PART I _ STUDIES OF THE SODIUM TUNGSTATE- HYDROGEN PEROXIDE OXIDATION OF 2-SUBSTITUTED-l,3,5,6,7- ' PENTACHLOROBICYCLOB 2. 01 AHEPTA- 2.6- DIEM 4- oNEs “ ,. PART II . 4 , . THE REACTION OF , . '- TETRACHLOROCYCLOPROPENE WITH POLYCHLOROETHYLENES VIhesi‘s for the Degree of Ph. D. , - ‘ MICHIGAN STATE UNIVERSITY ‘ CHARLES WILLIAM BAUER I I 1969 ......................... LIBRARY Michigan State I [Inivenfiqy THESIS I This is to certifg that the thesis entitled PART I STUDIES OF THE SODIUM TUNGSTATE-HYDROGEN PEROXIDE OXIDATION OF 2—s.UBSTITUTED-l,3,5,6,7-PENTACHL0R0- BICYCLO p.243 HEPTA-2,6-DIEN-u-0NES PART II THE REACTION OF TETRACHLOROCYCLOPROPENE WITH POLYC HLOROET HYLENES? presented by Charles William Bauer has been accepted towards fulfillment of the requirements for Doctor of PhilasnphL degree in_Qh.emia_t2y 44/ Major profe Date // '2’ é 0—169 ABSTRACT PART I STUDIES OF THE SODIUM TUNGSTATE-HYDROGEN PEROXIDE OXIDATION OF 2-SUBSTITUTED-1,3,5AL7-PENTACHLOROBICYCLO[3.2.0]HEPTA- 2,6-DIEN-4-ONES PART II THE REACTION OF TETRACHLOROCYCLOPROPENE WITH POLYCHLOROETHYLENES BY Charles William Bauer This study was initiated as an effort to prepare new members of the oxocarbon series (1). This goal was not achieved, but several new and interesting compounds were .prepared. Sodium tungstate—hydrogen peroxide oxidation of the 2—substituted-1,3,5,6,7—pentachlorobicyclo[3.2.0]hepta—2,6— dien-4—ones 12/ 12/ and Zg resulted in the formation of the same two major products. 0 014 Cl 1 a. X = Cl Z X = OH . X=OCH3 Charles William Bauer One of the major products was identified as 1,2,4,6,7,8— hexachloro-Z-hydroxy—3—oxabicyclo[4.2.0]oct-7—en-5—one léx Assignment of this structure to compound lg is rather un— satisfying because of the presence of the chlorohydroxy function. However, the available evidence suggests this to be the correct structural assignment. H0 01 13 The second major product has been identified as tetra- chloro—3-cyclobutene—cis—1,2—dicarboxylic acid £$. 011+ 0023 0°25 14 A third product resulting only from the reactions in- volving 2g was identified as the methyl half-ester of the diacid $32 The formation of this product only in the reac— tions of'Zg has implications relating to the mechanisms of the reactions involved. Some suggestions concerning pos— sible mechanisms are included as are suggestions for further study. Charles William Bauer PART II This study was initiated in an effort to determine if the reaction of tetrachlorocyckmmopene with trichloroethyl- ene in the presence of anhydrous aluminum chloride might be similar to the reaction of hexachlorocyclopentadiene with trichloroethylene (2,3). c16 c19 A1013 + C12C=CHC1 :11 The results are quite different however in that no bicyclic products analogous to $1 were isolated. The product is rather bis(trichlorovinyl)cyclopropenone Q2, /J3=O / C12C=CCl-C = C-CCl=CClZ 52 The same reaction using 1,2—dichloroethylene yielded bis(1,2—dichlorovinyl)cyclopropenone 2%. Some mechanistic considerations are discussed. References 1. R. West and D. L. Powell, J. Amer. Chem. Soc., §§/ 2577 (1963). 2- E. T. McBee and J. S. Newcomer, Ibid., 1;, 952 (1949). 3- A. Roedig and L. Hornig, Ann. Chem., 598, 208 (1956). PART I STUDIES OF THE SODIUM TUNGSTATE-HYDROGEN PEROXIDE OXIDATION OF 2-SUBSTITUTED—1,3,5,6,7—PENTACHLOROBICYCLO[3.2.0]HEPTA- 2,6-DIEN-4-ONES PART II THE REACTION OF TETRACHLOROCYCLOPROPENE WITH POLYCHLOROETHYLENES BY Charles William Bauer A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1969 4’42 75$ 7’/’90 To Sue and My Parents ii ACKNOWLEDGMENTS The author wishes to express his gratitude to Dr. Eugene LeGoff for his patient guidance and encouragement during the course of this work and for providing financial support. Appreciation is also extended to the Department of Chemistry at Michigan State University and the National Science Foundation for providing support in the form of a National Science Foundation Traineeship from September 1965 to September 1969 and to the Department for teaching assist— ships at various times during this period. Sincere appreciation is extended to Mrs. Lorraine Guile who carried out the sometimes difficult task of ob— taining mass spectra. A final sincere word of appreciation is extended to my fellow graduate students who contributed much to making the last four years an enjoyable experience. TABLE OF CONTENTS PART I INTRODUCTION . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . A. Products . . . . . . . . . . . B. Discussion of Possible Mechanisms C. Conclusions . . . . . . . . . . . D. Suggestions for Further Study EXPERIMENTAL . . . . . . . . . . . . . . . A. General Procedures . . . . . . . B. Sodium Tungstate—Hydrogen Peroxide Oxidation of 2— Substituted- 1, 3, 5, 6, 7—pentachlorobi- cyclo[3. 2. 0]hepta- -2, 6—dien- 4— —ones 7a, 7b, and 7c . . . . . . . . C. Preparation of the Acetate 11 of Compound 13 D. The Ketalization Reaction of Compound 13 E. Sodium Tungstate—Hydrogen Peroxide Oxida— tion of Compound 13 . . . . . . F. Attempted Oxidations of Compound 13 1. The Attempted Oxidation of 13 Using Jones Reagent . . . - 2. The Attempted Oxidation with Chromium Trioxide in Acetic Acid . 3. The Attempted Oxidation with Chromic Acid . . . . . . . G- The Attempted Hydrolysis of 13 H. The Thermolysis of 13 . . . . . o rw I. The Determination of Hydrogen Chloride as a Product of the Thermolysis of 13 iv 31 50 52 54 54 54 56 57 57 58 58 59 59 59 60 60 TABLE OF CONTENTS (Cont.) J. Test for Enol Formation In 13 . . . . . . . K. Hydrogenation of Tetrachloro— 3— —cyclobutene— cLs——1,2——dicarboxylic Acid 14 . . . . . . . L. Esterification of Tetrachloro— —3—cyclobutene- cLs-1,2—dicarboxylic Acid 14 . . . . . . M. Preparation of the Anhydride 21 of Tetra- chloro—3— —cyclobutene— cLs— —L 2 —dIbarboxylic Acid 14 . . . . . . . . N. Esterification of Anhydride 21 . . . . . O. Esterification of the Methyl Half—Ester 15 of Tetrachloro— 3— —cyclobutene— cLs— —1, 2-di—w carboxylic Acid 14 . . . . . . . . . . . P. Hydrolysis Reactions of 7a, ZR, and 32 with Sodium Tungstate Present“ . . . . . . . . . Q. Hydrolysis Reactions of 7a and 7c Using On_ly Ethanol and Water . . . . . . . . . . R. Reaction of 2—Hydroxy—1,3,5,6,7—pentachloro— bicyclo[3.2.0]hepta—2,6—dien—4-one 12 with Sodium Carbonate . . . . . . . . . . . . . 1. Acidification of Product 16 . . . . . PART II INTRODUCTION . . . . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . . . . A. B. The Reaction of Tetrachlorocyclopropene With Trichloroethylene . . . . . . . The Reaction of Tetrachlorocyclopropene With 1, 2—Dichloroethylenes . . . The Reaction of Tetrachlorocylcopropene With Tetrachloroethylene . . . . . . The Reaction of Tetrachlorocyclopropene With l,1-Dibromo-,1,1—Dichlorow and 1,1— Dimethylethylene . . . . . . . Conclusinns . . . . . . . . . . . Page 61 62 62 63 64 64 65 66 67 67 69 73 73 77 81 82 83 TABLE OF CONTENTS (Cont.) Page EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . 84 A. The Reactions of Tetrachlorocyclopropene With Trichloroethylene . . . . . . . . . . 84 1. The Reaction Using a 10 to 3 Ratio of Tetrachlorocyclopropene and Aluminum Chloride . . . . . . . . . . . . 84 2. The Reaction Using a 10.0 to 7.5 Ratio of Tetrachlorocyclopropene and Aluminum Chloride . . . . . . . . . . . . . . . 85 B. The Thermolysis of Bis(trichlorovinyl)cyclo— propenone 52’ . . . . . . . . . . . . . . . 85 C. The Reactions of Tetrachlorocyclopropene With 1,2-Dichloroethylenes . . . . . . . . 86 1. The Reaction Using a Mixture of cis— and trans—1,2—Dichloroethylene and a 10.0 to 7.5 Ratio of Tetrachlorocyclo— propene and Aluminum Chloride . . . . 86 2. The Reaction Using a Mixture of cis- and trans—1,2-Dichloroethylene and a 4 to 1 Ratio of Tetrachlorocyclo- propene and Aluminum Chloride . . . . 87 3- The Reaction Using trans- 1, 2—Dichloro— ethylene . . . . . . . . . . . . . 88 4. The Reaction Using cLs——1,2—Dichloro— ethylene . . . . . . . . . . . . . . 89 D- The Reaction of Tetrachlorocyclopropene with Tetrachloroethylene . . . . . . . . . . . . 90 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . 91 APPENDIX . . . . . . . . . . . . . . . . . . . . . . 95 Vi LIST OF TABLES TABLE Page 1. Product yields from the oxidation of 7a, 7b, and 7c . . . . . . . . . . . . . . . . 8 2. Product yields from the hydrolysis of 12/ 22/ and ZS . . . . . . . . . . . . . . . . . 9 3. Chemical shifts of ethers and analogous hydrocarbons . . . . . . . . . . . . . . . . 20 4. Product yields from the hydrolysis of 7a, 7b, and 7c in the presence of sodium tungstatew . 65 5. Product yields from the hydrolysis of 7a and 7c with ethanol and water . . . . . . . . 66 6. Yields of cyclopropenones from the various dichloroethylenes . . . . . . . . . . . . . . 78 7. Mass spectrum of compound l2, . . . . . . . . 95 8. Mass spectrum of compound 11, . . . . . . 96 9. Mass Spectrum of compound 1§, . . . . . . . . 97 l SCHEME 11. 12. 13. LIST OF SCHEMES Reactions involved in the test for enol formation . . . . . . . . . . . . . . . . . . Spectral and chemical properties of 1,2,4,6, 7,8—hexachloro—2—hydroxy—3-oxabicyclo[4.2.0]— oct—7—en—5—one 13 and compounds 17 and 12 derived from lgffl'. . . . . . . TV. . . . . . Proposed mechanism of CrVI oxidation of secondary alcohols . . . . . . . . . . . . . Possible mechanistic route from £2 to lg . . Reactions of tetrachloro—3—cyclobutene-cis— 1,2—dicarboxylic acid 14’ . . . . . . . . . . Reaction sequence involving perchloro—1,2— dimethylene cyclobutane . . . . . . . . . . . Mechanistic proposal for the hydrolysis of 2—substituted~1,3,5,6,7-pentachlorobicyclo— [3.2.0]hepta-2,6—dien—4—ones 72 and ZS to Z2 Mechanistic proposal for the formation of £2 Mechanistic proposal for the formation of lg Mechanistic scheme for the conversion of 2,3- dihydroxytetraphenylcyclopentenone fig to 2—benzoyltriphenylfuran £1 . . . . . . . . . Mechanistic proposal for the formation of diacid 14 from diketone 2§, . . . . . . . . . Mechanistic proposal for the formation of the methyl half-ester 1E of diacid £4 . . . . . . Mechanistic proposal for the formation of bis(trichlorovinyl)cyclopropenone E2, . . . . viii Page 15 18 21 24 25 27 32 35 41 43 46 48 76 LIST OF FIGURES FIGURE Page 1. Infrared spectrum of 1,2,3,5,6,7—hexachloro— bicyclo[S.2.0]hepta—2,6—dien-4—one 13 . . . . 98 2. Infrared spectrum of 2—hydroxy«1,3,5,6,7- pentachlorobicyclo[3.2.0]hepta—2,6—dien—4—one IR . . . . . . . . . . . . . . . . . . . . . 99 3. Infrared spectrum of 2—methoxy—1,3,5,6,7— pentachlorobicyclo[3.2.0]hepta—2,6—dien—4~one . . 100 7c . . . . . . . . . . . . . . . . . . . 4. Infrared spectrum of 1,2,4,6,7,8—hexachloro—2— hydroxy—B-oxabicyclo[4.2.0]oct-7—en-5—one 12 101 5. Infrared spectrum of the acetate 11 of compound 1g, . . . . . . . . . . . . . . . . 102 6. Infrared spectrum of ketal 1§'. . . . . . . . 103 7. Infrared spectrum of diacid 14 . . . . . . . 104 8. Infrared spectrum of the anhydride 2; of diacid 14 . . . . . . . . . . . . 105 9- Infrared spectrum of the dimethyl ester 22 of diacid 14 . . . . . . . . . . . . . . . . 106 10. Infrared spectrum of cis-cyclobutane—1,2— dicarboxylic acid . . . . . . . . . . . . . . 107 11- Infrared spectrum of the methyl half—ester 12 of diacid 14' . . . . . . . . . . . . . . . . 108 12. Infrared spectrum of the sodium salt 1g of compound ZQ . . . . . . . . . . . . . . . . 109 13. Infrared spectrum of bis(trichlorovinyl)- cyclopropenone 22, . . . . . . . . . . . . . 110 14- Infrared spectrum of bis(trichlorovinyl)- acetylene Q21 . . . . . . . . . . . . . . 111 15- Infrared spectrum of bis(1,2—dichloroviny1)— cyclopropenone Qfi . . . . . . . . . . . . . . 112 ix ———_7 LIST OF FIGURES (Cont.) FIGURE 16. 17. 18. Page Ultraviolet spectrum of bis(trichlorovinyl)- cyclopropenone gg‘. . . . . . . . . . . . . . 113 Ultraviolet spectrum of bis(trichlorovinyl)- .. 