by Emde (3) that terpenes may result from a synthesis from sugars through a coupling of molecules similar to levulinic acid with subsequent decarboxylation and with the addition of smaller products of sugar metabolism. A similar hypothesis was proposed by Hall (4), who suggested that terpenes may be formed from the condensation and degradation of sugar derivatives; "half molecules" which may give the desired structures were postulated. Undoubtedly enzymes play an important part in the synthesis of terpenes in plants, but the nature of the mechanism involved is obscure. Previously, it has been sho,m that most terpenes possess a regular headto-tail union of isoprenoid units. In later years, an increasing number of 11 terpenes with an irregular union of isoprenoid units have been found. With the desire to add to the limited information on sesquiterpenes the structural investigation of a sesquiterpene alcohol was undertaken. The arrangement of the isoprenoid units in this particular terpene would be of interest. 12 Experimental Section Isolation of Partheniol In extensive studies conducted in our laboratory on the constituents of guayule, a crystalline substance which melts at 125-126° c. was isolated from the entire plant from which the leaves had been removed. Initial isolation of the compound was carried out by exhaustive extraction of the guayule bark with cold alcohol; after the concentration of the extract, the crystalline substance separated out in the cold (0-5° c.). In later experiments the de-leafed plant was extracted with 86-100° petroleum ether. The concentrated petroleum ether solution was chromato- graphed on silicic acid columns; development was carried out with petroleum ether. The unknown substance was adsorbed in a band in the lower section of the column. alcohol. Elution of the compound from the adsorbent was carried out with After recrystallization from alcohol the substance gave a melting point of 125-126° c. By qualitative test it was indicated that the crystal- line compound was an ester. The test consisted essentially of reacting hy- droxylamine with the unknown ester of carboxylic acid to form a hydrox~mic acid which gives a colored inner complex iron salt with ferric chloride. The ester, on hydrolysis, yields cinnamic acid and an unknown sesquiterpene alcohol of melting point 127-128° C. Analysis for carbon and hydrogen showed that the empirical fornrula for the sesquiterpene alcohol is c15 H2 4 o. In 1911 Alexander (5) obtained, on prolonged steam distillation of the acetone-soluble portion of guayule, a crystalline product which melted at 127-128° c. Alexander thought that this crystalline solid which liquefied 13 on long standing was a sesquiterpene alcohol. This tendency to liquefy on long standing has also been observed with the unknown sesqui terpene alcohol. It is quite probable tha.t the crystalline alcohol obtained by Alexander is identical with this compound. In studies conducted on the isolation and improvement of the quality of rubber Walter (6) isolated an unknovm crystalline ester from the sticky exudate of injured roots and bark of the guayule shrub. He also obtained this substance from a concentrated acetone extract of guayule rubber. On recrys- tallization from acetone the ester gave tetragonal crystals melting at 1261260 c. The substance was soluble in organic solvents and was insoluble in water • . Hydrolysis of the ester in an alcoholic alkali solution gave a sesquiterpene alcohol and cinnamic acid. From Parthenium argentatum, Gray (gua.yule) Walter proposed the name partheniol for the sesqui terpene alcohol and parthenyl cinnamate for the ester. of' long crystalline needles. Partheniol was obtained in the form The resublimed crystals gave a melting point of 131° c. and were soluble in organic solvents and slightly soluble in hot water. From analysis for carbon and hydrogen Walter proposed c15H26 o as the empirical formula for partheniol. Since parthenyl cinna.mate melted at the same temperature as the crystalline ester which we have obtained, it is apparent tha.t the two compounds are identical. Further evidence to support this was the isolation of cinnamic acid from the hydrolyzed ester. Vl~ lter has proposed C15H2sO for the empirical formula of partheniol, as compared to our .formula, c16ttz 4 O, and has reported a melting point of 131° C. The melting point of p..rtheniol has been found to be 127-128° c. in this laboratory; no change in the melting point was observed 14 after repeated recrystallization of the compound from petroleum ether, benzene, Melting points of 129.5-130.5° c. and 132-132.5° c. have been and acetone. obtained from partheniol which had been recovered from a test for a conjugated double bond; it is probable that molecular rearrangement had occurred during the reaction. In the early period of the study, partheniol was obtained from the hydrolysis of parthenyl cinna:mate isolated from the guayule plant. The maxinru.m yield of parthenyl cinnamate which was obtained from fresh ground plant was approximately 0.05 percent, or about 0.03 percent of partheniol. Upon inves- tigation it was found that a 0.2-0.3 percent yield of partheniol w.ay be obtained from guayule resin which is a concentrated extract of crude guayule rubber. This resin is a by-product in the manufacture of guayule rubber. The method used for the isolation of partheniol from guayule resin is as follows. About 25 gallons of crude resin was extracted three times with five-gallon portions of ether and twice with a mixture of five gallons of ether and two gallons of methanol. Methanol was used in the last two ex- tractions in order to decrease the solubility of rubber and its oxidized products. After the removal of the solvents from the extract, the ether-soluble portion of the guayule resin was saponified with 0.5-1.0 N alcoholic solution of sodium hydroxide for approximately three hours at a bath temperature of about 90° c. The alkaline alcoholic solution was then neutralized with dilute sulfuric acid. Extraction of the hydrolyzed product ~ra.s carried out with sev- eral two-gallon portions of ether; the solvent was removed at the reduced pressure of an aspirator. under high vacuum. The residue (about five gallons) was distilled In a preliminary distillation, the fraction which distilled 15 up to 150° c. at l-10 mately two liters. f pressure was collected; this fraction was approxi- Further fractionation was carried out at 1-10 )l pressure with a short helix-packed column and a variable take-off still head. The fraction which was collected at 100-125° C. was approximately 1.2 liters of viscous oil. By diluting the oil slightly with petroleum ether and inoculat- ing with a few crystals of partheniol a solid mass of partheniol vvas obtained, after allowing to stand in the cold room for several days. The crystalline partheniol was filtered on a Buchner funnel and the mother liquor was again placed in the cold room in order that further crops of crystals may be obtained. After recrystallizing twice from 60-70° petroleum ether the partheniol was washed with three one-liter portions of hot 30-60° petroleum ether. The washings were carried out by placing the partheniol in the petroleum ether and allowing the solvent to boil vigorously. Subsequently, the mixture was cooled to room temperature and the crystals were filtered on a Buchner funnel. The purified crystalline partheniol was free of oily odor and possessed only a very delioate odor. A total of about 150 g. of purified partheniol, m.p. 127- 128° c., was obtained. The scheme used in the isolation of partheniol from crude guayule resin is summarized in the following diagram. 16 crude resin ~ ether insoluble resin her extraction ether-soluble resin ! ether removed residue ! hydrolysis with alcoholic sodium hydroxide neutralization with sulfuric acid hydrolyzed product ~ extraction ether-alcohol-insoluble, hydrolyzed resin ether-alcohol-soluble, hydrolyzed resin i ether, alcohol removed residue I di:tillation under high A residue cuum crude oil fraction t !fractional distillation under high vacuum fractionated oil tI inoculation with partheniol crystals crude partheniol A mother liquor lre~ rystallization from 1 etroleum ether purified partheniol 17 Quantitative and Qualitative Analyses of Partheniol* I. Analysis for carbon and hydrogen Calculated for c1£¾ 4 0: Found: C, 81.76%; H, 10.98% (a ) C, 81.77%; H, 10.92% (b) C, 81.73%; H, 10.98% (a) 2.280 mg. substance gave 6.832 mg. carbon dioxide and 2.226 mg. water (b ) 2.164 mg. substance gave 6.481 mg. carbon dioxide and 2.124 mg. water II. Determination of molecular weight (Rast) M.W., 220.34 M. W., 216.2 . 0.412 mg. substance in 3.420 mg. camphor gave a depression of 21.5 III. Determination of optical rotations 60 ~ ]~ : 83.6° (cone. = 1.47% in absolut.e alcohol) 116.5° (cone.: 1.29% in chloroform) IV. Tests for primary and secondary alcohol groups in partheniol A mixture of 7.8 g. of partheniol and 5.3 g. of phthalic anhydride was heated for 6 1/2 hours at a bath temperature of 130° C. under an atmosphere of nitrogen. No ester of phthalic acid was formed; hence, the alcoholic group was tertiary. A qualitative test with carbon disufide and sodium hydroxide reagents * Micro-analyses were done by G. Oppenheimer, G. Swineha~ and B. Dandliker 0 C. 18 failed to give a xanthogenate. The failure of the alcohol to react with these reagents again indicates that it is tertiary. V. Determination of unsaturation of partheniol by hydrogenation A. 9.404 mg. of partheniol in glacial acetic acid with 3.240 mg. of platinium dioxide took up 1.9 ml. of hydrogen at STP. Calculated for c15 H240: Found: B. 1.9 ml. hydrogen at STP for 2 double bonds 1.9 ml. hydrogen at STP; 2 double bonds 13.360 mg. of partheniol in glacial acetic acid with 6.248 mg. of platinium dioxide took up 2.66 ml. of hydrogen at STP. Calculated for c15H24 o: Found: 2.72 ml. hydrogen at STP for 2 double bonds 2.66 ml. hydrogen at STP; 2 double bonds VI. Tests for conjugated double bonds A. Reduction with sodium and alcohol Two grams of partheniol in 10 ml. of aIJ\Yl alcohol was heated in a 40-ml. flask at a bath temperature of 135° c. for 6 1/4 hours. Sodium (4.5 g.) in small pieces was subsequently added; at periodic intervals more amyl alcohol was introduced until all the sodium had reacted. After allowing to cool, water was slowly added to the reaction mixture; at this point a substance precipitated out. The alcohol phase was separated and combined with an ether solution of the precipitated material. After washing with water till neutral and drying over anhydrous sodium sulfate, the alcohol-ether solution was filtered and the solvent was removed at a reduced pressure. The crystalline residue was recrystallized twice from low boiling petroleum ether; the substance melted at 132-132.5° C. A mixture of the compound and partheniol gave a melting point of 130-130.5° c. 19 Analysis of recovered product for carbon and hydrogen: C, 81.76%; H, 10.98% C, 81.02%; H, 11.79% C, 81.85%; H, 11.06%; therefore compound is c15H24 o It has been sho1rm from the analysis that no reduction of partheniol had occurred; the double bonds in partheniol were, therefore, not conjugated, but isolated. The slightly higher melting point of the recovered partheniol is probably due to molecular