114 acetylene Qg' . . . . . . . . . . . . . . Ultraviolet spectrum of bis(1,2—dichloro- vinyl)cyclopropenone g4 . ... . . . . . . . . 115 PART I STUDIES OF THE SODIUM TUNGSTATE—HYDROGEN PEROXIDE OXIDATION OF 2-SUBSTITUTED-1,3,5,6,7-PENTACHLOROBICYCLO[3.2.0]HEPTA- 2,6—DIEN-4-ONES INTRODUCT ION An interest in the synthesis of aromatic oxocarbon compounds (1,2) led to the study reported herein. In a communication published in 1960, West pointed out that the anion of the then recently synthesized compound diketocyclobutenediol 1 (3), now commonly referred to as Squaric acid, represented "... one member of a general series of symmetrical electron delocalized anions, CnOn—Z’ salts of which may be considered as a new class of aromatic substances." (4). HO 1 OH In a subsequent paper (1), molecular orbital calcula- tions for a large number of oxygenated anions were reported. Included among those systems studied, in addition to the monocyclic species ranging from three— to eight—membered rings, were several bicyclic systems. To date, studies of only four of the arOmatic oxocarbon anions have been reported. Included are the dianions of squaric acid 2, croconic acid g, and rhodizonic acid 4 (1,2,5,6,7), and the tetraanion of tetrahydroxyquinone 2 (1,2,8). 2N 2th 201 Since no report concerning the synthesis and study of any of the bicyclic members for which the calculations had been reported (1,2) had been published, and since some com— pounds which appeared to be suitable precursors were avail— able from previous work, attempts were made to prepare some of the bicyclic oxocarbons. A major effort was expended toward the synthesis of the oxocarbon compound 2. 0 H 0 0 0H 0 ° Ho 0 £5, These attempts proved to be unsuccessful. However, as a result of the methods and approaches used in the attempted Synthesis of g, an interest in the oxidation and hydrolysis 4 of polychloro organic compounds as an approach to poly- oxygenated systems, including the oxocarbons, resulted. This interest and the ready availability of the 2—substi— tuted—l,3,5,6,7—pentachlorobicyclo[3.2.0]hepta—2,6—dien— 4—one system 1 (9), led to the study reported herein. 0 014 a. X = Cl 01 b. X = OH C. X = OCH3 x 7 The initial thought was to study the epoxidation of la expecting the eventual product to be the bicyclic triketone §’or the monocyclic triketone g’which could result from ring opening of §’which would be expected to be highly strained. 0 0 014 0.:J<::::]:i:] --——-—-——-—> O 014 O 0 £3. 2 Should the epoxide of 72/ 143., 12/ be formed, one might anticipate, on hydrolysis of 10, the formation of the Chlorohydroxy intermediate 1} which would be expected to lose hydrogen chloride to give triketone g. 0 014 011* H0 01 01 10 £1 5 The literature contains several reports of a—chloro epoxides and their reactions (10). Those systems reported in the literature contain only one a-chloro substituent, and with the exception of the 2-chlorobicyclo[2.2.1]hept- 2—ene Egg—oxide system lg, the a—chloro epoxides studied undergo molecular rearrangement with chloride migration. In addition to the products resulting from chloride migra— tion, those resulting from Wagner-Meerwein rearrangements are formed in the rearrangement reactions Of,lg (10f). 0 \ ‘01 12 Thus, Ig'would be of interest not only as a possible precursor to triketone gland the oxocarbons, but also be— cause a study of this type compound, i;§., an a,a'-dichloro— epoxide system, to our knowledge, has not been reported in the literature. The sodium tungstate—hydrogen peroxide oxidizing system was chosen for two reasons. For the epoxidation of an Q,E- unsaturated carbonyl system such as Z, a nucleophilic epoxidizing reagent is preferred (11). In the sodium tungstate—hydrogen peroxide system, the reactive species is thought to be the pertungstate anion (12) which would have the necessary nucleophilic character, and indeed the above System has been used as a nucleophilic epoxidizing reagent E 7 6 system (12). Also, this reagent combination had been used by us previously with rather good results. Reaction of the above combination with tetraethyl ethenetetracarboxylate, which is very hindered sterically and quite electron poor, gave a 60 per cent yield of the corresponding epoxide (13). The results of the initial epoxidation reaction using 12 were much different than anticipated, and therefore, a study of the epoxidation reactions of systems Z2 and 22 was conducted. Thus the research reported in Part I is that having to do with the study of the sodium tungstate—hydrogen peroxide oxidation of the 2—substituted—1,3,5,6,7—penta— chlorobicyclo[3.2.0]hepta-2,6-dien—4—ones 22” 22/ and 7g and the products of these reactions. RESULTS AND DISCUSSION A. Products The sodium tungstate-hydrogen peroxide oxidation of the 2—substituted—1,3,5,6,7—pentachlorobicyclo[3.2.0]hepta— 2,6—dien—4-ones 13, lb” and 12 afforded the same two major products in varying yields depending on which of the three starting compounds was used. The general range of yields of products from the various reactants is shown in Table 1. O 014 30%Ho c1 “— 13 + 14 . Na2WO4‘H20 M W C7H2O3C16 C6H204Cl4 I 7 a. X = Cl b. X = OH C. X : OCH3 The data in Table 1 suggest that the products formed may be at least in part arising from compound 12 since,:Q3 has been isolated from the reactions involving Ia and 12; This possibility will be discussed later when possible mechanisms are considered. - Hi, ' . Table 1. Product yields from the oxidation of 72/ ZR” and 7c. m Start— % Starting % Product 0 Product % Product % 13 14 15 ing Material a Com— Recovereda ZR« a a a pound C7H203C16 C6H2’04C14 C7H4O4Cl4 72, 40—45 0—15 15—20 15—20 0 7b, 0 0 20—25 40-50 0 lg, 10—15 0—20 35—45 20—30 2—20 aPer cent of product recovered after chromatography. Yields based on starting polychloro compound used. The possibility that 72 was resulting from the hydrol- ysis of 1a,and Zg,in the reactions involving these compounds was studied by allowing them to react under conditions identical to those for the oxidation reactions except that no hydrogen peroxide was added to the reaction mixture. The results of these reactions are given in Table 2 along with the results for some other hydrolysis reactions involving compounds 22/ lb” and 72. The results in Table 2, which show that lb is a product of the hydrolysis of both Zaiand 12/ along with the results shown in Table 1 would suggest that under the conditions used for the oxidation reactions the hydroxy compound lb] might be formed. The possible formation of 7b in the reac— tions of la and ZS will be discussed more fully later when possible mechanisms are considered. 9 Table 2. Product yields from the hydrolysis of 22/ lb, and Mg]. Starting % Recovered % % Material Starting Material 22’ 16 Sodium Salt 7a a 32 47 0 go“ a 16 16 55 lg a 0 13-22 48—60 R b 38 38 0 1c; b o 92 o 113 C o o 90 aReaction using sodium tungstate but no hydrogen peroxide. bReaction using neither sodium tungstate nor hydrogen per— oxide, rather only ethanol and water. CReaction using sodium carbonate rather than sodium tungstate and no hydrogen peroxide. On analysis of products 13 and 14, it became evident that neither was epoxide 12 or triketone §lor 9; The anal— ytical and mass spectral results agree in suggesting a formula of C7H203C16 for compound 13 (mp 129-1300). This result is surprising for several reasons. First, product 13’ contains six atoms of chlorine while compounds lb and 1g contain only five atoms of chlorine. Second, both chemical and spectral evidence show compound 1§,to contain a carbon— hydrogen bond in addition to an hydroxyl function. The infrared spectrum of 13 shows a carbon—hydrogen absorption near 3000 cm_1 and an oxygen—hydrogen stretching absorption 10 at 3350 cm—1. None of the three starting compounds con— tains a carbon—hydrogen bond. The nmr spectrum of 13 shows a sharp singlet at 5 6.25 (1H) and a broad singlet at 6 5.20 (1H) which disappears after treatment of the sample with deuterium oxide. The nmr and infrared spectra suggest that a single hydroxyl function and a single hydrogen bonded to carbon are present in compound 13, This structural arrangement was further substantiated by the formation of a monoacetate 11 (mp 183— 1840) on treatment of 13 with acetic anhydride. The anal- ysis of compound 11 corresponds to a compound of formula C9H4O4C16. The infrared spectrum of 11 shows no hydroxyl absorption. The nmr spectrum of £1 contains two sharp singlets at o 6.00 and 6 2.25 in a ratio of one to three. The spectrum shows no change on treatment with deuterium oxide. The infrared spectrum of 13 provides significant in— formation in addition to that noted above, i4g., that carbon-hydrogen bonding is present and that an hydroxyl function is also present. In addition, a carbonyl absorp- 1 and a carbon-carbon double tion at approximately 1790 cm— bond absorption at approximately 1630 cm_1 are present. The absorptions which may be attributed to the carbonyl function and the double bond in the cyclobutene ring in 1 compound Za’occur at 1750 and 1620 cm_ respectively. This comparison suggests that the carbonyl function in l§.is no 11 longer part of an a,B—unsaturated system and that the cyclo— butene ring has remained intact. Chlorinated cyclobutene systems show characteristic infrared absorption in the region of 1600 to 1640 cm_1 (14), and specifically, the double bond absorption of hexachloro— 1 while hexa— cyclobutene is reported to occur at 1635 cm— chlorobutadiene shows absorptions at 1611 and 1570 cm—1 (15). This further suggests that the cyclobutene ring is present in compound 13; The ultraviolet spectrum of 7a (9) shows a maximum absorption at approximately 260 mu with a shoulder at ap— proximately 280 mu. Compound 13 shows only a broad region of absorption below 225 mu which suggests the presence of an isolated carbonyl function rather than one which is in conjugation with a double bond. Thus the ultraviolet and infrared spectral data would suggest that lg'contains a cyclobutene ring and an isolated carbonyl function which is no longer part of the Q,fi—un— saturated system as in the starting compounds 72“ 7b] and 223 One may then begin assigning a structure to compound 12 considering the above information. C14 12 The remaining portion of the molecule must then be composed of CZHZOZCIZ. One of the oxygen atoms is part of the hydroxyl group. Since only one carbonyl absorption is evident in the infrared spectrum of 1§,and one hydroxyl function is present, the remaining oxygen atom must be in- volved in an etherifinkage. Based on the empirical formula of 13, 112;! C7H203C16, and the rule of rings plus double bonds (16), compound 1§,must contain four rings and/or double bonds. Considering the previous evidence that one carbonyl function and a cyclobutene ring are present, one is left with the choice of either one additional ring or one additional double bond being present in the molecule. The evidence would suggest that the remaining oxygen is in- volved as an ether rather than an hydroxyl or carbonyl func- tion. Thus the two remaining carbon atoms would have to be joined by a double bond if there were to be an additional double bond rather than a ring structure in the molecule. This is unlikely since there is no conjugation evident in the system and since such an arrangement would necessitate that the hydroxy function be present as part of an enol system. It thus appears that a bicyclic system containing an ether likage is the more likely choice. The mass spectrum of 1§I(Table 7) is very complex and therefore difficult to analyze. That of its acetate £1 (Table 8) is much less complex and lends itself more readily to analysis. I r 13 The first major fragmentation loss in the mass spec— trum of lilis that of C303Cl3. The peak at m/e = 43 cor- responding to C2H30 is the most intense in the spectrum while that at m/e = 44 corresponding to C02 is also quite intense. One might then consider that the loss of C3H303 actually occurs as the loss of C2H3O and C02. Loss of C3H303 (or C2H30- and C02) yields an even electron ion C6HOC16+ which appears to lose CO and CHC12+ giving a neutral fragment of high stability corresponding to C4Cl4 m/e = 188, 190, etc. This neutral fragment should contain three rings and/or double bonds and might be tetra— chlorocyclobutadiene. Peaks in the spectrum corresponding to CO, m/e = 28 and CHC12+, m/e = 83, 85, and 87 are present in significant intensities. A significant feature in the mass spectra of both 5% and its acetate 11,18 the presence of a rather intense peak at m/e = 44 which may be assigned to C02. The presence of carbon dioxide as a product in the fragmentation of 13 suggests that two of the oxygen atoms are bonded to a common carbon atom. This indicates that the ether oxygen is bonded to the carbon bearing either the hydroxyl or the carbonyl function. To this point, the only chemical evidence cited con— cerning the structure of compound lg‘is the fact that a monoacetate is formed when 12.18 allowed to react with acetic anhydride. Additional chemical evidence concerning the structure of lg’is available. IIIIIIT________________________________—_________———_VI" 14 On heating lg’to its melting point of 130°, a gas is evolved which is acidic to moist pH paper and gives a posi— tive test for halogen (17). This suggests that the gas evolved is hydrogen chloride. Attempts to isolate some identifiable compound on pyrolysis of 13 were unsuccessful and only a yellow—orange oil was isolated. In an effort to determine whether the hydroxyl func— tion and the single hydrogen atom were bonded to the same carbon atom, i;g., whether 13 contained a secondary hydroxyl function, attempts were made to oxidize 13; Oxidations using reagent systems ranging from the relatively mild neutral Jones reagent (18) to acetic acid—chromium trioxide and to aqueous Chromic acid resulted only in the recovery of start— ing material. These results suggest that either the secondary hydroxyl function is not present or that for some reason it is unreactive. The first possibility seems more likely since the oxidation of alcohols under the above conditions is not very subject to steric effects (18). A test for the presence of an enol was conducted in an effort to determine whether the hydrogen atom in 13 is q to the carbonyl function. The reactions involved in this test are outlined below (19). The reaction upon which the test is based is the oxidation of iodide to iodine by an d—bromoketone. Compound 13 gives a positive result using the test de- scribed above. Blank tests using lfi’but no bromine and using none of compound 13 do not give positive results. Therefore, 15 OH ..._) \ 8:2 3 Br H Br -HBr ° 0 ZHI MMMMN H Br Scheme 1. Reactions involved in the test for enol formation. the results would suggest that an a—bromoketone is formed from an enol form of 13; However, an effort to isolate a bromoketone from the direct bromination of 13 resulted only in the recovery of starting material. Thus, acceptance of the positive results of the above mentioned test as evi— dence for the presence of an enol of 13 and hence the hydro- gen atom being a to the carbonyl function may be ques— tioned. One additional reaction of lg’appears to be quite pertinent to the question of its structure. Reaction of 1§,with ethylene glycol in benzene using a catalytic amount 16 of p—toluenesulfonic acid afforded a crystalline substance 18 (mp 157.5—159.0°) which, based on mass spectral and chem— ical analysis, is a compound of formula 09H504C15. This compound displays a carbonyl absorption in the infrared at 1760 cm_1 and a carbon—carbon double bond absorption at approximately 1630 cm-1. In the nmr spectrum of 1§x a single peak is present at 6 6.77 in addition to a broad region of absorption between 6 4.0 and 6 5.0. The peak at 6 6.77 is significant since its presence suggests that the proton giving rise to the low field absorption in 13 is still present in compound 18” Points of major significance concerning this compound are the fact that no hydroxyl absorption is observed in the ir, a carbonyl function is still present even though a ketal has apparently been formed, the new product contains one chlorine atom fewer than the starting compound, and a proton absorbing at quite low field is still present. Consideration of this information together with the previously noted information concerning 13 suggests that apparently the ketal of l§.is formed which loses hydrogen chloride resulting in the formation of a new carbonyl func— tion. The fact that the low field absorption is still present in the nmr of 18'suggests that the single hydrogen atom in 13 is bonded to a carbon atom also bearing a chlor— ine atom rather than the hydroxyl function because the hydox— Yl function is not present in 18. Loss of hydrogen chloride apparently from the ketal of 12.1“ the above reaction and 17 from 13 when it is heated to 130°, indicates the presence of a chlorohydroxy function in 13; Consideration of the available information concerning 12/ its reactions, and the reactions involved in its formation lead to the suggestion of the structure below for this compound. 01 o 5 01 H 1" 7 5 O 2 1 8 HO 01 12 1,2,4,6,7,8—Hexachloro—2—hydroxy-3- oxabicyclo[4.2.0]oct—74en—5-one A brief summary of the above results leading to the structure suggested for compound 13 is given in Scheme 2. Duplicate chemical analyses and mass spectral data con— firm that 13 has a formula C7H203C16. The spectral proper- ties are consistent with the proposed structure. The uv spectrum suggests no conjugation is present in the molecule; the ir spectrum is consistent with the presence of the hydroxyl function, the carbonyl function, and the chloro— Cyclobutene ring, and the carbon—hydrogen function; the nmr spectrum is also consistent for a molecule containing a Single proton absorbing at quite low field (5 6-52) in 18 13 (mp 129—1300) C7H203C16 uv (methanol) No absorption above 225 mu. ir KBr) 3350 (O-H), 3000 (C-H), 1790 (c=o), and 1630 cm‘1 chlorocyclobutene C=C). . nmr (acetone-d ) 0 6.52 (s); (acetonitrile) 0 6.25 (s, 1) and 0 5.20 (broad s, 1). Absorption at 0 5.16 not present after treatment of the sample with D20. is; 130° Test for enol formation. A Loss of HCl. Positive; No product No isolable product Oxidation1solated however. Starting material only. a acetic \\\ ethylene glycol 7 anhydride p-toluenesulfonic acid, benzene ’1‘; (mp 183—1840) 2.13. (mp 157.5—1590) ir (KBr) no o—H, 1810 (C=O), ir (KBr) no o—n, 1760 (C=O), 1760 (c:o), and 1630 cm"1 and 1630 cm 1 (chloro— (chlorocyclobutene C=C). cyclobutene C=C). nmr (CDCla) 0 6.00 (s, 1) and nmr (CDCla) 0 6.77 (s, 1) é-2.25 (s, . and 0 4.0 to 0 5.0 (bm, 4) Scheme 2. Spectral and chemical properties of 1,2,4,6,7,8— hexachloro-Z—hydroxy—B—oxabicyclo[4.2.0loct—7— en-5—one 13 and compounds lzland 18 derived from 2.2:- 19 addition to the absorption due to the hydroxyl function which disappears after treatment with D20. That such a low field absorption might be predicted for a methine proton is discussed below. The proton at C—5 in compound 19 is reported to ab— sorb at 0 4.56 (20). Replacement of a methylene group with an ether oxygen causes the protons on adjacent methylene groups to be shifted downfield by approximately 2 ppm as shown in Table 3. Thus one might predict that the proton at C—4 in compound 13 would absorb in the region of 0 6.5. The results given in Scheme 6 for compounds 13” 11” and 18’ are all quite consistent with this prediction. 12, The loss of hydrogen chloride from 13 on heating is quite significant since a system containing the chlorohydroxy function would be expected to lose hydrogen dfloride very readily. Indeed, the very fact that the structure proposed for 13 contains such a functionality is the major argument against that structure. However, the great majority of evidence tends to support the structure suggested. 20 Table 3. Chemical shifts of ethers and analogous hydro— carbons (21). Compound Chemical Shift Difference in of -C_2- (0) Chemical Shifts Pentane 1.27 2.09 ppm Diethyl Ether 3.36 Cyclopentane 1.51 2.12 Tetrahydrofuran 3.63 Cyclohexane 1.45 2.11 Tetrahydropyran 3.56 Loss of hydrogen chloride from 13 should yield the lactone shown below which was not isolated. It is believed that at the temperature to which lglmust be heated for loss 0 01 H 01" -HCl H -——-——————> 0 O 0 HQ 01 13 22 Of hydrogen chloride to occur, i.e., 130°, the lactone is unstable. For this reason, no identifiable products could be isolated from the pyrolysis of 13. The results of the oxidation reactions suggest that the hydrogen atom and the hydroxyl function are not bonded to the same carbon atom. A probable mechanism for the 21 I as outlined by House (22) and oxidation of alcohols by CrV shown below requires the presence of a proton. A reaction proceeding by this mechanism would not be likely to occur with lglsince it would require loss of chlorine as an electron deficient species. £\\ 0 5\ _ + II Y/’CH OH + HCrO4 + H : > y/,CH-O Er OH + H20 . O O “ }\\ - + f—c;o&cr—0H —————> c=o + Hcro3 + H30 H O T /O\ H H Scheme 3. Proposed mechanism of CrVI oxidation of secondary Alcohols. A compound having the structure proposed for 13 might be expected to give a positive test for enol formation since the q-hydrogen atom is present. However, there is some question why the q—bromo compound has not been isolated. Attempts to prepare the bromo compound have not been exten- sive. The reaction of 13 with acetic anhydride to give ace- tate 11 further confirms the presence of a single hydroxyl function. The spectral properties of 11 are consistent with the structure proposed for it. 22 o 01 011* H o .9 euro-o 01 11 Of major significance is the ketalization reaction of 13 which yields compound 18; The significant points con- cerning this compound are that it contains five chlorine atoms rather than the six found in 13 and that no hydroxyl function is present. The structure proposed for 18 is that of the lactone shown below. [—1 o 0 01 a 014 o 0 13% The fact that only five atoms of chlorine are present in 18 and that the low field hydrogen absorption (0 6.77) is still present in the nmr strongly suggest that the two chlorine atoms in 13 which are not part of the cyclobutene ring are bonded to different carbon atoms. The loss of hydrogen chloride from 13 and the absence of the hydroxyl function and one atom of chlorine in 18 suggest that one of 23 the chlorine atoms and the hydroxyl function are bonded to the same carbon atom in lfix i;§., that a chlorohydroxy function is present. Apparently, the relief of strain which results from formation of the ketal of 13 is sufficient to allow for the loss of hydrogen chloride at lower temperature, i;§., at or below the 800 temperature of refluxing benzene in the ketalization reaction. The formation of 18 results in a system containing only one sp2 center in the six—membered ring. Thus, energetically one might predict that 18 should be no worse than 13; However, as noted previously, loss of hydrogen chloride from lg’would result in the formation of the keto lactone gglwhich contains two sp2 centers in the six—membered ring and is apparently not stable at 130°. These considerations are briefly outlined below. The spectral and chemical factors considered together suggest that compound 13 has the proposed structure. To propose a structure containing a chlorohydroxy function for a compound of such apparent stability is not particularly satisfying. However, after many doubts and reconsiderations, and for the lack of a better suggestion, this is the struc— ture which we would propose as best fitting the available information. The structural analysis of the remaining compounds proved to be somewhat less difficult. Compound 14 (mp 156.5—157.5°) analyzed for a compound of formula C6H204Cl4. The presence of a broad absorption 24 from 3000 to 2500 cm-1 in the infrared spectrum of 14’sug- gested that it contained a carboxylic acid function. A moderately strong absorption at ~1640 cm‘1 suggested that the cyclobutene ring was still present in the molecule. As noted above, the double bond absorption in hexachlorocyclo- butene is reported to occur at 1635 cm"1 (15). O ' I 01 014 O O a ' + 800 °1 01¢ . I HOCHZCHZOH m> H ° I O .. a. X BO 01 13 1300 800 or "HC]. 