rearrangement which may have occurred in the course of the reaction. B. Reaction with maleic anhydride (Diels-Alder reaction) A mixture of 600 mg. of partheniol, 300 mg. of maleic anhydride, and 3 ml. of anhydrous benzene in a 10-ml. flask was refluxed at a bath temperature of 100° c. for 10 1/2 hours. After cooling, the precipitated substance was filtered and the benzene filtrate was evaporated at reduced pressure. The remaining oil from the benzene solution was allowed to stand in the refrigerator for several days; no crystals were formed. The solid which had been recovered from the original reaction mixture was washed with two portions of benzene and was subsequently centrifuged; the substance melted at 134-136° c. and decomposed above 136° c. It was shown from the melting point, the decom- position temperature and solubility in water and alcohol that the compound was maleic acid. Hence, it may be concluded that no adduct was formed in the Diels-Alder reaction. A similar experiment, in which 20 mg. of partheniol and an equal amount of ethyl maleate in 0.75 ml. of anhydrous benzene were used, also showed no adduct formation. The recovered partheniol, after reo crystallization from cyclohexane gave a melting point of 129.5-130.5 c. 20 Analysis of recovered product for carbon and hydrogen: Calculated for c23 H36 o (conjugated product): C15Hz40: Found: C, 70.37%; H, 9.26% C, 81.76%; H, 10.98% C, 81.96%; H, 10.06% 2.287 mg. substance gave 6.869 mg. carbon dioxide and 2.055 mg. water VII. Test for homogeneity of partheniol In the isolation of partheniol from guayule concentrate, repeated recrystallization of the compound from petroleum ether gave a constant melting point of 127-128° C. The constancy of the melting point after further recrys- tallization from benzene and acetone was determined to check the purity of the compound. In order to further establish the purity of the isolated par- theniol and to investigate whether the substance is possibly a mixture of isomers, a chromatographic study of its derivative, plrthenyl-p-phenylazobenzoate, was undertaken. The red-orange parth~myl-p-phenylazobenzoate was chro- w~tographed instead of partheniol because its color serves as a convenient index of adsorption. A. Constancy of melting point of partheniol About one gram of partheniol (m.p. 127-128° c.) which had been purified by repeated recrystallization from petroleum ether was recrystallized twice from benzene and once from acetone. The melting point of partheniol was 127- 128° c. after the first recrystallization from benzene, 128-128.2° C. after the second recrystallization, and 126.5-128° c. after the final recrystallization from acetone. From the constancy of its melting point partheniol appears to be a pure substance. 21 B. Chromatographic studies 1. Preparation of parthenyl-p-phenylazobenzoate To 2 g. of partheniol in 15 ml. of pyridine in a 40-ml. flask, 2.4 g. of p-phenylazobenzoyl chloride in 20 ml. of anhydrous benzene was added. The reaction mixture was heated at a bath temperature of 90° c. for 1/2 hour. After allowing to cool the mixture was diluted with alcohol-free ether and washed with four 10-ml. portions of 3 N hydrochloric acid, followed by three 10-ml. portions of 1 N potassium hydroxide. The solution was then thoroughly washed with water and dried over anhydrous sodium sulfate. The residue which remained after the evaporation of the solvent mixture was recrystallized, first, from a mixture of benzene and petroleum ether, then, from acetone. After drying in a vacuum desiccator the parthenyl-p-phenylazobenzoate was used for chromatographic studies. 2. Chromatogr9:phy of parthenyl-p-phenylazobenzoate Approximately 1 g. of parthenyl-p-phenylazobenzoate dissolved in a mixture of 10 ml. of benzene and 50 ml. of petroleum ether (85-100° c.) was chromatographed on a silicic acid-Celite column (2:1) which had been initially washed with 250 ml. of petroleum ether. The column was developed with 400 ml. of petroleum ether (85-100° c.) and post-washed with 50 ml. of 30-60° petroleum ether. Two very faint narrow zones which were situated at the top of the column and about 7 cm. from the top, respectively were observed. A main band of parthenyl-p-phenylazobenzoate which was about 5 cm. wide was situated approximately 10 cm. from the top of the column. After extruding the column this zone was cut out and the benzoate was eluted from the adsorbent with ether; the ether solution wa.s dried over anhydrous sodium sulfate, filtered, 22 and the benzoate recovered. The compound vra.s recrystallized three times from acetone; the melting point of the pure parthenyl-p-phenylazobenzoate is 162.5164° c. Analysis of recovered parthenyl-p-phenylazobenzoate: Calculated for c28 H32 o2 Na: Foun : c, 78.471l; H, 7.53%; N, 6. 54<1o c, 78.48%; H, 7. 53%; N, 6.66% 2.675 mg. substance gave 7.693 mg. carbon dioxide and 1.800 mg. water 4.893 mg. substance gave o.297 ml. nitrogen at 24° c. and 743 mm. Hg The two very faint bands were probably impurities which were present in either the reagents used in the preparation of the benzoate or in partheniol itself. In any case, the total quantity of these impurities is insignifi- cantly small as compared to the amount of benzoate chromatographically separated. From the above chromatographic study, partheniol appears to be a homogeneous substance. 23 Absorption Spectrum of Partheniol The absorption spectrum was determined with the Beckman photoelectric spectrophotometer. A solution of a concentration of l.76 mg. of partheniol in 100 ml. of absolute alcohol was used. The spectrum, as given in Figure 5 1 did not show any characteristic maxima or minima within the range of the Beckman photoelectric spectrophotometer. 0.9~------------------------, 0 .8 0 .7 .,_ 0 .6 z IJJ 0 IJ.. ~ 0 .5 0 0 z 0 .,_ 0.4 0 z ....X w 0.3 0.2 0.I 210 220 230 WAVE Figure 5. 240 250 260 270 LENGTH, M}J Absorption spectrum of partheniol in absolute ethanol 24 Preparation of Parthenyl-3,5-dinitrobenzoate A solution of approximately 150 mg. of i;artheniol and an equal amount of 3,5-dinitrobenzoyl chloride in a mixture of 2 ml. of dry benzene and 5 ml. of anhydrous pyridine was gently refluxed for 3/4 hour. After cooling, the solu- tion was diluted with alcohol-free ether, , then washed with several portions of dilute hydrochloric acid, followed by washing with dilute sodium hydroxide. The ether solution was thoroughly washed with water till neutral and dried over anhydrous sodium sulfate; the solvent was then removed under reduced pressure leaving behind a viscous oil. The oil was diluted with alcohol, then water was added till the solution was turbid. The turbid solution was allowed to stand in the refrigerator for several days. The crystals formed were recrystallized from acetone three times and were then washed with alcohol . .After drying under vacuum over phosphorus pentoxide in a drying pistol at 56° C. the substance gave a melting point of 143-144° c. Analysis: Calculated for c22 Hz 6N205: Found: C, 63.75%; H, 6.31%; N, 6.76% (a) C, 63.88%; H, 6.61%; N, 6.67% (b) C, 63.97%; H, 6.51%; N, 6.90% (a) 3.040 mg. substance gave 7.116 mg. carbon dioxide and 1.795 mg. water 5.635 mg. substance gave 0.341 ml. nitrogen at 22° c. and 742 mm. Hg (b) 2.806 mg. substance gave 6.577 mg. carbon dioxide and 1.633 mg. water 2.842 mg. substance gave o.180 ml. nitrogen at 25° c. and 742 mm. Hg 25 Preparation of Parthenene A mixture of 5 g. of partheniol and 3.5 g. of fused potassium acid sulfate was heated in a 50-ml. flask fitted with a reflux condenser at a bath temperature of 180-190° c. for 1/2 hour. The oil was extracted with ether and the ether extract was washed with water, then dried over anhydrous sodium sulfate. The ether was removed under reduced pressure and the residual oil was distilled twice over potassium. In the second distillation at 60-65° C. and 100 p, 1.1 g. of parthenene was obtained. I. Analysis for carbon and hydrogen Calculated for c15¾2 : Found: C, 89.04%; H, 10.96% C, 88.90%; H, 11.15% 2.062 mg. substance gave 6.717 mg. carbon dioxide and 2.054 mg. water II. Physical constants n 26° D : l.5120 calculated: Found: Exaltation: III. 65.67 for 3 double bonds 66.04 0.37 Absorption spectrum A solution of 6.0 mg. of parthenene in 100 ml. of absolute ethanol was examined with the Beckman photoelectric spectrophotometer. At 212-330 3/' the extinction coefficient decreased with the 1ncrease of the wave length; no maximum nor minimum was detected. 26 Partial Hydrogenation of Partheniol ~ Raney Nickel A 5.04-g. sample of partheniol dissolved in 50 ml. of absolute ethanol was hydrogenated with approximately 2 g. of Raney nickel as catalyst at room temperature and at a pressure slightly above atmospheric. The rate at which the hydrogenation proceeded was about 30 ml. of hydrogen per hour for the first 15 hours and decreased thereafter; at the end of 27 hours the rate was 5 ml. of h;i7drogen per hour. The hydrogenation was stopped shortly after the required amount of hydrogen was taken up. For the 5.04-g. sample, 514 ml. of hydrogen is required at STP; in the hydrogenation 582 ml. of hydrogen was ta.ken up at S'l'P, giving 1.1 double bonds. After the removal of the Raney nickel and the ethanol, the remaining oil was distilled at 92-102° c. at 100 fi a total of 3.5 g. of dihydropartheniol was obtained. I. Analysis for carbon and hydrogen Calculated for c1£H26 o: Found: C, 81.02%; H, 11.79% C, 81.05%; H, 11.96% 2.036 mg. substance gave 6.047 mg. carbon dioxide and 2.177 mg. water II. Hydrogenation 12.955 mg. substance and 5.947 mg. platinium dioxide in cyclohexane gave no uptake of hydrogen 10.259 mg. substance and 5.063 mg. platinium dioxide in methylcyclohexaneglacial acetic acid (l:l) took up 1.84 ml. hydrogen at STP Found; 1.04 ml. hydrogen at STP for l double bond; 2.08 ml. hydrogen at STP for 2 double bonds 1.84 ml. hydrogen at STP; 2 double bonds (probably reduction of the hydroxyl group) 27 III. Physical constants 26° n D : 1.5035 calculated: Found: IV. 68.13 (c 15¾6 o), for 1 hydroxyl group 68.50 Preparation of dihydroparthenyl-p-phenylazobenzoate Approxiroo.tely lg. of dihydropartheniol in 15 ml. of dry pyridine was added to 1.3 g. of p-phenylazobenzoyl chloride in benzene-pyridine. The reaction mixture was heated at a bath temperature of 100° c. for 1/4 hour and was then diluted with alcohol-free ether. The pyridine was removed with 3 N hydrochloric acid; the p-phenylazobenzoic acid with 1 N potassium hydroxide. The ether-benzene solution was washed with water, dried over anhydrous sodium sulfate, and the solvents were removed under reduced pressure. In order to separate the substance from its impurities, the dihydroparthenylp-phenylazobenzoate was dissolved in 30 ml. of 85-100° petroleum ether and chromatographed on a column of silicic acid-Celite (2:1), 4.5 cm. x 20 cm., which had been pre-washed with about 250 ml. of 85-100° petrol ether. The chromatogram was developed with 150 ml. of 1:1 85-100° petroleum ether-100130° petroleum ether. The main zone of the azobenzoate was about 15 cm. from the top of the column; three small zones all of which were more strongly adsorbed than and well separated from the benzoate were observed. The main zone was cut out and the substance was eluted from the adsorbent with ether. ether solution was dried over anhydrous sodium sulfate and the ether was The 28 removed. The recovered dihydroparthenyl-p-phenylazobenzoate gave a m.p. of 121.5-122.5° c., after three recrystallizations from acetone. Analysis for carbon and hydrogen: C, 78.20%; H, 7.96%; N, 6.51% C, 78.40%; H, 8.20%; N, 6.58% 2.708 mg. substance gave 7.781 mg. carbon dioxide and 1.984 mg. water 4.884 mg. substance gave 0.293 ml. nitrogen at 24° c. and 743 mm. Hg v. Conclusion It has been shown that one of the double bonds in partheniol is easily hydrogenated. 