1955 -HCl \) 0 O O 01 01 0 o 39—» unidentified oil H O O 0 32 .13: Scheme 4. Possible mechanistic route from 1§.t0 18; A number of chemical transformationsinvolving acid £4 which give further information concerning its structure are outlined in Scheme 5 following. 25 C6H204Cl4 14 acetyl Hz/Pt chloride HCl methanol cis-cyclobutane-1,2— CegiCl4 dicarboxylic acid. Methanol HCl C3H604Cl4 22 Scheme 5. Reactions of tetrachloro—3—cyclobutenejgi§—1,2— dicarboxylic acid £4. Acid 14 may be converted to its methyl ester 22 in very good yield by treating a methanol solution of the acid with dry hydrogen chloride. Evidence that the cyclobutene ring is retained in the ester is again found in the infrared spectrum. The carbon-carbon double bond absorption occurs at ~1635 cm_1 almost identical to that absorption in the acid. The nmr spectrum of 22 displays only a single sharp peak at 0 3.75. Treatment of 14 with acetyl chloride affords an anhy— dride 21 (mp 137.0-138.5°) which may be converted to the methyl ester 22 as shown in Scheme 5. Compound 21 displays absorptions in the infrared typical for anhydrides, i;g., two carbonyl absorptions at 1860 and 1795 cm_1 (25), and in 1 addition an absorption at 1625 cm_ which may be assigned to the double bond of the cyclobutene ring as noted previously. 26 The acid may be hydrogenated yielding gig—cyclobutane- 1,2-dicarboxylic acid, identified by comparison of infrared spectrum with a standard spectrum (32). This,in addition to the above chemical and spectral evidence,strongly sug— gests that 14 is a tetrachlorocyclobutene dicarboxylic acid. The fact that the nmr spectrum of the dimethyl ester 22 shows only a single peak for the methyl groups limits the possible structures for diacid 12,t0 those two shown below. 014 002“ 011‘ 002a 0023 0023 14a 14b Some additional facts indicate that the diacid should be assigned structure 142 rather than 14g. The ultraviolet spectrum of the diester 22 has a maximum absorption at 212 mu (log 6 = 3.98). This low wavelength absorption suggests that there is no conjugation between the double bond and the ester functions. One might compare the uv spectrum with that of the dimethyl ester of 1—cyclobutene—gis-1,2— dicarboxylic acid which shows a maximum absorption at 232 mu (log 6 = 3.94) (26). Also, the position of the double bond absorption in the infrared spectrum at 1635 cm—1 suggests that the cyclobutene ring is not conjugated with the ester functions. 27 Further, it seems doubtful that under the conditions of the reactions involved that the double bonds in the starting polychloro compounds 72” lb” and 22 would isomer- ize. A further, and perhaps weak, point is that formation and isolation of the anhydride analogous to 14b might be difficult and even unlikely. A compound said to have structure 142 had been re— ported previously in the literature (27,28). Scheme 6 gives a brief outline of the report noted. ccl2 c12 HN03 1750 '—_———> C6H204Cl4 —> C6H204C14 C12 135° 5 min. mp 174° mp 198° CClZ a b N N l acetic anhydride orl acetyl chloride (reflux) H2/P Hz/Pt cis-cyclobutane-1,2— dicarboxylic acid Scheme 6. Reaction sequence involving perchloro—1,2— dimethylene cyclobutane. The authors suggest that the two compounds 3 and b are the Eli and trans isomers of tetrachloro-3-cyclobutene-1,2— dicarboxylic acid. In addition to the differences in the melting points, the two compounds above have infrared spec— tra much different from that of compound 12,(28)° The results outlined in Scheme 6 are somewhat puzzling. For example it seems unlikely that heating with acetyl 28 chloride or acetic anhydride should cause a gisjtrans isomer— ization in a fully chlorinated system such as 142, It would seem likely that such a procedure would at best result in the formation of an anhydride as was our experience with compound 14; If an anhydride were formed as an intermediate in the reaction of compound a,it seems curious that such would result in a gisjtran§_isomerization. There might also be some question as to why the gis— and trans—dicarboxylic acids should both give gig-cyclobutane—l,2—dicarboxylic acid on hydrogenation. We are unable to explain the results in Scheme 6, but we feel that product 14 is tetrachloro-3—cyclobutene—gi§— 1,2—dicarboxylic acid. Thus, 21 and 22 would be the analo— gous anhydride and dimethyl ester respectively. Considering our study, the results outlined in Scheme 6 would apparently require some alternative explanation. Compound 15 (mp 93.5-950) is unique in that it is a product only in the reactions involving lg; Purification of 15 was quite difficult, but an analytical sample was eventually prepared. Analysis of this sample showed it to be a compound of formula C7H605Cl4. The infrared spectrum of 15 is quite similar to those of diacid 14 and its di~ methyl ester 22, On treatment of a methanol solution of 15 with hydrogen chloride, the dimethyl ester 22 results. The analysis and spectral properties coupled with the fact that lfilis so readily converted to the dimethyl ester 22 suggest that 15 is the methyl half—ester of diacid 14 and that 29 apparently one molecule of bound water was present in the analytical sample. The empirical formula of the analytical sample might then be more correctly written as C7H4O4Cl4'H20. The nmr spectrum of a carefully dried sample of 15 displayed two singlet absorptions at 0 11.07 and 0 3.87 in a ratio of one to three. This is an agreement with the spectrum one might predict for compound 15, 014 002035 002K 15 NV Compound 15 will be discussed more fully when mechan— isms are considered. Also, at that time, a possible reason why a third major product is formed only in the reaction of Zglwill be discussed. One additional compound encountered during this study deserves some mention. As noted in Table 2 , a compound 16 was isolated from several different reactions. This compound was recovered in yields ranging from zero to greater than 90 per cent depending on reaction conditions. We have been unable to obtain a satisfactory analysis for this compound, but based on the evidence at hand, we feel rather certain concerning its identity. The infrared spectrum of lg‘dis— plays no carbonyl absorption but rather a strong broadened 1 region of absorption from 1585 to 1540 cm_ which is 30 characteristic of organic salts (25). Extraction with nitric acid of a residue remaining after burning a sample of 16 and evaporation of the acid yields a residue which gives a positive test for sodium ion using magnesium uranyl acetate reagent. Acidification of a methanol solution of 16 yields a white crystalline material identified as ZR which suggests that compound 16 is the sodium salt of ZR. 014 + .. 01 Na Base > 7.12, 12 Additional evidence that 12 is acidic and may react with weak bases is shown by its reaction with sodium car- bonate. In this instance, a nearly quantitative yield of the sodium salt of 72” 14g., 16, was isolated and then re— converted to 7b as described above. This reaction pathway is apparently not significant under the usual reaction conditions when hydrogen'peroxide is present since none of l§,is isolated from those reactions. Also, the reaction mixture becomes acidic apparently due to the presence of hydrogen chloride which might be formed when the diacid 14 is formed. Therefore the acidic nature of the reaction mixture would preclude the formation of 16x 31 However, sodium tungstate is apparently a sufficiently strong base to react with 2b to form lglunder neutral con- ditions. B. Discussion of Possible Mechanisms The fact that the same two major products 13 and 14 result from the reactions of all three starting compounds suggests that a common precursor may be involved. Evidence that this precursor might be 72 has been noted previously. The data in Table 1 which show that ZR is sometimes isolated from the reactions involving 72 and 72 and in Table 2 which suggest that 19 may be formed from 72 and ZS under the re- action conditions used in this study would give some sup— port to the suggested intermediacy of ZR. The conversion of 73 or ZS to Zb may be occurring by the pathway shown in Scheme 7 or some related mechanism. Product 13 is apparently formed from lb since both con— tain the hydroxyl function. Also, it seems unlikely that hydrolysis of the chloro or methoxy function to hydroxyl would occur after the oxidation of 72 or ZS' This point will become more evident when the probable structure of the product of the oxidation reaction is discussed. Evidence that the diacid product 14 may be formed from 13 is available in that the reaction of 13 under the usual reaction conditions affords a 30 per cent yield of 14 in addition to recovered starting material. Evidence suggest— ing that product 14 may be formed directly from Za’and ZR 32 Cl ¢_——_——_—> = Cl OH = OCH3 NNN II more ——-—> c 0— 014 c1 + x x 0' C14 1 3:; H20 X 2.1:», + -H a —-—:— 1 H0 22 > C 23 __;H_X__. (lbw W OH c14 X OH C14 0 7b Scheme 7. Mechanistic proposal for the hydrolysis of 2—sub- stituted—1,3,5,6,7—pentachlorobicyclo[3.2.0]hepta- 2,6-dien-4—ones 72 and Zg’to ZR. 33 is available. The indirect evidence for this is the isola— tion of product 1§,from the reaction of 72, This product which is the methyl half—ester of 14 could not be resulting from the esterification of 14 because the solvent used in the reaction is ethanol. Therefore, the methyl group must be coming from the starting methoxy compound, and 15 must be formed directly from 72; The connection between this fact and the suggestion that 14 may be formed directly from 1% and 3b,will be discussed later. The available evidence thus suggests that 13,15 formed from 1§Ias well as from 72 and 7b directly and that ZR is involved as an intermediate in the reactions of 72 and 223 The questions which must be discussed then are 1) by what mechanism may the formation of l§.be occurring, 2) how may diacid 12.be formed from 73, lb, and 13” and 3) by what mechanism might the methyl half—ester 15 be formed, and what connection does the mechanism of its formation have with the answer to question 2? Possible answers to these questions will be discussed below. One might profit by first considering how the expected oxidation reaction may proceed. As noted previously, the sodium tungstate-hydrogen peroxide reagent system was chosen because of its proven utility in the oxidation of a,6-un— saturated carbonyl systems (12,13). The available evidence (12,29) indicates that the reactive species is some form of pertungstate ion which for simplicity will be written as below. 34 O ‘0 4(1-0—0‘ 5 This is apparently the nucleophilic species which reacts with a,B—unsaturated carbonyl systems such as those in- volved in this study. Therefore, the initial oxidation of the compounds used in this study, 143,, the 2—substituted-1,3,5,6,7— pentachlorobicyclo[3.2.0]hepta—2,6—dien—4—ones IE” 22” and 12” may be occurring in the manner shown in Scheme 8. Considering the starting materials and the reagent system involved, it seems quite probable that 26 is the primary intermediate resulting from the oxidation reaction. The identity of xlis of importance. As noted previous- ly, it seems likely that in the reaction leading to the formation of £2, Zb’is involved; therefore 2gb’would be the likely intermediate. Also, it seems quite unlikely that hydrolysis of 222 or 262 to yield 2fib’would occur since the epoxide might be expected to be more subject to hydrolysis than the chloro or the methoxy function. There is evidence, however, that 262 and 2§glmay be involved in the reaction. Assuming that 262 is the precursor to product 12, one might then consider what further reactions such a molecule might undergo. Since 2gb‘is an unusual molecule rich in functionalities, predictions concerning its further reactiv- ity may be difficult, but one may start by considering pos— sible reaction pathways leading from the epoxide function. O 0 C14 C14 Cl ¢—-——-———> Cl 7 . — l + X ~ a X C X b X = OH O 1 c1, c1 / O O-W‘O‘lO X 0 2:», ° c14 ‘2 c1 + wo4 O b. X = OH c. x = OCH3 Scheme 8. Mechanistic proposal for the formation of 26. 36 014 01 08 26b The usual reactions of epoxides in solution are of two general types (30,31). The first and most thoroughly studied involves attack of a nucleophile on one of the carbon atoms of the epoxide ring resulting in cleavage of the carbon—oxygen bond of which the carbon atom undergoing attack is a member. The final product is an alcohol. This type of reaction may occur under basic, neutral, or acidic conditions. The reaction mixture studied contains water, ethanol, and pertungstate ion and is also acidic. There should be ample opportunity for nucleophilic attack on 262. There would seem to be no steric factor which would direct nucleo— philic attack to either the chloro or the hydroxy bearing carbon atom. Therefore, one might suppose that electronic factors would be more important. Accepting the proposed mechanisms for nucleophilic cleavage of epoxides under acidic conditions which suggest that the carbon atom being attacked has at some stage in the reaction positive char- acter (30), the preferred point of attack would be the carbon bearing the chlorine atom. The two intermediates to consider 37 are 27a and 27b with 27a being less favorable since the positive character would be on the carbon adjacent to the carbonyl function. O 01‘ It 5‘0 OH = HOH, 27a Nov EtOH, w05'2 Such attack, whatever the nucleophile might he, would yield a system which would be expected to lose chloride readily to give the a-diketone 22; 014 0 Y = —OH, -OEt, —o—o—wo3_ no, gisv The significant point in this discussion is that although a species such as 28 is probably involved in the reaction, it seems unlikely that 12 is resulting from a reaction path- way including 28. Thus, the first major type of reaction of epoxides in solution, i.e., nucleophilic attack with resultant cleavage of one of the carbon—oxygen bonds, does not appear to provide an answer to the question of how 12 might be formed. 38 .The second major type of reaction of epoxides in solue tion is that of rearrangement in which one of the substitu— ent groups migrates resulting in formation of an aldehyde or ketone. Such rearrangements may also involve ring con— traction or expansion and are usually catalyzed by Lewis acids although heat is sometimes sufficient to cause such rearrangement (30,31). Assuming, however, that such a re— action might occur under the conditions used, one might con— sider the possible products. Migration of the hydroxyl group is quite unlikely since it is a poor leaving group. Also, such migration would require positive character in the transition state at the carbon a to the carbonyl function which is unfavorable as noted above. Since migration of acyl groups is quite favorable (30), one might consider the product resulting from such a rearrangement. 0 Cl“ 0 014 o -———-—-————9 '\ ° 01—0l H OH 26b 23 There is no evidence that 22 or a product resulting from further reaction of 22 is formed in the reaction. Re- arrangement with migration of the hydroxyalkyl group which would also yield a bicyclo[2.2.0]hexanone is also unlikely for the same reasons. 39 The final possibility then is migration of the chloride atom. As noted in the introduction, few studies of chloro epoxides have been conducted, and the majority of these have been studies of the pyrolysis of monochloro epoxides (10). Results of thermal reactions (10e, 10f, 10g) and solution reactions (10b, 10c, 10d) indicate, however, that chloride has a strong tendency to migrate. Such migration in 26b might be expected to give product 22 or possibly 8. ° 01,, ° 01,, Cl ’ 4101 Egg ___Jz_14> O -———————9 O 1 0 1 0H 0 50 N 9. Again, no product like 12 would be expected from this re- action. Thus, a consideration of the normal reactions of epox— ides in solution is not very helpful in determining a pos— sible mechanism for the formation of 12; It does, however, seem quite probable that some of the systems which would result from the expected reactions of 262 such as 22 and 39 are involved in the formation of the other products, namely 14 and 15. This possibility will be di5cussed more fully later. One must now consider some alternative pathway leading to 1E, Assuming for the time that 262 is indeed the pre— cursor to 22” one is left with a single reaction pathway. 40 This involves cleavage of the carbon—carbon bond of epoxide 262/ Although such an apparent cleavage of the carbon- carbon bond is known in some thermal (32) and photochemical reactions (33) of epoxides, only one such reaction has been reported in the solution chemistry of epoxides (34). What appears to be happening is the addition of hydro— gen chloride across the carbon—carbon bond in 262; Such direct cleavage of the carbon—carbon bond is reported in the reaction of tetracyanoethylene oxide with pyridine (34). This is in effect what appears to be happening. In fact, 12’ may be resulting from some totally different pathway which has not been considered. A significant amount of thought has not led to any other more satisfactory mechanism for the formation of 122 Nevertheless, those mechanisms shown in Scheme 9 are not very satisfying. A four—center addition concerted or otherwise is not meant to be implied by structure 22; Rather, this structure is meant to show what is in effect occurring. Addition of hydrogen chloride in the revase manner would lead to product 32 which is much more satisfying structurally in that it contains no chlorohydroxy function. This does not appear to be the final product. The reaction involving 34 is somewhat more satisfying in that 24 seems a reasonable intermediate. Also, a reac— tion somewhat similar to that leading to the formation of §4 is reported in the literature (35). Treatment of the dihydroxytetraphenylcyclopentenone 36 with acid yields the 41 o o c14 H C14 c1 .1 II] __.________> o 0 OH c1 OH 31 13 O 0 C14 C C14 Cl C I] ———* > o 0 OH H OH 2% ii 0 0'- 0 C14 c1 C14 cl C14 - /// c1 > < > O o 0 ~ - + Q + o .q. H \H H 2:1 26b +H+ o o H H Cl4 Cl4 C1 — C1 < + c1 0 O + Cl OH OH 12. 22 Scheme 9. Mechanistic proposal for the formation of 12, 42 benzoyltriphenylfuran 21, The proposed mechanism is that shown below. The first step is the one of significance since in this step the carbon—carbon bond is broken. The addition of chloride to species 25 in the manner shown is not very satisfying for two reasons. First, this addition leads to the chlorohydroxy function. One might suspect that loss of a proton from 35 would be more favor— able than addition of chlorine. Second, if chloride addi- tion is occurring by some mechanism which involves chloride as a nucleophile, one might expect the addition of chloride ion to the reaction mixture to have some effect on the yield of 12, However, addition of sodium chloride to the reac- tion mixture seemed to have little effect on the yield of 22. Thus, the mechanism by which 12 is formed is very much open to speculation. It is clear, however, that 12 contains six atoms of chlorine while Za and Zg contain only five. Thus, addition of one atom of chlorine, by whatever the mechanism, is required. Compound 12 has proven to be a very difficult compound in two respects. The structural analysis has been difficult, and although the available evidence tends to support the proposed structure, such a structural assignment is rather unsatisfying because of the presence of the chlorohydroxy function. It is felt, however, that 12 may be stable toward the loss of hydrogen chloride from this functionality because of the nature of the molecule. That is, loss of hydrogen 43 “ “‘0. 1+ ¢ “(an O n “*1 0 )6 fl 0H 5 OH 0 ‘+0\ 5 23 H + fl (3'3 I ‘03 fl 0 (fl Hf ¢on ¢ ¢ ” / If 0 "'""""" 0 \ ¢ :f 0 ¢ 0 ¢ ¢ 21 Scheme 10. Mechanistic scheme for the conversion of 2,3— dihydroxytetraphenylcyclopentenone 26 to 2-ben— zoyltriphenylfuran 21. :44 chloride would require the formation of a system of high strain and would therefore be a high energy process. This is evidenced by the fact that 12 must be heated to 130° before loss of hydrogen chloride occurs. That loss of hydrogen chloride may occur at lower temperature in the appropriate system is suggested by the fact that the pro- duct of ketalization of 12, i;e., 12 has no hydroxy func- tion and contains only five atoms of chlorine. The second major difficulty related to 12,13 that of suggesting a plausible mechanism for its formation. Here again, one has difficulty, if not failure, in suggesting an adequate and at the same time satisfactory mechanism. Suggestions concerning the mechanism are very Speculative, and the possibility certainly exists that some totally rea— sonable and sufficient mechanism for the formation of £2 which we have not thought of may be available. We feel, however, that the systems involved in this study are rather different, and that one might therefore expect their chemistry to be unique in some respects. The second major question to be discussed concerns the formation of the diacid 14; There appear to be several possible mechanisms leading to 14 which are worthy of con— sideration. The first to be considered will be those from the epoxide 26b. As noted previously, considering the usual reactions of epoxides in solution, one might predict the formation of an a~diketone such as 28 from the reaction of 45 £22: Such diketones are known to undergo Baeyer-Villiger oxidation with peroxide as well as with peracids to give diacids in aqueous media (36). One might then expect 24 to be formed by a mechanistic pathway like that shown in Scheme 11. It then seems quite likely that lg’could result from the further reaction of 262, There is also indirect evidence to suggest that 14 may be formed directly from Zalrather than by a pathway involving the conversion of 22 to 22; The epoxide of 1a, 143., 262/ if formed might be expected to react further to give 32 which could yield 14 as shown in Scheme 11. 