29 Preparation of Azulenes from Partheniol There are found in literature many references to blue colored fractions in essential oils. In 1894 Piese gave the name Azulena to the blue substance obtained from wormwood oil. The general t r end of thou ght at that time was that the blue oil was an oxygenated compound of alcoholic nature. From the properties of azulene its nucleus was considered to be related to the sesquiterpene skeleton. Later, it was found that azulenes were bicyclic hydrocarbons. Since the only simple colored hydrocarbons lmown at that time were f'ulvene and its derivatives, the benzofulvene structure was proposed for azulenes. Many of the reactions of azulenes were explained on the basis of a benzof'ulvene structure. With the actual synthesis of azulenes, however, this concept of a benzofulvene structure was discarded in favor of a conjugated cyclopentacycloheptene nucleus. Hilckel et al (7 ,8 ) were the first t o synthesize cyclopentenocycloheptanone by ozonizing octalin to form a diketone which was subsequently treated with aqueous sodium carbonate or dilute sodium hydroxide. The reactions are as presented below. 0 co d , 10octalin 0 II C) <JO 0 30 Pfau and Plattner (9) condensed cyclopentenocycloheptanone with various Grignard reagents and dehydrogenated the products with nickel or sulfur to form various azulenes with different substituted groups. These investigators (10) also prepared the parent azulene by reducing the cyclopentenocycloheptanone, first, to the dihydro compound, then, to the carbinol which was subsequently dehydrogenated with palladium-charcoal. The reactions involved are given below. f 19 <D RMgX ➔ cO 0 OH ·II I <JO H2 cO R I s > cO Pd-C >ca A recent method for the synthesis of azulene is t o treat the desired substituted indane with diazo-aoetio ester, followed by hydrolysis of the ester. The bicyclic acid is then decarboxylated and dehydrogenated with palladium-charcoal or sulfur to form the desired azulene. are as follows. The reactions 31 Pd-C COOR co From bond energy considerations* it may be shown that the azulene nucleus is less stable than that of the naphthalene nucleus. Hence, it is not surpris- ing that an azulene may be readily transformed into a naphthalene derivative by treatment with hydroiodic acid and red phosphorus, as given below. HI+ P 0\-1 Two stable valence bond structures may be fornn.tlated for the parent azulene; these are as given. cO co * Heats of combustion of 1,4-dimethyl-7-isopropylazulene and 3,8-dimethyl5-isopropylnaphthalene are 2022.~9 and 1993 .4 Gals., respectively (Perrottet, Traub, and Briner, Helv. Chim. Aota 23, 1260-8, 1940). 32 Generally azulenes may be formed from essential oils by dehydrogenation with sulfur, selenium, palladium-charcoal or nickel. There may be found re- ported in literature a number of azulenes each of which was named after the natural oil from which it is derived. Among these supposedly different azu- lenes are possibly identical compounds, since the chemical nature of most of these azulenes apparently has not been investigated. In an investigation on a numb·er of azulenes Ruzicka and Haagen-Smit (11) suggested that kessazulene is identical to S-guaiazulene. The number of actually different azulenes which may be derived from natural sources is probably limited; a few which have been reported are ~riefly discussed below. s~Guaiazulene (m.p. 30° c.) which is obtained by the S-dehydrogena.tion of guaiol is a blue compound. and other essential oils. It may be derived also from Eucalyptus globulus Camazulene (m.p. 132° c.), the structure of whioh has not been entirely elucidated, has been obtained from camomile oil. Vetiv- azulene (m.p. 32° c.), a violet compound, is derived from sulfur dehydrogenation of vetivone or vetiver oil. The structural forrulae of S-guaiazulene and vetivazulene are presented in I and II, respectively. I II In studies on azulenes, the phosphoric acid method of isolating the azulene has been extensively used. Azulenes are usually characterized by such 33 addition products as the picrates and the styphnates. More stable crystalline addition products with trinitrobenzene and trinitrotoluene also may be prepared. The purification of an azulene is afforded by the easily decomposable polynitro addition products; the separation of the regenerated azulene from the polynitro reagent is accomplished by chromatographic adsorption on activated a ltimina • In the investigation of the structure of partheniol, the selenium dehydrogenation of a small sample of the sesquiterpene alcohol was carried out; a blue violet azulene was obtained. The induction of an azulene from partheniol suggests that the sesquiterpene alcohol possesses a oyolopentacycloheptane nucleus. In a more extensive study of the preparation of azulenes from par- theniol, it has been found that dehydrogenation with sulfur and with selenium yield azulenes which are not identical. Previous studies by Ruzicka and Haagen-Smit on guaiol (11) have shown that dehydrogenation of guaiol with sulfur and selenium also yielded different azulenes. With sulfur, the blue S-guaiazulene was formed; with selenium, a violet azulene, or probably a mixture of azulenes was obtained. The preparation of and the studies made on the azulenes from partheniol are presented in this section. I. Dehydration of partheniol Twenty grams of partheniol and 12.6 g. of fused potassium acid sulfate were placed in a 200-ml. flask and heated at a bath temperature of 165-170° c. for an hour. The reaction mixture ~~s transferred into a distilling fla sk and the reaction flask was rinsed with ether; the ether washings were combined with the w.ain fraction in the distilling flask. After the re moval of the 34 solvent by distillation, the remaining oil was fractionally distilled under high vacuum. The crude parthenene which distilled at 78-100° c. and 0.15 nnn. Hg was collected and dehydrogenated as described below. In a separate experiment, 24.6 g. of crude parthenene was prepared from 30 g. of partheniol according to the procedure described above. II. Dehydrogenation of parthenene ~~th sulfur (9 1 12) Fifteen grams of crude parthenene was heated with 4.7 g. of sulfur under a reduced pressure of 30-50 mm. Hg at a bath temperature of 180-185° c. for 3 hours. The blue azulene which ~~s formed was distilled directly from the reaction flask. The fraction obtained at 70-100° C. and 10-15 nun. Hg was 7.0 g. A. Purification of S-azulene by the phosphoric acid method The crude blue azulene was dissolved in five times its volume of purified petrol ether and the solution was cooled to below o0 c. in an ice-salt bath. Approximately 0.3-ml. portions of 85% phosphoric acid were added with thorough agitation of the mixture after each addition. The addition of the phosphoric acid reagent was continued until the petrol ether solution was practically discolored; a total of 2.5-3.0 ml. of the reagent was used. The petrol ether solution was then drawn off and the phosphoric acid addition product which was formed was washed with four 15-ml. portions of petrol ether. To decompose the addition product finely crushed ice ¥ras added. liberated azulene was taken up in about 25 ml. of peroxide-free ether. The The ether solution was washed, first, with several 5-ml. portions of 1 N sodium hydroxide, then, with water until a neutral reaction with litmus ~~s obtained. 35 The ether was evaporated and the azulene was steam distilled. The steam dis- tillate containing the azulene was extracted with peroxide-free ether, the ether solution was dried over anhydrous sodium sulfate, filtered and the ether removed. The azulene was transferred into a micro-claisen flask and distilled 0 over sodium; the fraction which distilled at 140-150 and 50 f pressure was approximately 250 mg. C. (bath temperature) The percentage yield based on crude parthenene is about 1.7 percent. B. Purification of azulene by means of the 1,3,5-trinitrobenzene addition product Approximately 250 mg. of the blue azulene obtained from S-dehyd.rogenation, 300 mg. of 1,3,5-trinitrobenzene and 1.5 ml. of absolute alcohol were placed in a small test tube and heated to boiling. The stoppered teat tube and its contents were allowed to cool in the refrigerator. Ai'ter approximately an hour the crystalline trinitrobenzene addition product which had formed was filtered, washed with several small p~vtions of absolute alcohol and the excess alcohol was removed by suction. The thoroughly dried addition product was placed on a chromatographic column 2 cm. in diameter and packed with activated alumina 14 cm. high. On the addition of purified petrol ether to the column the trinitrobenzene addition product of azulene immediately decomposed into the original azulene and trinitrobenzene. The trinitrobenzene, being strongly adsorbed on alumina, formed a zone at the top of the column. The azulene was adsorp;e d a short distance below the trinitrobenzene zone. By continued addition of petrol ether to the column the azulene zone was washed free of weakly adsorbed contaminating substances; a total of 75 ml. of petrol ether was used. To elute the azulene from the adsorbent about 60 ml. of a 36 mixture of 1:1 benzene-petrol ether was passed through the column. The fil- trate containing the azulene was collected and the solvent was distilled off. A trinitrobenzene addition product of the recovered azulene was again prepared and the cycle of the decomposition an~ the regeneration of the azulene was repeated. The final product -Of azulene was distilled under reduced pressure. Ap- proximately 125 mg. of the purified blue azulene was obtained. Ana. l:1rsis of purif'ied azulene: C, 90.85%; H, 9.15% C, 90.93%; H, 9.24% 1.934 mg. substance gave 6.444 mg. carbon dioxide and 1.596 mg. water c. Absorption spectrum of azulene obtained from sulfur dehydrogenation of parthenene A solution of 3.62 mg. of S-azulene in purified hexane was made up to a total volume of 10 ml. The absorption spectrum of the azulene in hexane ~~s determined with a Beclanan photoelectric spectrophotometer using a slit width of 0.02 to 0.30 mm. The spectrum was characterized by three maxim~ at 603, 656, and 728 ~ and undulations at 557, 682, 630, and 698 rip, as shown in Figure 6. The absorption spectrum of S-guaiazulene, as reported by Plattner, is characterized by three maxima at 603, 661, and 732 1;Jll and undulations at 556, 577, 631, and 699 llJl (13). D. Preparation of polynitro derivatives of the sulfur azulene from partheniol 1. Trinitrobenzene addition product of S-azulene from partheniol A mixture of approximately 42 mg. of purified S-azulene, an equal amount IJJ IX - 500 540 580 620 660 WAVE LENGTH, M).J 700 740 780 Absorption spectrum of S-guaiazulene prepared from S-dehydrogenation of parthenene (hexane solution). 