0 o o 01,, 01,, 014 01 -m1 01 -——+ no “’"’ ° 0 BO 110 01 01 01 26a 11 30 W m It is therefore possible that 14,18 being formed from Ia directly and from 22 derived from 13. The indirect evidence which suggests that Za may be reacting directly to yield 14 is found in the reaction of IE, As was noted previously, only in the reaction of lg is a third product observed. This product was identified as the methyl half—ester of 14, and the fact that it is the methyl rather than the ethyl ester indicates that it is formed directly from 123 A possible mechanism for the formation 46 0 C14 C14 . . O O Baeyer-Villiger > Oxidation O HO Y 32 HO Y 33 32’ Y = Cl Hydrolysis ‘HY 0 C14 HO O HO HO Y Q2 1 my 0 0 Cl 0 > 12, ‘————- 0 HO O E 0 251 Y = -Cl. -OH, —OEt, —o—o—w03_ Scheme 11. Mechanistic proposal for the formation of diacid 14 from diketone 22: 47 of this product is shown in Scheme 12. Oxidation of 42 to the keto acid and its oxidation to 15 are quite reasonable reactions under the conditions (36). Two additional points concerning Scheme 12 should be noted. First, compound 41 could react to give 14 in the manner shown in Scheme 11. This reaction may be competing with that leading to the formation of léx and perhaps for this reason, the yield of 15 is low. Second, the reaction of Za’or Zbgin the manner shown in Scheme 12 would yield 24. Because of this, one is not able to say whether Za and ZS are reacting in this manner. Another possible reaction pathway leading to 14 is that shown in Scheme 13, which involves direct formation from lb rather than the intermediate epoxide 26b: This sequence involves Baeyer-Villiger oxidation of the S-diketo form of 12, Such compounds, $43., B—diketones, sometimes react abnormally under Baeyer—Villiger oxidation conditions (36); there might then be some question about this sequence. It is a possibility nonetheless. One additional consideration is the formation of £4 from 12 which has been shown to occur under the reaction conditions as noted previously. The mechanism suggested below seems possible. However, the attempted hydrolysis of 12 using 2§_hydrochloric acid afforded only starting material. Thus, one might question the hydrolysis of l§.t° £11. 48 O Cl4 O c14 C1 C1 hydrolysis: H O HO OCH3 OCH3 26c -HC1 O O H C14 C14 0 < O \ F— CH30 ,> H_' 00113 0 42 41 1. oxidation 2. hydrolysis 0 c1 HO 4 CH30 o 12 Scheme 12. Mechanistic proposal for the formation of the methyl half—ester 15 of diacid £4. 49 HO O Baeyer-villiger Oxidation 01 c14 43 hydrolysis 1. oxidation 2. hydrolysis HO 1:1, Scheme 13. Mechanistic proposal for the formation of diacid $4 from 2b. 50 r- _ 0 0 0 cl 014 c1 014 C1 c14 [ + H 1 [CH H [I H I I HO m __I ___> O H-O + . C1 /> HO c1 H-O c1 0 1:2», — 2:2 46 J o H c14 _ l J i6, L> O I l > 14 HO O 44 Obviously there are many possible routes to 13/ and those presented here are only suggestions. The reactants and reaction conditions used are such that mechanistic sug- gestions are at best speculative and at worst futile con- sidering the information at hand. It is therefore obvious that a great amount of research concerned with the study of the mechanisms of these reactions might be conducted. C. Conclusions One might conclude from the previous discussion that more questions have been posed than have been answered. This is indeed true, and perhaps not too surprising when one considers the reactants and reactions studied. Some rather unusual and little studied systems were subjected to 51 unusual reaction conditions; one might therefore expect some unusual results. Several major problems are involved in a study dealing with highly chlorinated organic compounds. Unlike the chemistry of polyfluoro organic compounds which has been and is being studied quite extensively, the chemistry of polychloro organic compounds has been studied much less ex- tensively. However, the chemistry of polychloro compounds, like that of polyfluoro compounds, is unique in that the reactions of such compounds are different from the reactions of "normal" organic compounds. Another major difficulty unique to the study of poly- chloro organic compounds is that of structural analysis. That such is the case is obvious fromtflfis study as well as from the fact that numerous examples of errors in struc- tural assignments may be found in the literature (9,23,24). No extremely valuable tool such as nuclear magnetic reso- nance is available for the structural analysis of polychloro organic compounds. A technique which does show great promise is that of nuclear quadropole resonance which is only now beginning to be used in the study of polychloro organic compounds (24,37,38). This has not yet become a routine analytical tool such as nmr. Therefore, struc- tural analysis of polychloro organic compounds at this time depends largely upon classical organic reactions which may take unusual and unexpected directions when applied to the unique systems involved. The obvious answer to the question 52 of the structure of lfi would be an X-ray analysis. There is some question whether such a study is warranted, however. Definitive conclusions concerning this study are dif— ficult to make. An unusual structure has been proposed for compound lgiwhich is apparently formed in a unique manner. Further, we have assigned a structure to compound 12 which had previously been assigned to compound having properties ’° very much different from those of 12.(27'28) As noted previously, there is a possibility that the structure we have assigned to 12 is in error. However, we feel that the available evidence supports the proposed structure and feel obliged to suggest it if for no other reason than that we cannot suggest a more satisfying struc~ ture which fits the available data. The discussion of mechanistic possibilities is totally speculative.' As such, it is quite open to criticism, further speculation and discussion, and hopefully, further study. Therefore, at this stage, any conclusions about mechanisms would be unwarranted. D. Suggestions for Further Study One obvious suggestion is that of a nuclear quadropole resonance study of a series of polychloro organic compounds of known structure such as Z and related compounds so that one might begin to build a fund of empirical data such as that available for the analysis of nmr and ir spectra. The usefulness of such data in the analysis of unknown structures is obvious. h;— 53 Further study concerning the mechanisms of the reac— tions involved in this study would be quite useful. The possibilities in this direction are many. A third suggestion for further study would be to ex— tend the reaction to related systems and to use different oxidation methods. Thus, it is obvious that this study has been quite pre— liminary and has suggested many new problems while solving few. Addendum: An alternative structure for compound 1%, suggested by Professor William Reusch, is that of the y-lactone shown below. Studies to distinguish between this structure and that proposed for compound ngin the text of this report are to be carried out. <214 l EXPERIMENTAL A. General Procedures Infrared spectra were recorded on a Perkin—Elmer Model 237B spectrophotometer. The nmr Spectra were obtained using a Varian A—6O instrument with chemical shifts reported as 6 values measured from an internal standard of tetramethyl- silane. The uv spectra were recorded on a Unicam Model SP-800 spectrophotometer using 1 cm quartz cells. Mass spectra were determined using a Hitachi Perkin-Elmer Model RMU—6 low resolution instrument. Melting points were determined on a Thomas Hoover melting point apparatus and are uncorrected. Microanalyses were performed by Spang Microanalytical Laboratory, Ann Arbor, Michigan or Galbraith Laboratories, Inc., Knoxville, Tennessee. B. Sodium Tungstate—Hydrogen Peroxide Oxidation of 2—Sub— stituted—1,3,5,6,7—Pentachlorobicyclo[3.2.0]hepta-2,6— dien—4—ones 7a, 7b, and 7c The same general procedure was used in the reaction of all three compounds. A description of the procedure used in a typical reaction is given below. 54 55 To a suspension of 10 mmol of the polychloro compound 12” Z?» or ZS (9) and 1.65 g (5 mmol) of sodium tungstate dihydrate in 25 ml of absolute ethanol was added 20 ml of 30% hydrogen peroxide over a 15 min period. The suspension was then stirred at room temperature for 45 min and at re— flux for 2.5 hr. The solvent was then removed under vacuum and the residue chromatographed directly or extracted thoroughly with ether. In the latter case, the extracts were washed once with saturated sodium chloride solution, dried over magnesium sulfate, and the solvent removed. The oil which remained was then chromatographed. In either case, chromatography was on a column of 70 g of silicic acid. Elution with chloroform yielded only starting material when 13 or Zg were the reactants. Using the more polar carbon tetrachloride—ethyl acetate (10:3) solvent system, IQ, lg” and 15 (IE only from the reaction of ZS) were iso— lated. Using the still more polar carbon tetrachloride— methanol (10:4) solvent system, the diacid 13 was isolated. Yields of the various products are given in Table 1. Recrystallization of product 12 from carbon tetra— chloride provided an analytical sample of colorless crystals: mp 129—1300; uv (methanol) absorption below 225 mu; ir (KBr) Figure 4; nmr (acetone—d6) 6 6.52 (s); nmr @cetonitrile) 5 6.25 (s, 1) and 5 5.20 (broad s, 1); absorption at 6 5.20 disappears after deuterium exchange; mass spectrum Table 7. 56 Anal. Calcd for C7H2C1603: C, 24.24; H, 0.58; Cl, 61.34. Found: C, 24.10; H, 0.63; Cl, 61.35. C, 24.02; H, 0.56; Cl, 61.50. Recrystallization of product lé’from benzene afforded an analyticalsample as white crystalline flakes: mp 156.5- 157.5°; uv (H20) absorption below 225 mu; ir (KBr) Figure 7. Anal. Calcd for C6H2Cl404: c, 25.75; H, 0.72; c1, 50.67. Found: C, 25.57; H, 0.72; Cl, 50.55. Recrystallization of product 15 two times from carbon tetra- chloride—methylene chloride and careful drying afforded an analytical sample of fine white powder: mp 93.5—95.0°; ir (KBr) Figure 11, nmr (CDCla) 5 11.07 (s, 1) and - 3.87 (s, 3). Anal. Calcd for C7H4O4Cl4~H20: C, 29.95; H, 1.94; Cl, 45.46. Found: C, 27.01; H, 1.87; C1, 45.50. C. Preparation of the Acetate $1.0f Compound lg A reaction mixture consisting of 175 mg (0.5 mmol) of £1, 2 ml of acetic anhydride, and a few drops of acetyl chloride was stirred at 80° for 6 hr. The clear yellow solu— tion was then allowed to stand overnight at room temperature. Chromatography of the yellow—brown oil, which remained after removal of the solvent under vacuum,on a column of 5 g of silicic acid using chloroform as the eluent afforded an analytical sample of 11 as small white crystals: mp 183- 184°; ir (KBr) Figure 5) nmr (CDCls) 5 6.00 (s, 1) and a 2.25 (s, 3); mass spectrum, Table 8. 57 Anal. Calcd for C9H4O4C16: c, 27.30; H, 1.04; cl, 54.70. Found: C, 27.88; H, 1.04; Cl, 54.80. D. The Ketalization Reaction of Compound 12 A reaction mixture composed of 347 mg (1 mmol) of 12” 93 mg (1.5 mmol) of ethylene glycol, and a few crystals of p—toluenesulfonic acid in 20 ml of benzene was refluxed for 24 hr using a Dean-Stark apparatus for azeotropic removal of the water formed in the reaction. The benzene solvent was distilled off under vacuum leaving an oily residue which was chromatographed on a column of 10 g of silicic acid. Elution with chloroform yielded 165 mg (46%) of an oil which crystallized on standing for a few days. Recrys— tallization from carbon tetrachloride—hexane afforded an analytical sample of 1§ as white crystals: mp 158—159°; ir (KBr), Figure 6; nmr (c0c13) a 6.77 (s, 1) and a 4.0 to 6 5.0 (bm, 4); mass spectrum, Table 9. Anal. Calcd for C9H5O4Cl5: C, 30.50; H, 1.42; Cl, 50.02. Found: C, 30.65; H, 1.53; Cl, 50.07. E. Sodium Tungstate—Hydrogen Peroxide Oxidation of Com— pound 13 _.