460 Figure 6. o.o 0.1 0.2 zo 0.3 r- 0 z 0.4 0 0 IJJ ( .) IJJ tt 0.5 ~ 0.6 0.7 0.8 0.9----------------------------------. 37 of trinitrobenzene, and 0.75 ml. of absolute ethyl alcohol was heated to boiling. The trinitrobenzene addition product which was formed was allowed to crystallize at a temperature of -10° c. with absolute alcohol and dried. The crystals were filtered, washed Recrystallization of the trinitrobenzene addition product was carried out three times f'rom absolute ethyl alcohol and once from absolute methyl alcohol. pentoxide for several days. The crystals were dried over phosphoric The purified 1,3,5-trinitrobenzene addition pro- duct of the sulfur azulene melted at 150-151° c. The melting point of the corresponding addition product of S-guaiazulene, as reported in literature, is 151-151.5° C. (9) Analysis of the trinitrobenzene addition product: Calculated for Cz 1H21N3 o6 : Found: C, 61.34%; H, 5.14%; N, 10.21~ C, 61.61%; H, 4.99%; N, 10.12% 2.993 mg. substance gave 6.757 mg. carbon dioxide and 1.333 mg. water 2.760 mg. substance gave 0.254 ml. nitrogen at .£3° c. and 742 nnn. Hg 2. Picrate of S-azulene from partheniol Approximately 26 mg. of the purified S-azulene, 26 mg. of picrio acid, and 0.3 ml. of absolute ethanol were placed in a test tube and heated to boiling. cold room. The stoppered tube with its contents was allowed to cool in the The picra.te of the azulene which precipitated from the reaction mixture was f'iltered, washed with absolute ethanol and recrystallized from the same solvent. The melting point of the picrate was 120.5-121° c.; that of the piorate of S-guaiazulene, as reported in literature, is 122.5-123° C. (9). 38 Analysis of picrate of S-azulene: C, 59.01%; H, 4.95%; N, 9.83% C, 58.79%; H, 4.88%; N, 10.09% 2.947 mg. substance gave 6.349 mg. carbon dioxide and 1.286 mg. water 2.810 mg. substance gave 0.258 ml. nitrogen at 24° C. and 744 mmi Hg 3. Trinitrotoluene addition product of s-azulene from partheniol A mixture of 30 mg. of the purified azulene, 30 mg. of trinitrotoluene, and 0.3 ml. of absolute ethanol was heated to boiling and the trinitrotoluene addition product which was formed was allowed to crystallize in the cold room at o0 c. The crystals were filtered, washed with absolute ethanol, and dried. The melting point of the trinitrotoluene addition product after recrystallization from absolute ethanol was 88-89° c.; that of the corresponding addition product of S-guaiazulene was reported in literature to be 89° c. (9) Analysis of trinitrotoluene addition product of S-azulene from partheniol: Calculated for c22 ¾rN3 ~o: OU : c, 62.ll%; H, 5.45%; N, 9.88% c, 62.30%; H, 5. 74%; N, 10.06% 2.786 mg. substance gave 6.360 mg. carbon dioxide and 1.430 mg. water 2 .922 mg. substance gave o.267 ml. nitrogen at 24° c. and 745 mm. Hg E. Conclusion The azulene obtained from the sulfur dehydrogenation of parthenene has been characterized by various polynitro derivatives which agree with those reported for S-guaiazulene. that of S-guaia.zulene. 'r he absorption spectrum is also identical with It is concluded that the sulfur azulene from partheniol is s-guaia.zulene, and hence, partheniol possesses a 1,4-dimethyl,7-isopropylcyclopentacycloheptane configuration as in guaiol. 39 III. Selenium dehydrogenation of parthenene A mixture of 24.6 g. of crude parthenene which was prepared from 30 g. of partheniol and 19 g. of selenium was heated in a 100-ml. flask in a metal bath at a bath temperature of 285-295° c. for 7 hours. The reaction product .was transferred to a claisen flask, rinsing the flask with a small amount of ether. The ether was evaporated and the azulene was distilled under reduced pressure. A total of 14.5 g. of crude purple azulene which distilled at 90- 100° c. and 1 mm. Hg was obtained. A preliminary purification of the crude azulene was carried out by means of the phosphoric acid method which was used for the purification of the sulfur azulene (Section II-A). Distillation of the azulene over metallic sodium gave further purification. At this stage 0.5 g. of azulene was obtained. Still further purification of the azulene was made by means of the trinitrobenzene addition product decomposition method as described in Section II-Bin connection with the work on the s-azulene from partheniol. This process of purification was repeated; the final yield of pure Se-azulene was 350 mg., which is 1.4 percent of the crude parthenene originally used. Analysis of purified Se-azulene: Calculated for c15H18 : Found: C, 90.86%; H, 9.15% C, 90.93%; H, 9.34% 1.925 mg. substance gave 6.414 mg. carbon dioxide and 1.606 mg. water A. Absorption spectrum of Se-azulene A solution of 1.75 mg. of purified azulene in hexane was made up to a total volume of 5 ml. The absorption spectrum of Se-azulene in hexane, as 40 determined by means of a Beckman photoelectric spectrophotometer, is characterized by three maxima at 553, 567, and 578 l1lJl and undulations at 490, 500, 533, 603, 616, 627, and 664 o/• as shovm in Figure 7. B. Preparation of polynitro addition products of Se-azulene from partheniol l • . Preparation of the trinitrobenzene addition product The 1,3,5-trinitrobenzene addition product of the Se-azulene from partheniol was prepared from 78 mg. of the azulene according to the procedure used for the preparation of the corresponding addition product of S-azulene and is as described in Section II-D. product was 103-104 0 The melting point of the crystalline addition C. Analysis of trinitrobenzene addition product of Se-azulene from partheniol: Calculated for C21H2 1 N3 05: Found: C, 61.34%; H, 5.14%; N, 10.21% C, 61.10%; H, 5.07%; N, 10.18% 2.884 mg. substance gave 6.457 mg. carbon dioxide and 1.306 mg. water 3.535 mg. substance gave 0.329 ml. nitrogen at 24° c. and 742 rrnn. Hg 2. Preparation of the picrate of Se-azulene from partheniol A mixture of 60 mg. of azulene and 50 mg. of picrate was used in the preparation of the picrate of Se-azulene according to the procedure given in Section II-D. Recrystallization from absolute ethanol gave a product with a melting point of 105.5-107.5° c. 480 560 LENGTH, 520 WAVE Mµ 600 640 680 720 760 Absorption spectrum of azulene prepared from Se-dehydrogenation of parthenene (hexane solution). 440 Figure 7. 400 0.1 w 0.2 X .... u z .,_ 0.3 0 u z 0 w 0:4 I.L I.L (.) .,_ 0.5 z w 0.6 01--------------------------------. 41 Analysis of picrate 0£ Se-azulene from partheniol: Calculated for c21 HzpN3 0x: oun : c, 59.01%; • H, 4.96J~; N, 9.83% c, 58.68%; H, 4.75%; N, 9.87% 2.978 mg. substance gave 6.404 mg. carbon dioxide and 1.264 mg. water 2.840 mg. substance gave o.254 ml. nitrogen at 23° c. and 745 nnn Hg 3. Preparation of the styphnate (2,4,6-trinitroresorcinol addition product) of Se-azulene from partheniol The styphnate of Se-azulene was prepared according to the general procedure used for the preparation of polynitro addition products. point of the compound was found to be 110-111° c. The melting Because the quantity of the styphnate obtained was small, it was not possible to carry out an analysis of the substance. C. Conclusion The azulene obtained from the selenium dehydrogenation of parthenene has not been identified; it is probably admixed with other azulenes. 42 Preliminary Qualitative Ozonization 2£_ Partheniol and Tetrahydroparthenene r. Ozonization of partheniol Five grams of partheniol was ozonized in glacial acetic acid. The ozonide was decomposed by hydrogenation in 1:1 acetic acid-ether with a platinium catalyst (Adams) at below o0 c. Because the decomposition of an ozonide with catalytic hydrogen is exothermic, the reaction mixture was cooled in an ice- ' salt bath during the first phase of the hydrogenation in order that a secondary reaction of "acid rearrangement" of the ozonide would not occur to any appreciable extent. The hydrogenation was later allowed to proceed at room ·temperature till the required amount of hydrogen was taken up. A test for formaldehyde in the gas wash bottle with dimedon reagent was positive, while a test for acetone with p-nitrophenylhydrazine was negative. It was thus shown that probably a methylene group was present in partheniol and, hence, one of the two double bonds in the compound is possibly exocyclic. From the ozonized product was separated a neutral fraction by making the solution alkaline and extracting with ether. An acid fraction was obtained by acidifying the alkaline solution and extracting with ether. A qualitative test for methyl ketone in the neutral fraction was positive; a qualitative test for aldehyde with Schiff 1 s reagent was also positive. By distilling at 0.6 mm. Hg and 98-100° c., 0.5 g. of the neutral fraction was collected. tillable. A greater portion of the neutral fraction was not dis- The amount of acid fraction was negligible. 43 A. Analysis of neutral fraction 1. Analysis for carbon and hydrogen C, 67.13%; H, 9.01% C, 66.11%; H, 8.72% C, 66.74%; H, 8.69% 2.117 mg. substance gave 5.177 mg. carbon dioxide and 1.644 mg. water 2. Determination of active hydrogen and carbonyl group (Zerewitinoff) 25.166 mg. substance gave 1.90 ml. methane at STP; 4.05 ml. methane missing (STP) 2.21 ml. methane at STP for 1 hydroxyl group; 4.46 ml. methane missing (STP) for 2 carbonyl groups 3. Determination of methyl ketone group 7.4 mg. substance used up 42.14 mg. iodine, equivalent to 1.90 or 2 CH3co- groups for c 1411z o (MW: 254.3) 2 4 4. Physical constants n d 24° D = 1.4762 24° = 1.062 40 ~' calculated for C14H2z04: Found: II. 66.20, for 1 hydroxyl gr.o~p and 3 carbonyl groups 67.5 Ozonization of tetrahydroparthenene Approximately 0.4 g. of tetrahydroparthenene was ozonized in carbon tetrachloride; the ozonide was decomposed by adding a small amount of water and heating to a boil in a water bath. A test for formaldehyde in the wash bottle with dimedon reagent was positive; a test for acetone with p-nitrophenylhydrazine 44 was negative. It is thus indicated that a methylene group was probably pres- ent in tetrahydroparthenene. Hence, it is possible that in partheniol a ter- tiary alcohol group and a methyl group are attached to the same carbon atom. 45 Quantitative Determination of Methylene Group~ Partheniol and in Tetrahydroparthenene by Ozonization In preliminary qualitative ozonization experiments which were described in the previous section, it was indicated that one of the double bonds in partheniol is possibly exocyclic. It was also shown that probably the ter- tiary hydroxyl group and a methyl group are attached to the same carbon atom. In order to determine whether these preliminary observations are justified quantitative oxonization experiments were carried out; these experiments are described below. I. Ozonization of partheniol A 1.5-g. sample of partheniol was ozonized in 25 ml. of glacial acetic acid at 17° c. for 40 minutes (03 cone., 1.24 g./hr.). The formaldehyde was collected in a vra.sh bottle containing water and the carbon dioxide liberated was absorbed in two ascarite tubes. At the completion of the ozonization, 50 ml. of water was introduced into the reaction flask without disconnecting the system. Decomposition of the ozonide was carried out by heating this aqueous solution at a bath temperature of 100° C. for 2 3/4 hours. A sus- pension of mercuric oxide in water was then added from a separatory funnel and the reaction