____._m To a suspension of 694 mg (2 mmol) of product 13 and 330 mg (1 mmol) of sodium tungstate dihydrate in 4 ml of absolute ethanol was added 4 ml of 30% hydrogen peroxide over a period of 15 min. The suspension was stirred at room temperature for 1 hr and at reflux for 2.5 hr. The 58 solvent was removed under vacuum and the residue chromato~ graphed on a column of 14 g‘of silicic acid using in the order listed, chloroform, carbon tetrachloride—ethyl ace— tate (10:3), and carbon tetrachloride—methanol (10:4) as eluents. Two products were recovered and identified as the starting compound (45%) and tetrachloro-3—cyclobutene— 1,2—dicarboxylic acid, 14” (30%) by comparison of melting points and spectra with authentic samples. The diacid product was also converted to its dimethyl ester as described elsewhere. This product was in all respects identical to the diester prepared previously. F. Attempted Oxidations of Compound 13 1. Oxidation Using Jones Reagent (18). To a solution of 347 mg (1 mmol) of 13 in 6 ml of acetone cooled in an ice bath was added 0.5 ml of Chromic acid solution prepared by dissolving 2.0 g (0.02 mol) of chromium trioxide in 5 ml of water, adding 3.0 g (0.03 mol) of concentrated sulfuric acid, and diluting to 10 ml. This solution was allowed to warm to room temperature over a 1 hr period and then stirred at room temperature for 3.5 hr. After addition of 2 ml of isopropyl alcohol and 210 mg of sodium bicarbonate, the mixture was filtered. The filter cake was washed thoroughly with acetone. Removal of the acetone from the combined filtrates yielded a green oil which on chromatography on a column of 12 g of silicic acid afforded 84% of the starting compound 13; 59 Repetition of the reaction as above yielded only the starting compound. 2. Oxidation with chromium Trioxide in Acetic Acid (10f). A solution of 347 mg (1 mmol) of l§.in 3 ml of glacial acetic acid was added to a solution of 98 mg (0.98 mmol) of chromium trioxide in 5 ml of glacial acetic acid. The re- action mixture was stirred 8 hr at room temperature. Evaporation of the acetic acid was accomplished by passing a stream of air over the reaction mixture. The residue was taken up in 20 ml of water which was then extracted thoroughly with ether. Removal of the ether after drying yielded only a crystalline material identified as 13. 3. Oxidation with Chromic Acid. A solution of 150 mg (0.5 mmol) of potassium dichromate in 2 ml of water and 1 ml of concentrated sulfuric acid was added to a suspension of 174 mg (0.5 mmol) of 13 in 1 ml of water. After stirring at room temperature for 2.5 hr, the reaction mixture was extracted thoroughly with ether. Re- moval of the solvent after drying over magnesium sulfate, yielded an oil which crystallized on standing. This material was identified as 13, G- The Attempted Hydrolysis of 13 A solution of 278 mg (0.8 mmol) of l§,in 3 ml of tetra— hydrofuran and 3 ml of 23 hydrochloric acid was stirred at 60 room temperature for 2 hr and at reflux for 6 hr. One hour after the start of reflux an additional 1 ml of THF was added. On cooling, the solution separated into two phases. The lower aqueous phase was separated from the upper organic phase and extracted with ether. The extracts were combined with saturated sodium chloride solution, dried over magnes— ium sulfate, and the solvent removed. Recrystallization of the residue afforded only starting material. H. The Thermolysis of 13 A sample of 175 mg (0.5 mmol) of 13 in an evacuated sealed tube was heated to 120—1300 for 0.5 hr. The yellow- red oil which resulted was chromatographed on a column of 6 g of silicic acid. Elution with chloroform gave 123 mg of red-orange oil. Distillation of this oil gave a golden yellow oil. This material was not identified. A very puz- zling fact is that the infrared Spectrum of the distilled sample contained several absorption bands in the region 3000 to 2900 cm—1 which is caracteristic of carbon—hydrogen ab~ sorptions (25). This absorption was not present in the ir of the sample before distillation. I. The Determination of Hydrogen Chloride as a Product of the Thermolysis of 13 A piece of moist pH paper gives an indication of acidity when placed over a capillary tube containing 13 which is heated to 130°. 61 A test for halide ion which involves the use of a test paper containing silver ferrocyanide (17) which is moistened with a ferric sulfate solution was used to determine whether halide is evolved when 13 is melted. A positive test is indicated by a blue color ascribed to the formation of Prussian blue according to the formula shown below. _ Fe SO - Ag4[Fe(CN)6] + x —3-(—4—)3> 4AgX +Fe[Fe(CN)6] A procedure like the above using the appropriate test paper gave a positive result. One may thus conclude that hydrogen chloride is evolved when 12,15 heated to 130°. J. Test for Enol Formation in 13 The reactions involved in this test have been discussed in the text of this report (19). A few drops of bromine water was added to an ethanol solution of 13. After a few minutes, a saturated solution of sulfosalicylic acid was added to the above solution until the yellow—orange color disappeared. A few drops of 5% potassium iodide solution and starch solution were then added. After a few minutes, the solution took on a purple Color which became more intense on standing. A control experiment like the above except that 13 was not added to the ethanol did not become colored on standing overnight. Similarly, a control in which no bromine was added to the ethanol solution of £2 did not become colored. 62 These results suggest that an a—bromoketone is formed from 13 on treatment with bromine. However, as noted in the text, such a compound has not yet been isolated. K. Hydrogenation of Tetrachloro—3—cyclobutene—cis—1,2- dicarboxylic Acid £4 The hydrogenation of 570 mg (2 mmol) of the diacid £4 in 75 ml of water using 50 mg of platinum oxide catalyst was carried out at room temperature and atmospheric pres- sure. The reaction was stopped after the system had taken up 260 ml (12 mmol) of hydrogen. Removal of the water under vacuum afforded a crystalline material the infrared spec- trum of which was identical to that of gig—cyclobutane—1,2— dicarboxylic acid. The material melted over a wide range. Attempted recrystallizations were not successful. Chrom— atography using carbon tetrachloride—methanol (10:4) eluent afforded a white crystalline material with properties identical to those of cis—cyclobutane-l,2-dicarboxylic acid (27:39); ir (KBr) Figure 10. L. Esterification of Tetrachloro-3~cyclobutene—cis—1,2— dicarboxylic Acid {4 Hydrogen chloride gas was bubbled into a cooled solu— tion of 280 mg (1 mmol) of diacid 14 in 10 ml of anhydrous methanol. The flask was stoppered with a drying tube con— taining calcium chloride and the solution stirred at room temperature for 18 hr. On removal of the solvent, a white 63 crystalline material was present. This material chromato— graphed on a column of 5 g of silicic acid using chloroform eluent gave 234 mg (76%) of white crystalline diester 22 mp 66—680. Recrystallization from 95% ethanol—water af— forded an analytical sample as fine white needles: mp 68.5—69.207 ir (KBr) Figure 9; uv max (heptane) 212 mu (6 = 9500); nmr (CDCla) 0 3.75 (s). Afléi- Calcd for C8H6O4Cl4: C, 31.20; H, 1.96; Cl, 46.05. Found: C, 31.25; H, 1.95; Cl, 46.15. In subsequent preparations of the ester, it was found that chromatography is not necessary. The crude product recovered after removal of the methanol may be recrystallized directly. M. Preparation of the Anhydride 21 of Tetrachloro—B—cyclo— butene—gis-l,2—dicarboxylic Acid £4 A solution of 280 mg (1 mmol) of diacid 14 and 4 ml of acetyl chloride was refluxed for 2 hr. The flask was then Stoppered and stored in a refrigerator for 2 days. On re— moval of the solvent under vacuum, a crystalline solid was left. Purification by sublimation two times afforded 135 mg (51%) of analytical sample: mp 137-138.5°; ir (KBr) Figure 5. Anal. Calcd for c603c14: c, 27.52; H, 0.00: Cl, 54-15- Found: C, 27.36; H, 0.15; Cl, 54.07 64 N. Esterification of Anhydride 21 A solution of 32 mg of the anhydride in 3 ml of anhy— drous methanol saturated with hydrogen chloride gas was stirred at room temperature for 4 hr. Removal of the sol— vent and chromatography of the residue gave 5 mg of white crystals identified as the dimethyl ester of tetrachloro— 3-cyclObutene—1,2—dicarboxylic acid by comparison of its infrared spectrum and melting points with those of an authentic sample prepared from the diacid £4. 0. Esterification of the Methyl Half-Ester 15 of Tetra— chloro—3-cyclobutene—cis-1,2-dicarboxylic Acid 14 A solution of 420 mg of the methyl half-ester of tetra— chloro—3-cyclobutene-1,2—dicarboxylic acid 15” which was used without purification other than chromatography, in 15 ml of methanol saturated with hydrOgen chloride was stirred at room temperature for 24 hr. The solvent was removed, the residue extracted with ether, the extracts dried over magnesium sulfate, and the ether removed. Chro- matography of the residue on a column of 12 g of silicic acid using chloroform eluent afforded 240 mg of crystalline material mp 66-680. Recrystallization from 95% ethanol— water yielded fine white crystals identified as the dimethyl ester of tetrachloro—3—cyclobutene—1,2—dicarboxylic acid 22 by comparison of melting point and infrared spectrum. 65 P. Hydrolysis Reactions of 13, 72/ and ZS with Sodium Tungstate Present The same general procedure was used for all three sys— tems. To a suspension of 5 mmol of the 2-substituted—1,3,5, 6,7—pentachlorobicyclo[3.2.0]hepta—2,6—dien-4—one 22” 1b, or Igland 0.83 g (2.5 mmol) of sodium tungstate dihydrate in 10 ml of absolute ethanol, 7 ml of water was added over a period of 5 min. The white suspension was stirred at room temperature for 15 min and at reflux for 2.5 hr. The solvent was then removed under vacuum and the residue chromatographed on a column of 40 g of silicic acid. The solvents used in elution were chloroform, carbon tetra— chloride—ethylacetate (10:3), and carbon tetrachloride— methanol (10:4) in that order. The results are shown in tabular form below. Table 4. Product yields from the hydrolysis of la, 1b, and 1 7c. Starting % Starting % % Material Material 2-Hydroxy-(Zb) Sodium Salt lg Recovered Recovered ZS (2—Chloro—) 32 47 0 12 (2—HydrOXy—) 16 16 55 lg (2—Methoxy~) 0 22 60 66 Q. Hydrolysis Reactions of thand ZEIUsing Only Ethanol and‘Water The same procedure was used in the reaction of Za’and 7c. “A, To a suspension of 5 mmol of the 2—substituted-1,3,5, 6,7—pentachlorobicyclo[3.2.0]hepta—2,6—dien-4—one Zg’or 2E, in 10 ml of absolute ethanol, 7 ml of water was added over a 5 min period. The suspension was then stirred at room temperature for 35 min and at reflux for 2.5 hr. The sol— vent was then removed under vacuum and the residue chromato- graphed on a column of 40 g of silicic acid using progres— sively more polar eluents starting with chloroform, then carbon tetrachloride—methyl acetate (10:3) and finally carbon tetrachloride—methanol (10:4). The results are given in the table below. Table 5. Product yields from the hydrolysis of Z3 and 13 with ethanol and water. Starting % Starting % Material Material 24Hydroxy— (ER) Recovered 72 (2—Chloro-) 38 38 22(2Mmmmq—) O 92 67 R. Reaction of 2-Hydroxy—1,3,5,6,7:pentachlorobicyclo— [3.2.0]hepta-2,6-dien-4—one zg’with Sodium Carbonate A suspension consisting of 1.47 g (5 mmol) of Zb and 0.27 g (2.5 mmol) of sodium carbonate in 10 ml of absolute ethanol to which 7 ml of water had been added over a period of 5 min, was stirred at room temperature for 15 min and at reflux for 2.5 hr. The solvent was then removed under vacuum and the residue recrystallized from ethanol—benzene. A nearly quantitative yield of product 16 as fine white needles resulted: dec. «800°; ir (KBr) Figure 12. 1. Acidification of Product £6 Hydrogen chloride gas was bubbled through a methanol solution of the above product for approximately 3 minutes Almost immediately, large quantities of white solid formed. The reaction mixture was filtered and the solvent removed from the filtrate. A white crystalline product shown to be the starting hydroxy compound 22 was isolated in 82% yield based on 72 used in the initial reaction. PART II THE REACTION OF TETRACHLOROCYCLOPROPENE WITH POLYCHLOROETHYLENES 68 INTRODUCTION The bicyclic systems studied in Part I of this thesis are derived from 6—H—Nonachlorobicyclo[3.2.0]hept—2—ene 41 which is the product of the novel reaction of hexachloro— cyclopentadiene with trichloroethylene in the presence of anhydrous aluminum chloride at 80 to 110° (39,9). 