mixture was heated in a boiling water bath for an additional hour. When the reaction was complete, an excess of buffered dimedon reagent was added to the water in the wash bottle and the mixture was allowed to stand overnight. The amount of' formaldehyde ,dimedon which precipi ta. ted out on standing was negligible. The quantity of' carbon dioxide which was absorbed 46 in the ascarite tubes was determined gravimetrically and was found to be 121 mg., expressed as formaldehyde. Theoretically, from 1.5 g. of partheniol, which is assumed to possess one exocyclic double bond, 205 mg. of formaldehyde should be formed. Experimentally, only 121 mg. of formaldehyde was obtained, thus accounting for 60 percent of an exocyclic double bond. II. Ozonization of tetrahydroparthenene A 0.85-g. sample of tetrahydroparthenene was ozonized in 25 ml. of glacial acetic acid for 14 minutes (o3 cone., 1.24 g./hr.). The formaldehyde and car- bon dioxide which were formed were determined gravimetrically. The amount of formaldehyde which was precipitated with dimedon was negligible; the quantity of carbon dioxide expresse~ as formaldehyde was 80 mg. Assuming the compound contains one exocyclic double bond, the theoretical amount of formaldehyde formed is 124 mg. Hence, 64 percent of the exocyclic double bond in tetra- hydroparthenene was accounted for. III. Conclusion It was again indicated by quantitative ozonization that one of the double bonds in partheniol is probably exocyclic and thus gives rise to a methylene group. It is not known at this stage whether the methylene group is attached to the nucleus of the molecule as a separate unit, or is part of an isopropylene side cha.in. As shown in the quantitative as well as the qualitative experiments, the tertiary hydroxyl group in partheniol is probably attached to the same carbon atom as a methyl group, thus accounting for the methylene group in tetrahydroparthenene. 47 Determination of the Position of the Hydroxyl Group !!!_ Partheniol A scheme of reactions which was used for the determination of the position of the hydroxyl group in partheniol is essentially as follows. Partheniol is first saturated with hydrogen to fo:nn tetrahydropartheniol which is subsequently dehydrated into tetrahydroparthenene. As the hydroxyl group is pro- bably on the same carbon atom as a methyl group (work described in previous section) there are three possible positions at which the double bond may be formed in the tetrahydro compound. It was thought that the double bond may be predominantly in a given position; in this case, by cleavage of the molecule at the double bond by means of ozonization, the major products of the cleavage may be obtained for study. By characterizing these products, it is possible to locate the hydroxyl group in partheniol. In another series of experiments. designed to determine the position of the hydroxyl group in partheniol, a chromic acid oxidation procedure was used. As in the ozonization experiments, characterization of the products of oxidation was desired in order to place the hydroxyl group. I. Preparation of tetrahydropartheniol In a typical experiment 32.8 g. of partheniol was hydrogenated in approximately 200 ml. of glacial acetic acid with 0.4 g. of platinium dioxide as catalyst. The reaction was carried out at room temperature and at a pressure slightly above atmospheric. When no further uptake of hydrogen was observed, the reaction was assumed complete. After the removal o:f the catalyst by fil- tration, the glacial acetic acid solution was concentrated under vacuum. 48 Qualitative tests indicated that esterification of partheniol had occurred during the hydrogenation. To hydrolyze the parthenyl acetate, approximately 100 ml. of a 0.5 N alcoholic solution of' potassium hydroxide was added; the reaction mixture was heated for two hours in a water bath. After neutral- ization of the base, the alcohol was evapore.ted under vacuum. The crude tetrahydropa.rtheniol was extracted with ether and the ether solution was dried over anhydrous sodium sulfate. The ether having been removed, the hydrogenated product was distilled under high vacuum; the main fraction of tetrahydropartheniol, 28.1 g., was distilled at 70-85° c. and 0.01 mm. Hg. A. Analysis of tetrahydropa.rtheniol 1. Analysis for carbon and hydrogen Calc-ulated for C15H28 o: Found: C, 80.29%; H, 12.58% C, 80.16%; H, 12.72% 2.424 mg. substance gave 7.209 mg. carbon dioxide and 2.707 mg. water 2. Physical constants n 25.5° D d25.4 40 = 1.4820 0 = o.9342 MR, calculated= 68.60, for 1 hydroxyl group Found = 68.47 II. Pre:_i;aration of tetrahydroparthenene A mixture of approximately 38.5 g. of tetrahydropartheniol and 24 g. of fused powdered potassium bisulfate was heated for 45 minutes at a bath temperature of a.bout 160° c. After allowing the reaction mixture to cool, the 49 crude dehydrated product was extracted with ether. The ether solution was washed, first, with a very dilute solution of sodium hydroxide, then, with water till the solution was free from alkali. After drying over sodium sul- fate, the ether solution was filtered and the ether removed. Distillation of the crude hydrocarbon was carried out under high vacuum. The hydrocarbon was distilled three times, twice over potassium. The main fraction which was obtained from a final distillation over potassium was distilled at 5964° c. and a pressure of 0.02 mm. Hg; a yield of 23 g. was obtained. A. Analysis of tetrahydroparthenene 1. Analysis for carbon and hydrogen Calculated for c15H26 : Found: C, 87.30%; H, 12.70% C, 87.31%; H, 12.83% 2.303 mg. substance gave 7.368 mg. carbon dioxide and 2.640 mg. water 2. Physical constants 25° n D d25 40 - 1.4880 0 = o.9025 calculated: 66.6, for one double bond Found: 65.9 3. Hydrogenation 25.504 mg. substance and 6.799 mg. platinium dioxide in glacial acetic acid took up 2.77 ml. hydrogen at STP Calculated for c15H2 6 0: Found: 2.77 ml. hydrogen at STP for 1 double bond 2.77 ml. hydrogen at STP; 1 double bond 50 III. A• Ozonization of tetrahydroparthenene (14) An 18.2-g. sample of tetrahydroparthenene in approximately 40 ml. of glacial acetic acid was ozonized at about 15° c. for 3 hours. To decompose the ozonide, 18 g. of zinc dust suspended in a small amount of water was slowly added; the reaction temperature was kept at 50-70° c. for approximately l/2 to l hour. A test for peroxides was made with titanic sulfate and dilute sulfuric acid; no peroxides were found. The excess zinc and zinc salts were removed by filtration on a sintered glass funnel. The acetic acid solution was neutralized with a slight excess of dilute potassium hydroxide and the neutral solution was extracted with several portions of ether. An acid frac- tion was obtained by acidifying the aqueous phase remaining from the ether extraction with dilute sulfuric acid and extracting with several portions of ether. The neutral and acid ether fractions were both thoroughly washed and dried over anhydrous sodium sulfate. After filtering and removing the ether, the neutral fraction was distilled at 0.125-0.150 rmn. Hg; the fractions collected were redistilled at 5-50 ?• 1. First fractionation of neutral fraction at 0.125-0.150 mm. Hg A B C D (a) 68-78° c. 78-93° c. 93-108° c. 108-120° c. 6.4 g. 1.4 g. 2.4 g. 2.8 g. Analysis of neutral fraction A for carbon and hydrogen C, 80.71%; H, 11.61% C, 83.83%; H, 12.51% 1. 724 mg. substance gave 5.296 mg. carbon dioxide and 1.927 mg. wate·r 51 (b) Analysis of neutral fraction B for carbon and hydrogen C, 80.71%; H, 11.61% C1 75.58%; H, 11.00% C, 77.56%; H, 11.36% 3.098 mg. substance gave 8.805 mg. carbon dioxide and 3.145 mg. water (c) Analysis of neutral fraction C for carbon and hydrogen Calculated for c16~ 6o2 : Found: C, 75.58%; H, 11.00% C, 75.17%; H, 11.37% 2.345 mg. substance gave 6.459 mg. carbon dioxide and 2.383 mg. water 2. Second fractionation of neutral fractions at 5-50 p. 1. 2. 3. 4. 5. 6. 51-61° c. 61-76° c. 76-96° c. 96-135° c. 135-140° c. residue 2.1 g. o.a g. 1.3 g. 1.8 g. 3.0 g. 0.1 g. (a) Analysis of neutral fraction l (i) Analysis for carbon and hydrogen Calculated for c14H24 o: 01sH25°2: Found: C, 80.71%; H, 11.61% C, 76.58%; H, 11.00% C, 83.48%; H, 12.99% 2.177 mg. substance gave 6.660 mg. carbon dioxide and 2.528 mg. water (ii) Active hydrogen and carbonyl group determination (Zerewitinoff) 22.766 mg. substance gave 0.40 ml. methane at STP; 0.77 ml. methane missing (STP) Calculated for c 14 H24 o (208.3): c15H26 o 2 (238.4): 2.45 ml. methane missing (STP) for l carbonyl group 4.28 ml. methane missing (STP) for 2 carbonyl groups 52 (iii) Methyl ketone group determination 21.7 mg. substance used up 20.96 mg. iodine; equivalent to 0.3 CH3 cognoup for c15 H26 o2 (MW= 238) and o.27 CH3CO- group for C14H240 (MW= 208) (b) Analysis of neutral fraction 2 (i) Analysis for carbon and hydrogen Calculated for C14Hz 40: Cl~2602: Found: C, 80.71%; H, 11.61% c, 75.58%; H, 11.00% C, 81.59%; H, 12.92% 1.947 mg. substance gave 6.821 mg. carbon dioxide and 2.247 mg. water (ii) Active hydrogen and carbonyl group determination (Zerewitinoff) 19.535 mg. substance gave 0.52 ml. methane at STP; 0.60 ml. methane missing (STP) Calculated for c H24 o (208.3): 14 c15H26 o2 (238.4): (iii) 2.10 ml. methane missing (STP) for 1 carbonyl group 3.67 ml. methane missing (STP) for 2 carbonyl groups Methyl ketone group determination 14.0 mg. substance used up 7.74 mg. iodine; equivalent to O.l CH3COg~oup £or ci 5H26 o2 (MW: 238) and 0.15 CH3CO- group for C14Hz40 (MVV : 208.3} (o) Analysis of neutral fraction 3 (i) Analysis for carbon and hydrogen C, 80.71%; H, 11.61% C, 75.58%; H, 11.00% C, 77.47%; H, 11.99% 2.098 mg. substance gave 5.956 mg. carbon dioxide and 2.248 mg. water 53 (ii) Active hydrogen and carbonyl group determination (Zerewitinorr) 19.794 mg. substance gave 1.25 ml. methane at STP; 1.88 ml. methane missing (STP) calculated for c14Hz 4 0 (208.3): 015H2s O2 (238.4): (iii) 2.13 ml. methane missing (STP) for l carbonyl group 1.87 ml. methane missing (STP) for l carbonyl group; 3.74 ml. methane missing (STP) for 2 carbonyl groups Methyl ketone group determination 9.5 mg. substance used up 12.11 mg. iodine; equivalent to 0.4 CH3 cogroup for C1 5H26 o2 (MW= 238) B. In another experiment, 23 g. of tetrahydroparthenene was ozonized in 60 ml. of glacial acetic acid. The solution was tested for unsaturation with a 5-percent solution of bromine in carbon tetrachloride; the test was negative indicating that the reaction was complete. For the reductive decomposition of the ozonide, the zinc-acetic acid method was used. The ozone reaction pro- ducts were separated into a neutral and an acid fraction. From the distilla- tion of the acid fraction at 145-165° c. and 75 p pressure a 2.3-g. fraction was obtained. Fractions from distillation of neutral fraction at lO-2Op: 1. 2. 3. 4. 60-80° c. 80-105° c. 105-120° c. residue 6.0 g. 3.5 g. 2.0 g. 2.6 g. Both neutral and acid fractions showed a strong positive qualitative test for the aceto or methyl ketone group. Investigation of the contents of the water wash bottle showed that only 54 formaldehyde was present. The formaldehyde was identified by means of its dimedon derivative and its 2,4-dinitrophenylhydrazone derivative. 1. Carbonyl derivatives of the neutral fractions Attempts to prepare a semioarbazone from the neutral fractions by means of an alcoholic acid-buffered solution of semicarbazide have been unsuccessful thus far. The preparation of a 2,4-dinitrophenylhydrazone also proved unsuccessful. 2. Absorption spectra of neutral ozone fractions The spectra of the three neutral ozone fractions in ethanolic solutions were determined with the Beckman photoelectric spectrophotometer. All three samples showed maxima in the region 230-236 r,t, as shown below. Neutral fractions 1. 2. 3. 3. 60-80° c. 80-105° c. 105-120° C. Position of maxima wave length, rrp236 • 5 235 230 Hydrogenation of the carbonyl compound (neutral fraction) Neutral fractions 2 and 3 were combined with a neutral ozone fraction from a previous ozonization. This combined fraction (10.3 g.) was hydrogenated in 100 ml. of absolute alcohol at room temperature under a pressure of 2.7 atmospheres. When no further drop in pressure was observed, the reaction was stopped. The catalyst was filtered out and the alcohol removed under reduced pressure. The remaining residue was then distilled under high vacuum; the three fractions 55 collected are listed below. Distillation at 60 f: 88-95° c. 95-115° c. 115-135° c. 1. 2. 3. 0.6 g. 2.0 g. 4.6 g. Fractions 2 and 3 were combined and taken as the principal alcohol fraction; this combined fraction was dehydrated to form a hydrocarbon which was subsequently dehydrogenated with sulfur. The procedures for these reactions are given in the following sections. (a) Dehydration of the alcohol A mixture of 6.5 g. of the alcohol (fractions 2 and 3 combined) and 4 g. of fused potassium acid sulfate was heated at a bath temperature of 200° c. for 1/2 hour. The reaction mixture was extracted with ether and the ether solution was dried over anhydrous sodium sulfate. The oil remaining after the removal of the solvent was fractionally distilled; the fractions collected are given below. Distillation at 75?: 1. 2. 90-115° c. 115-140° c. 2.4 g. 0.9 g. Analysis of fraction 2: (1) Analysis for carbon and hydrogen Calculated for c15H30o2 : Found: c, 74.32%; H, 12.48% c, 73.99%; H, 12.44% 2.677 mg. substance gave 7.258 mg. carbon dioxide and 2.977 mg. water 56 (2) Active hydrogen and carbonyl group determination 19.712 mg. substance gave 2.36 ml. methane at STP; 0.95 ml. methane missing (STP) 3.64 ml. methane missing (STP) for 2 carbonyl groups; 1.82 ml. methane missing (STP) for 1 carbonyl group (b) Dehydrogenation with su lf'ur A mixture of 2.4 g. of fraction land 1.5 g. of sulfur was heated for 5 hours at a bath temperature of 230° c. The crude product was transferred into a micro claisen flask and distilled under vacuum. The distilled product was dissolved in petroleum ether and chromatographed on an activated alumina column. 4. No trace of azulene was detected on the column. Acid ozone fraction Distillation of the acid ozone fraction was carried out at 75p. and 145165° c.; 2.3 g. was obtained. The acid fraction gave a positive qualitative test for methyl ketone, thus indicating that it is probably a keto acid. This fraction was combined with an acid fraction from a previous ozonization giving a total of 3.2 g. (a) Degradation of the keto acid with potassium hypobromite (Haloform reaction) A reagent solution containing 29 g. of potassium hydroxide, 6.8 nµ.. of 0 bromine, and 170 ml. of water was prepared at o c. The keto acid (3.2 g.) was added to the potassium hypobromite reagent and the mixture was allowed to come to room temperature. The reaction mixture was agitated for two hours on a shaking machine after which the excess potassium hypobromite was decomposed 57 with sodium bisulfite; the bromoform was removed by extraction with ether. The aqueous solution was then made acid and extracted with ether. The ether solution was thoroughly washed with water, dried over anhydrous sodium sulfate, and the solvent was distilled. The residue was combined with a crude acid fraction (less than 2 g.) which was obtained from a haloform reaction on a mixture of 6 g. of a neutral ozone fraction and 2.6 g. of a similar neutral fraction from a chromic-acid oxidation; qualitative tests showed that each of the latter fractions contained methyl ketone and aldehyde. Distilla- tion of the combined acid fraction was carried out as shown below. Distillation at 50 f: 3. 100-140° c. 140-155° c. 155-170° c. (i) Analysis of fraction l 1. 2. 0.9 g. 0.3 g. 2.2 g. Hz Calculated for c14 4 0 3 : Found: C, 69.96%; H, 10.07% C, 67.66%; H, 9.94%; Br, 6.13%; equivalent weight, 228 2.401 mg. substance gave 5.953 mg. carbon dioxide and 2.132 mg. water 6.379 mg. substance gave 0.919 mg. silver bromide 10.275 mg. substance r .e quired 4.49 ml. 0.01 N sodium hydroxide (ii) Analysis of fraction 3 C, 50.16%; H, 6.92%; Br, 23.84% C, 49.59%; H, 7.47%; Br, 29.53%; equivalent weight, 192; saponification equivalent, 129 2.565 mg. substance gave 4.661 mg. carbon dioxide and 1.712 mg. water 6.945 mg. substance gave 4.819 mg. silver bromide 12.569 mg. substance required 6.54 ml. 0.01 N sodium hydroxide 84.l mg. substance saponified 35 minutes at 125° c. with potassium hydroxide in diethylene glycol required 6.50 ml. O.l N hydrochloric acid 58 (b) Reaction of dicarboxylic acid (acid fraction 3) with acetic anhydride (Blanc's reaction (15)) The bromodicarboxylic acid (acid fraction 3), about 400 mg., was debrominated with 18 ml. of a 1 N alcoholic solution of potassium hydroxide. The alkali was neutralized with dilute sulfuric acid and the alcohol was removed -under reduced pressure. After acidifying with dilute sulfuric acid the res- idue was extracted with ether; the ether extract was washed with three small portions of water, dried over anhydrous sodium sulfate and the ether was removed under reduced pressure. An attempt to hydrogenate the residual liquid in glacial acetic acid with platinium dioxide as catalyst resulted in the uptake of only 2 ml. of hydrogen. The dicarboxylic acid was recovered by fil- tering out the catalyst and removing the glacial acetic acid under vacuum. About l ml. of acetic anhydride was then added to the recovered acid and the mixture was refluxed for an hour. The mixture was transferred to a micro claisen flask and the acetic acid and acetic anhydride were removed under the reduced pressure of an aspirator. One milliliter of absolute ether was used to wash out the acetic anhydride in the side arm of the micro claisen flask; the ether was subsequently removed under vacuum. The remaining oil was distilled at a bath temperature of 240° c. and a pressure of 0.5 mm. Hg. A drop of oil was obtained and was analyzed. Analysis: Calculated for c13 H o (cyclic ketone): 20 Found: C, 81.20%; H, 10.48% C, 54.18%; H, 8.03% 2.847 mg. substance gave 5.653 mg. carbon dioxide and 2.044 mg. water 59 c. Since the results obtained from the Blanc reaction described above were not conclusive the experiment ~~s repeated. A 10.5-g. sample of tetra- hydroparthenene was ozonized in 20 ml. of glacial acetic acid for 1 1/2 hour (1.45 g. o3/ hr.). Since the acid fraction was desired the ozonide was decom- posed with a 7-percent solution of hydrogen peroxide by heating in a water bath for an hour. Decomposition of the excess hydrogen peroxide was carried out by heating for an additional 1/2 hour in the presence of traces of platinium black. acid. The mixture was concentrated in order to remove the excess A neutral and an acid fraction were obtained by making the concen- trate basic and acid respectively and extracting with ether. On testing the neutral fraction with Schiff's reagent £or aldehyde a positive reaction was obtained; the oxidation was apparently incomplete. The neutral fraction was thus further reacted with a 5-percent solution of hydrogen peroxide in glacial acetic acid for an hour in a hot water bath with vigorous stirring. The ex- cess peroxide was again decomposed with traces of platinium black and the mixture was concentrated and separated into a neutral and an acid fraction. The total acid fraction obtained was 2.5 g.; the neutral fraction, 4.5 g. Distillation of the acid fraction was carried out e.s shown below. Distillation at 75 p.: 176-190° c. 190-210° c. 0.65 g. 0.5 g. 60 Analyses for carbon and hydrogen: Fraction A: Calculated for Cl5¾603: C14H240&: Foun : c, 7o.sa%; H, 10.30% c, 65.60%; H, 9.44% c, 68.36%; H, 9.96% 2.098 mg. - substance gave 6.256 mg. carbon dioxide and 1.866 mg. water Fraction B: Calculated for c14 ~ 4 o4 : Found: C, 65.60%; H, 9.44% C, 66.51%; H, 9.70% 2.074 mg. substance gave 5.055 mg. carbon dioxide and 1.798 mg. water l. Reaction with sodium hypobromite (Haloform reaction) Fractions A and B were reacted separately with sodium hypobromite (1.7 g. NaOH, 2 g. Br 2 , and 23 ml. water) for 2 hours. The excess sodium hypobrom- ite was decomposed with a small amount of sodium bisulfite. After removing the neutral fraction containing bromoform the alkaline solution was acidified with sulfuric acid and an acid fraction was extracted with ether. The acid fractions obtained from original fractions A and B were distilled separately. Distillation of fraction A at 10 f and a bath tempera- ture of 190-200° c. gave 0.3 g. of oil; the distillation of fraction Bat the same pressure and at a bath temperature of 200-230° c. gave 0.15 g. of oil. Analysis of distillate from fraction A: Calculated for c14 H o4 : 24 Found: C, 65.60%; H, 9.44% C, 64.34%; H, 9.52% 2.181 mg. substance gave 5.143 mg. carbon dioxide a.nd 1.855 mg. water 61 2. Reaction of dicarboxylic acid with acetic anhydride A mixture of 0.3 g. of dicarboxylic acid from fraction A and 0.6-0.7 ml. of acetic anhydride was heated in a sealed Carius tube for 1/2 hour at 240° c. (bath temperature). The cooled tube was opened and its contents was trans- ferred into a micro-claisen flask with ether. The acetic anhydride was re- moved and the ren~ining oil was distilled under high vacuum. A fraction of approximately 50 mg. which distilled at 10 f pressure and a bath temperature of 200-210° c. was obtained. Analysis for carbon and hydrogen: Calculated for c14 I½_ 4 o4 (acid): c14 H22 o3 \anfiydride): c13 r1z 0 o (ketone): Found: C, 65.60%; H, 9.44% C, 70.55%; H, 9.30% C, 81.20%; H, l0.48J/, C, 68.17%; H, 9.17% 2.147 mg. substance gave 5.363 mg. carbon dioxide and 1.759 mg. water 3. Cone lus ion The analysis of the product obtained from Blanc's reaction on the dicarbe.nltc acid shows that it is probably a mixture of the original its anhydride. acid and Since a ketone was not formed from the dicarboxylic acid, it is indicated that the hydroxyl group in partheniol is situated in the 5-membered ring. IV, oxidation of tetrahydropartheniol with chromic acid (16) A 22.4-g. sample of tetrahydropartheniol in 112 ml. of glacial acetic acid was treated drop~~se with a solution of 19.5 g. of chromic trioxide in a mixture of 112 ml. of water and 240 ml. of glacial acetic acid. The addition of the oxidizing agent was completed in an hour; the oxidation, in 62 three hours. The volatile products were steam distilled; the distillate was extracted with ether and the ether solution was thoroughly shaken with an aqueous solution of ,sodium carbonate • .After this treatment the ether solution was washed with water till neutral, dried over anhydrous sodium sulfate, filtered and the ether was removed. tained. A crude neutral fraction of 14.8 g. was ob- This fraction was distilled twice; the final distillation at 20 f (0.02 mm. Hg) gave the following fractions. Distillation at 20 f; 1. 2. 3. 57-70° c. 70-85° c. 85-95° c. 2.6 g. 3.8 g. 3.3 g. Analysis of fraction 2: Found: C, 82.86%; H, 13.34% 2.001 mg. substance gave 6.080 mg. carbon dioxide and 2.385 mg. water Qualitative tests for the methyl ketone or aceto- (CH3 co-) group showed that fraction 1 was strongly positive while fractions 2 and 3 were weakly positive. A qualitative test for the carbonyl group with azobenzenephenyl- hydrazine sulfonic acid showed fraction 1 to be strongly positive; fractions 2 and 3 were also positive. A. Absorption spectra of the neutral fractions The absorption spectra of the neutral fractions were recorded with a Beckman photoelectric spectrophotometer, using absolute alcohol as the solvent. The nature of the absorption spectra of the three neutral fractions are as shown. 63 Neutral fractions 1. 2. 3. B. Position of maxima wave length, up 57-70° c. 70-85° c. 85-95° c. no maxima 293 286 Carbonyl derivatives of the neutral fractions Attempts to prepare a semicarbazone from the neutral fractions by means of an alcoholic, acid-buffered solution of semicarbazide or an alcoholicpyridine solution of semicarbazide have proved unsuccessful thus far. The preparation of a 2,4-dinitrophenylhydrazone has likewise proved unsuccessful. c. Hydrogenation of the neutral fraction Approximately 6.6 g. of a neutral fraction (ketone fractions 2 and 3 combined) was hydrogenated in 60 ml. of absolute alcohol under 2.4 atmospheres of pressure with 6 g. of Raney nickel as catalyst. moles of hydrogen was taken up. A total of 0.15 The catalyst was filtered out and the fil- trate containing the hydrogenated product was concentrated. The alcohol ob- tained from the hydrogenation was fractionally distilled under high vacuum. Distillation at 100 p (0.1 mm. Hg): 1. 2. 63-70° c. 70-80° c. 0.9 g. 3.55 g. 1. Analysis of fraction 2 (a) Analysis for carbon and hydrogen Calculated for c14 H26 o: Found: C, 79.93%; H, 12.46% C, 79.89%; H, 12.12% 2.018 mg. substance gave 5.908 mg. carbon dioxide and 2.186 mg. water 64 (b) Optical rotation r.-J',2 L"""D 2. 6 0 rd -- 15.7° (Cone. -- 1 • 02 ;o, • in a b so 1u t e a l co ho 1) Dehydration of the alcohol with potassium acid sulfate A mixture of 4.3 g. of combined fractions land 2 and 2.9 g. of potassium acid sulfate was heated for 35 minutes at 200-210° C. The hydrocarbon which was formed was directly distilled from the reaction flask at 80-95° c. and a pressure of 0.5 mm. Hg; 3.5 g. of light blue oil was obtained. 3. Dehydrogenation of the hydrocarbon A rrdxture of the hydrocarbon (3.5 g.) and 2.4 g. of sulfur was heated for 4 hours at 23 0° c. After the evolution of hydrogen sulfide had ceased, the reaction product was directly distilled at 90-100° c. and 50 f (0.05 mm. Hg); a 1.5-g. fraction was collected. A 1.1-g. sample of this substance was further dehydrogenated by heating with 0.3 g. of selenium at 300° C. for 5 hours. The crude product was trans- ferred into a claisen flask and distilled; 0.6 g. of slightly blue distillate was obtained. No addition product with trinitrobenzene could be prepared from this fraction; the amount of azulene present was evidently not more than a trace. In order to isolate this small quantity of azulene, the entire fraction was dissolved in petrol ether and chromatographed on a column of activated alumina. A diffuse zone of azulene, which was formed on the column was eluted with a petrol ether-benzene mixture. was repeated three times. The chromatographic procedure The solvent mixture was distilled from the final eluate and the azulene recovered was dissolved in purified hexane. The 65 absorption spectrum of the purified azulene in hexane, as recorded with a Beckman photoelectric spectrophotometer,is characterized by a principal maxinmm at 580 11'9-1 with three other maxima at 540, 635, and 672 llJl, as shovm in Figure 8. The azulene has not been identified. 0 .15 r z w 0.10 - (.) u.. u.. w 0 (.) z 0 r (.) z r- 0 .05 X w 0.00L----4-'8-0_ __5_._0_0_ __55....1....0 _ __ 6...._ 0_ 0 _ _6_,J5L.....0---7-'-0-0_ _ _ 7....L.50--~ WAVE Figure 8. LEN GT H, Mµ Absorption spectrum of azulene prepared from neutral fraction from chromic acid oxidation of tetrahydropartheniol (hexane solution). 66 Determination of the Positions of the Double Bonds I. Partial ozonization of partheniol in petroleum ether Twenty grams of partheniol was ozonized in 5-g. portions in purified petroleum ether at about 10° c. The ozonization time required for the cleavage of one double bond was determined by calculations. During the ozonization, a portion of the ozonide precipitated out as a gelatinous mass. At the com- pletion of the ozonization, the solvent was recovered and used for the ozonization of the next 5-g. sample. The ozonide was decomposed immediately after ozonization with water, zinc dust and traces of silver nitrate and hydroquinone. Decomposition at the boiling point of water was carried out for 2.5-3.0 hours; the reaction mixture was allowed to stand overnight ' at room temperature. The ozone reaction pro- ducts were separated into a neutral and an acid fraction. By continuous ex- traction an additional 2 g. of neutral fraction was recovered from the mother liquor. The neutral fractions gave positive methyl ketone and aldehyde tests. Distillation of neutral fraction at 10 f: 98-140° c. 140-160° c. Refractionation of neutral fraction l at 5 f: A• B. c. 85-92° c. 92-110° c. 0 ll0-130 c. 0.6 g. 0.9 g. 2.75 g. Distillation of recovered neutral fraction at 40-60 )1: 75-95° c. 95-120° c. 1.4 g. 0.4 g. 67 A. Analysis of neutral fractions 1. Fraction A (a) Analysis for carbon a.nd hydrogen Calculated for c15H24 o3 : Found: C, 71.39%; H, 9.59% C, 72.68%; H, 9.3Q% 1.951 mg. substance gave 5.189 mg. carbon dioxide and 1.622 mg. water (b) Active hydrogen a.nd carbonyl group determination (Zerewitinoff) 11.047 mg. substance gave 1.18 ml. methane at STP; 1.98 ml. methane missing (STP) 0.98 ml. methane at STP for 1 hydroxyl group; 1.96 ml. methane missing (STP) for 2 carbonyl groups (c) Methyl ketone group determination 11.l mg. substance used up 37.82 mg. iodine; equivalent to 1.1 CH3COgroups for c15H24 o3 (MW: 252) (d) Physical constants 25° n - 1.4962 25° = 1.028 D d 40 ~, calculated: Found: 70.35 (c 15H24 o3 ), for 1 hydroxyl group, 1 double bond, and 2 carbonyl groups 71.78 2. Fraction B (a) Analysis for carbon and hydrogen Calculated for C15H2403: C14H2202: Found: C, 71.39%; H, 9.59% C, 75.63%; H, 9.97% C, 73.79%; H, 9.70% 2.119 mg. substance gave 5.730 mg. carbon dioxide and 1.836 mg. water 68 (b} Active hydrogen and oarbonyl group determination (Zerewitinoff) 12.440 mg. substance gave 1.27 ml. methane at STP; 2.45 ml. methane missing (STP) 1.1 ml. methane at STP for l hydroxyl group; 2.5 ml. methane missing (STP) for 2 carbonyl groups ( c) Methyl ketone group determ.inat ion 26 mg. substance used up 110.3 mg. iodine; equivalent to l.4 CH3 cogroupd'or c15H24 o3 (MW: 252) 12.4 mg. substance used up 50.89 mg. iodine; equivalent to 1.36 CH3cogroups for c15H24 o3 (MW= 252) (d) Physical constants n25° D d 24° 40 -- 1.4992 = 1.031 MR, calculated: Found: 70.35 (c 16H2 4 o3 ), for 1 hydroxyl group, 1 double bond, and 2 carbonyl groups 71.77 3. Fraction C (a) Analysis for carbon and hydrogen Calculated for c15H240 3 : Found: C, 71.39%; H, 9.59% C, 72.23%; H, 9.62% 2.443 mg. substance gave 6.466 mg. carbon dioxide and 2.101 mg. water (b) Active hydrogen and carbonyl group determinati~n (Zerewitinoff) 10.913 mg. substance gave 1.21 ml. methane at STP; 2.30 ml. methane missing (STP) 0.97 ml. methane at STP for l hydroxyl group; 1.94 ml. methane missing (STP) for 2 carbonyl groups 69 (c) Hydrogenation 23.927 mg. substance took up 7.94 ml. hydrogen at STP Found: (d) 6.38 ml. hydrogen at STP for 2 carbonyl groups and 1 double bond, 8.50 ml. if tertiary gydroxyl group was reduced F3.74 : F4 Physical constants n 25° D d 24° 40 =1.4960 -- 1.033 ~, calculated: Found: 70.35 (c 15H24 o3 ), for l hydroxyl group, 1 double bond, and 2 carbonyl groups 71.43 4. Fraction a (a) Analysis f' or carbon and hydrogen Calculated for C15H240 3 : Found: C, 71.39%; H, 9.59% C, 71.61%; H, 8.88% 2.053 mg. substance gave 5.387 mg. carbon dioxide and 1.629 mg. water (b) Methyl ketone determination 14.0 mg. substance used up 66.25 mg. iodine; equivalent to 1.6 CH3COgroups for c15H24 o3 (MW= 252) 5. Fraction b Calculated for C15H2403: Found: C, 71.39%; H, 9.59% C, 68.50%; H, 9.27% 2.206 mg. substance gave 5.538 mg. carbon dioxide and 1.827 mg. water 70 B. Degradation of methyl ketone group in fraction C with potassium hypobromite Fraction C was reacted with a solution of potassium hypobromite in the usual manner. From the reaction 1.3 g. of crude acid was obtained. An attempt to distill the hydroxyl acid under high vacuum was unsuccessful because the compound decomposed during the distillation. 71 Discussion Partheniol has been found to possess two double bonds .by, .. hydrogenation experiments in glacial acetic acid with platinium dioxide as catalyst. The hydrogenation of dihydropartheniol in methylcyclohexane-acetic acid showed, however, an uptake of two moles of hydrogen. reduction of the hydroxyl group. This is probably caused by the Such a reaction has been reported in exper- iments on guaiol (17,18); it was fourfd that on hydrogenating guaiol in cyclohexane with platinium dioxide one mole of hydrogen was taken up, while on hydrogenating in glacial acetic acid with platinium black as catalyst two moles of hydrogen was consumed. Parthenene which contains three double bonds did not show any maxima or minima in its absorption spectrum, as determined with the Beclonan photoelectric spectrophotometer; it is thus indicated that the compound does not possess any conjugated double bond. The determination of molecular refractirity showed, however, an exaltation of 0.37. It is possible that the exaltation is due to a semicyclic double bond; such small exaltations of semicyclic double bonds have been reported for f -pinene and for camphene which give an exaltation of 0.2 and 0.5, respectively. Azulenes have been formed from parthanene by selenium and sulfur dehydrogenation. tained. By selenium dehydrogenation an unknown violet azulene was ob- The melting points of the polynitro derivatives of the Se-azulene are as follows: trinitrobenzene addition product, m.p. 103-104° c.; piorate, m.p. 105.5-107.5° c.; styphnate, m.p. 110-lll0 c. The absorption spectrum 72 of Se-azulene is characterized by three maxima at 553, 56_7 , 578 ~ and undulations at 490, 500, 533, 603, 616, 627, and 664 '1• Plattner (13), in a comprehensive study on azulenes, ·round that the absorption spectrum of an azulene is influenced by the introduction of alkyl substituents. Plattner further states that the position rather than the size of the alkyl subs·bituent is responsible for variations in the absorption of the azulene. By the comparision of the absorption bands of various violet azulenes, Plattner was able to deduce that Se-guaiazulene, a violet azulene, is probably substituted in the 2,4,7-positions. Since the violet azulenes in- vestigated by Plattner have been found to possess a 2-substituent, it may be speculated that, as the Se-azulene from partheniol is violet . in color, possibly a shift of the substituent from the 1-position to the 2-position had occurred during the process of dehydrogenation. It is also possible that in the Se-dehydrogenation of partheniol, a mixture of azulenes rather than a single azulene was formed. The azulene obtained from the sulfur dehydrogenation of parthenene has been characterized by various polynitro derivatives and by its absorption spectrum and has been identi.fied as S-guaiazulene. From this finding the conf'iguration of partheniol has been elucidated as 1,4-dimethyl-7-isopropylcyclopentacycloheptane (I) as in guaiol. With the conf'iguration known the positions of the hydroxyl group and the two double bonds remained to be determined. This problem has been approached by means of ozone and chromic acid oxidation experiments. From the detection of a methylene group in the preliminary ozonization of tetrahydroparthenene it was deduced that the hydroxyl group is probably 73 situated on the same carbon atom as a methyl group; in this case, the hydroxyl group is situated either in the 1- or 4-position in the partheniol con:figuration or in the isopropyl side chain as shown in diagrams III, IV, and v. III IV V Formula V may be eliminated, however, because no acetone was deteo~ed. Con- firmation of this deduction has been furnished by the subsequent quantitative determination of the methylene group in tetrahydroparthenene by an ozonization procedure in which 64 percent of an exooyclio double bond was accounted for. In the dehydration of tetrahydropartheniol the double bond which is formed could be situated in three possible positions; the three ppssible structures of tetrahydroparthenene are as presented in diagrams VI, VII, and VIII, if the hydroxyl group is originally in the 1-position and are as given in diagrams IX, X, and XI, if the hydroxyl group is originally in the 4-poaition. VI VII VIII It 74 IX X XI is plausible that 64 percent of a methylene group which was found indicates a predominance of the exocyclic double bond. It is also conceivable that the value 64 percent may be too low; in consideration of the experimental error in general quantitative ozonization procedures this fairly wide discrepancy is acceptable. In the ozonization experiments on tetrahydroparthenene formaldehyde and methyl ketone have been detected; it is thus substantiated that the hydroxyl group and a methyl group are attached to the same carbon atom. From a study of the analytical data it appears that one or the other of the two cyclic ketones, XII and XIII, depending on the position of the hydroxyl group, probably XII XIII 75 constitutes two fractions, land 2, which were obtained from the ozonization of tetrahydroparthenene (p. 51). The discrepancies in the analyses 0£ these fractions are possibly due to the presence of hydrocarbons. A third fraction from the ozonization experiment (fraction 3) probably consists of a mixture of carbonyl com.pounds XIV and X:V or XVI and XVII, depending whether the hydroxyl group in partheniol is in position 1 or 4. XIV xv X:VII The presence of active hydrogen in the neutral ozone fractions of tetrahydroparthenene indicates that the aldehyde group probably was partially reduced to an alcohol group during the course of the reduction of the ozonide with zinc and acetic acid. The spectra of the neutral ozone fractions in 76 alcohol showed maxima at 230-236.5 ryi,; the position of the :maxima is lower than that usually shown by a carbonyl group, which is about 280 np• It has been found in ozonization experiments on guaiol (19) that the dihydroxyl ketone, XVIII, showed a maximum at 281.5 np in an alcoholic solution, whereas, the corresponding dehydrated compound, XIX, which is an unsaturated ketone, gave a spectrum in alcohol with a maximum at 265 o/• XVIII It is possible that the XIX absorption spectra of the neutral ozone fractions showed maxima at lower wave lengths because of tautomeric equilibrium between keto and enol forms of the carbonyl compound {20). From neutral fractions which were from ozone and chromic acid oxidations and which were treated with potassium hypobromite several acid fractions showing the presence of bromine were obtained. The analysis of the first fraction gave 67.66 percent carbon, 9.94 percent hydrogen, and 6.13 percent bromine (p. 57); excluding the small amount of bromine, the data agree reasonably well with the empirical fort!D.lla, c14H24o3 (69.96% C, 10.07% H). Taking into con- sideration the finding that the tertiary hydroxyl group is probably on the same carbon atom as a methyl group, formulae XX and XX.I may be proposed for the 77 c14¾4o3 compound. For the third acid fraction (fraction 3, p. 57) the forllJllla of c14¾304 Br, XXII and XXIII, may be proposed. cool-{ xx y~· e>r cool-I Br XX.II "' XX.III Ozonization and oxidation products have not been able to elucidate whether the hydroxyl group in pa rtheniol is in the 1- or 4-posi tion. To deter- mine the exact position of the hydroxyl group Blanc's reaction was carried out on the dicarboxylic acid which was obtained from the oxidation of tetrahydropa.rthenene; the structure of the acid is either XXIV or XXV, depending on the position of the hydroxyl group. Blanc 1 s rule states that when dicar- boxylic acids are heated with acetic anhydride and the resulting product slowly distilled, ketones are obtained from 1,6- or 1,7-dicarboxylic acids 78 XXIV XXV while anhydrides are formed from 1,4- and 1,5-dicarboxylic acids. Accordingly, if' the two carboxylic groups in the dioarboxylio acid from tetrahydroparthenene were situated in the 7-membered ring a ketone would be obtained; if' the two oarboxylic groups were situated in the 5-membered ring an anhydride would result. Apparently, no ketone was obtained from the reaction of the dicarbox- ylio acid with acetic apJlydride; this indicates that the carboxylic groups in the dicarboxylio acid are not situated in the 7-membered ring but are situated in the 5-membered ring as in structure XXIV. Hence, it probable that the structure of tetrahydropartheniol is as shown in diagram III, with the hydroxyl group at the 1-position. 79 The presence of a methylene group has been detected in the ozonization experiments on partheniol, thus indicating that one of the two double bonds is exocyclic. As the hydroxyl group is probably in the 1-position, the po- sition of the exocyclic double bond is restricted to two positions as shown in diagrams XXVI and XXVII. XX.VI XXVII Further indication of an exocyclic double bond was shovm in the partial hydrogenation of partheniol in alcohol with Raney nickel catalyst. The rela- tive rapid uptake of one mole of hydrogen is indicative of an exocyclic double bond. In the partial ozonization of partheniol in petroleum ether which was carried out for the purpose of the selective cleavage of the exocyclic double bond, it vms found that the neutral fraction is probably a mixture of C15H24 o3 and c14n22 o2 • the forrrula The structure of C1sH240 3 is probably as given in diagram XXVIII; c14 n22 o2 may be represented by structure XXIX. a rrdxture of products, namely, c15 The formation of nz 4 o3 and c14nz 2 ~, from the partial ozon- ization experiment in petroleum ether shows that the degradation products resulting from the cleavage of both the exocyclic and cyclic double bonds were obtained. 80 0 II LI C-ri \ XXIX XXVIII A neutral fraction which was obtained from the complete oxidation of partheniol with ozone has been found to contain 66.74 percent carbon, 8.69 percent hydrogen, one hydroxyl group, two carbonyl groups, and two methyl c15¾4 o4 and C14H2 2o4 , agree reasonably well with the carbon and hydrogen values. The formula, c16H24 o4 , cannot ketone groups. Two empirical formulae, represent, however, a possible ozonized product of partheniol because for two cyclio double bonds and one hydroxyl group the compound should contain five The formula, c14H22 o , which indicates one exocyclic double 4 bond, is acceptable and is structurally as shovm in XXX and :XXXI, depending atoms of oxygen. on the position of the exocyclic double bond. l O ij XXX 0 c// H 1~ These structures are in 0 0 XXXI 81 accordance with the Zerewitinoff determination which showed the presence of at least two carbonyl groups. Formula XXXI is discarded since it does not con- tain two methyl ketone groups and forrrllla XXXII is not considered because an aldehyde test with S.c hiff''s reagent was positive. The remaining formu.la, XXX, indicates that the structure of partheniol is as represented in XXXIII. XXXII XXXIII. Structures XXXIV, XXXV, XXXVI, XXXVII, and XXXVIII may be eliminated because no conjugated double bond system is present in parthenene. The absence of a conjugated double bond in parthenene has been indicated by spectral studies. XXXIV XXXV XXXVI 82 XXXVII XXXVIII This is also shovm by the difficulty in the dehydration of partheniol. Since it has been shown by qualitative tests that the two double bonds in partheniol are not conjugated fornmlae XXXIX and XL may also be eliminated. Structure XLI is not considered because the oxidized product of such a compound would not give two methyl ketone groups which have been found in the degradation XXXIX XL 83 products from the ozonization of partheniol. The remaining possible structure, XXXIII, is in agreement with the oxidized product XXX and hence may be proposed as the probable structure for partheniol. 84 Summary 1. Partheniol, a sesquiterpene alcohol (m.p. 127-128° c.), has been isolated from guayule. 2. Partheniol has been found to possess two double bonds, one of whioh is exocyclic, 3. An unlrno~m azulene or a mixture of azulenes has been obtained from the selenium dehydrogenation of parthenene. 4. S-g;uaiazulene has been obtained from the sulfur dehydrogenation of parthenene; hence, the skeletal st ructure of partheniol is as follows. 5. The hydroxyl group in partheniol has been found to be tertiary and is probably situated at the 1-position. 6. The proposed structure of partheniol is presented below. Bibliography (1) RuzickR, 1., Meyer, J., Helv. Chim. Acta!, 506 (1921) (2) Ruzicka, L., Meyer, J . , Mingazzini, M., Helv. Chim. Acta.!!, 346 (1922) (3) Emde, H., Pharm. Acta Helv. 4, 121-8 (1929) (4) Hall, J.A., Relationship in Phytochemistry, Chem. Rev. 20, 305 (1937) (5) Alexander, P., Ber.!!, 2320-28 (1911) (6) Walter, E.D., J. Am. Chem. Soc.~, 419-21 (1944) (7) Hl'ickel, w., Gercke, A., Gross, A., Ber.~, 563 (1933) {8) Buckel, w., Schnitzpahn, L., Ann ~ , 274 (1933) (9) Pfau, A.s., Plattner, Pl. A., Helv. Chim. Acta 19, 858 (1936) (10) Pfau, A.S., Plattner, Pl. A., Helv. Chim. Acta 20, 224 (1937) (11) Ruzicka, L., Haagen-s~1t, A.J., Helv. Chim. Acta 14, 1104 (1931) (12) Melville, J., J. Am. Chem. Soc., 55, 2462 (1933) (13) Plattner, Pl. A., Helv. Chim. Acta~, 283-94E (1941) (14) Plattner, Pl. A., Lemay, 1., Helv. Chim. Acta 23, 897 (1940) (15) Blanc, H.G., Compt. Rend. 144, 1356 (1907) Gilman, Org. Chem., 2nd ed., Vol. II, P• 1358 (John Wiley & Sons, Inc., 1944) (16) Plattner, Pl. A., Magyar, G., Helv. Chim. Acta 25, 581 (1942) (17} Sennnler, F.W., Risse, F., Ber.~, 2305 (1913) (18) Ruzicka, L., Haagen-Smit, A.J., Helv. Chim. Acta_!!, 1131 (1931) (19) Plattner, Pl. A., Magyar, G., Helv. Chim. Acta 24, 191 (1941) (20) Brode, W.R., Chemical Spectroscopy, 2nd ed., P• 221-222 (John Wiley & Sons, Inc., 1943)