01 6 ”9 a a) + 0120:0301 __‘_1£11__, :11 No mechanistic details concerning this reaction have been published. Speculation might suggest that possibly the pentachlorocyclopentadienyl cation 48 is involved in the reaction (9a). Evidence for the possible existence of this species is presented in the literature as noted below (40,41). 01 01 Cl 01 Cl 2?. 69 70 In the mass spectrum of hexachlorocyclopentadiene, the most abundant ion may be assigned the formula C5Cl5+ (41) which according to the rule of rings plus double bonds (16) would correspond to a system containing three rings and/or double bonds. The authors suggest that this ion is indeed the pentachlorocyclopentadienyl cation 48 which fits the classification noted above in that it has the required three rings and/or double bonds. Spectral evidence that the pentachlorocyclOpentadienyl cation 48 exists as a triplet in the ground state has been published by Breslow and coworkers (42). Schafer and Fritz (43) suggest structure 42 shown below rather than a symmetrical pentachlorocyclopentadienyl cation structure for a red resinous material isolated from the reaction of hexachlorocyclopentadiene with aluminum chloride. 2.9. There appears to be some justification for considering the intermediacy of a species with some cationic Character. The red-purple color reported by Breslow (42) and by Schafer and Fritz (43) and observed in the reaction of hexachloro— pentadiene with trichloroethylene suggests the possibility Of an intermediate possessing some cationic character. 71 A mechanism involving 48 or some related species such as 49 in an electrophilic reaction with trichloroethylene and subsequent cyclization to give product 41 might be sug— gested in view of the above information. The thought that perhaps the analogous reaction using tetrachlorocyclopropene rather than hexachlorocyclopenta— diene under the same reaction conditions might afford some information concerning the mechanism of the reaction led to the study reported herein. Tetrachlorocyclopropene 50 reacts exothermically with anhydrous aluminum chloride to form trichlorocyclopropenium tetrachloroaluminate Ei,(44)' 014 01 + 11015 H @ 11014 01 c1 529. E}. Considering this favorable reaction to form the aromatic trichlorocyclopropenium ion 51,0ne might suspect that 51 would be the reactive species involved in an electrophilic reaction with trichloroethylene. If the product formed in this reaction should prove to be the bicyclic analogue of 41, one might then have some slight additional support for the involvement of the penta- chlorocyclopentadienyl cation 48 in reaction a. Great care would need to be exercised in comparing results from 72 the two reactions, for the two cationic species 48 and EL are very different. The pentachlorocyclopentadienyl cation 48 is at best predicted to be an unstable system while the trichlorocyclopropenium ion 51 has been shown to be a stable species which exhibits properties characteristic of an aromatic system (44). For the above reasons and with the realization that even if a product analogous to 41 should result from the reaction of tetrachlorocyclopropene 52 with trichloro— ethylene in the presence of aluminum chloride great care in drawing conclusions would be necessary, the reaction of tetrachlorocyclopropene with trichloroethylene was studied. Since the results of this study were rather interest- ing, the study was extended to include the reaction of tetra— chlorocyclopropene with 1,2—dichloroethylene (gig and trans) and with tetrachloroethylene. After the work was in progress, a study concerning the reaction of tetrachlorocyclopropene with 1,1—dibromo—, 1,1—dichloro~, and 1,1—dimethylethylene was reported (45). RESULTS AND DISCUSSION A. The Reaction of Tetrachlorocyclopropene with Trichloro— ethylene Chromatography of an oily material isolated from the . reaction of tetrachlorocyclopropene with trichloroethylene in the presence of anhydrous aluminum chloride afforded, in addition to an intractable yellow—orange oil, a crystal— line product 52 (mp 151—1520). The appearance of the infrared spectrum of 52 with ab— sorption bands at 1905 (m), 1865 (s), and 1825 (s) cm~1 immediately suggested the possibility that compound 52 might be a cyclopropenone since cyclopropenones characteristically show absorption in this region (46}. The mass spectrum and elemental analysis of 52 showed it to be a compound of ’ formula C7OC16. From the mass spectrum, it appears that a facile cleavage of carbon monoxide occurs as might be ex— pected for a cyclopropenone (46a,47). On melting, 52 was observed to evolve gas. The red liquid which resulted solidified on cooling to room temperature. When a sample of compound 52 was heated in an evacuated sealed tube to 150° for 15 minutes, a new orange crystalline compound 53 (mp 52.0-52.5°) was isolated. Analysis showed this compound to have an empirical formula of CGClG' The 73 74 mass spectra of leand 53 are very similar in the region corresponding to C6C16, $43., m/e = 282, and below. This suggests that leis formed by the loss of carbon monoxide from 52; The infrared spectrum of 53 contains very few bands. Of major significance,however, is a weak band at 2190 cm—1 which may be attributed to a disubstituted acetylenic bond (25). There is no infrared absorption near 756 cm"1 which is the region of EC-Cl absorption (48). The above evidence suggests that 52 is a bis(trichloro— vinyl)cyclopropenone and that 53 is bis(trichlorovinyl)- acetylene formed by the loss of carbon monoxide from 52’ a reaction which is general for cyclopropenones (46a, 46c, 47). O C / \ C12C=CCl-C ;: C-CClzcclz __1. c12c=cc1—csc-Cc1=cc12 2____ 5,22 The ultraviolet spectra of 52 and §E.Will be discussed later. Suggesting a possible mechanism for the formation of cycloprOpenone Qg'from the reaction of tetrachlorocyclopro- pene with trichloroethylene is not too difficult when one considers some other examples of reactions of 75 tetrachlorocyclopropene, and the fact that the product of the reaction is a cyclopropenone is not too surprising. Several examples of electrophilic substitution reac— tions of trichlorocyclopropenium tetrachloroaluminate 51 with benzene and substituted benzenes to give aryltrichloro- cyclopropenes and diarylcyclopropenones have been reported (49,50). Thus, one might conSider the possibility of a similar electrophilic substitution reaction with trichloro- ethylene. However, trichloroethylene might be expected to be a rather poor system to enter into an electrophilic substitution reaction as the electron rich partner because of the presence of three electronegative chlorine atoms. Also, since 51 is an aromatic species, its reactivity as an electrophile might be greatly decreased. The results suggest that a favorable combination of factors is present in the above reaction and that the tri- chlorocyclopropenium ion does apparently undergo an electro— philic substitution reaction with trichloroethylene. Per— haps fil’is after all a more reactive electrophile than might be anticipated because it is in fact a cation in addition to being an aromatic Species. At any rate, the following mechanism may be suggested. (Scheme 14) There is a question whether the product of step g is vi or vii. Precedence for the formation of 1,1—dichloro compounds from chlorocyclopropenium (44) and chlorotropylium salts (51) on hydrolysis may be found in the literature. Also, in a few instances, compounds such as vi have been 76 C1 Cl 8) + Cl2C=CHCl > 69 c1 c1 c1 CHCl-CClz C1 C1 + Cl 1” <9 _H > A Cl CHCl-CClz Cl CCl=CC12 Cl Cl i Cl c) i + AlC13 ———-—> E AlCl4e ii Cl d) " G 11 + C12C=CHCl > ClZC-CHCl cc1=cc12 - Cl iii C1 + e) iii -H > ClzC:CCl‘ACC]—=CC12 Cl iv Cl f) iv + AlCl3 > Cl2C=CCl CC1=CC12 v C12 9) V + H20 > A Cl2C=CCl cc1=0012 vi or + Cl 0H2 C12C=CC1 CC1:CC12 vii M iv + H20 > £2 <1 HF- vii aim Scheme 14. Mechanistic proposal for the formation of bis— — S trichlorovinyl )cyclopropenone 5,2". 'I 77 isolated (44,45). Thus, one might be inclined to favor the pathway involving hydrolysis of vi. B. The Reaction of Tetrachlorocyclopropene with 1,2—Di— chloroethylenes The reaction of a mixture of gig: and E£§n§71,2-di— chloroethylene with tetrachlorocyclopropene in the presence of anhydrous aluminum chloride gave similar but much less satisfactory results than the reaction with trichloroethylene. A maximum of 16 per cent yield of crystalline bis(1,2-di- chlorovinyl)cyclopropenone Qé’was isolated. The analytical 'and spectral data support the cyclopropenone structure for 54, No evidence on which to base the stereochemistry of 54 is available. ClCH=CCl*C = C-CCl=CHCl 54 I'W When trgngfl,2—dichloroethylene, bp 49—520, was used in the reaction, a five per cent yield of 54 and a seven per cent yield of a,B-dichloroacrylic acid resulted. Using gigfl,2-dichloroethylene, bp 58-600, yields of 17 per cent of the cyclopropenone 54 and 11 per cent of the acrylic acid were isolated. The differences in yield may or may not be significant since the reaction conditions were some— what different for the various reactions. What is signifi— cant is that at best a yield of less than 20 per cent of 78 cyclopropenone éé’was isolated. An effort to maximize yields was not put forth, but several reactions under vari- ous conditions gave yields which did not vary greatly. There is some evidence to suggest that possibly the lower boiling points of the 1,2—dichloroethylenes, and hence the lower refluxing temperatures of the reaction mixtures as compared to that of trichloroethylene, may be significant. Table 6. Yields of cyclopropenones from the various dichloroethylenes. Reflux % Yield of Polychloroethylene bP (Eigfi prgggigge 1. tggngfl,2—Dichloroethylened 49-520a 8.0 5 2. gig-1,2-dichloroethylened 58—60Oa 5.0 17 3. Mixture of 1 and 2d 13.5 12 4. Mixture of 1 and 2e 10.0 16 5. 1,1,2-Trichloroethylenee 87Ob 5.0 47 a . . . . . . . BOiling pOints of isomers as observed on distillation of commercial mixture of cis— and trans—1,2—dichloroethylene. bBoiling point recorded on label. CData taken from reactions giving highest yields. Data for 1,2, and 3 from reactions using a ratio of tetra— chlorocyclopropene to aluminum chloride of 10 to 7.5. e . . Data for 4 and 5 from reactions u51ng a ratio of tetra— chlorocyclopropene to aluminum chloride of 4 to 1 and 10 to 3 respectively. Considering the data in Table 6, one might suggest that reaction temperature is of significance although the data 79 in Table 6 are by no means conclusive on this point. A very significant factor is the amount of aluminum chloride used. For example, in the reaction of trichloro- ethylene with tetrachlorocyclopropene, when the ratio of tetrachlorocyclopropene to aluminum chloride is ten to three, the yield of cyclopropene is approximately 50 per cent. If however, the ratio is ten to seven and one-half, the yield of cyclopropenone is less than five per cent. In both in— stances, a significant amount of red—orange intractable oil is present. This is thought to be a polymeric material. There is evidence presented in the literature which shows that high polymers are formed in the reactions of polychloro compounds catalyzed by aluminum chloride when significantly more than catalytic amounts of aluminum chloride are used (52,53). A study of the effect of excess aluminum chloride on the reactions using the 1,2—dichloroethylenes was much less conclusive as the data in Table 6 show. In the reactions in which a ratio of tetrachlorocyclopropene to aluminum chloride of ten to seven and one—half was used, a large amount of intractable oil resulted, and the maximum yield of 54 was 12 per cent. In a reaction using a ratio of tetrachloro- Cyclopropene to aluminum chloride of four to one, the yield 0f ngwas 16 per cent. These results are for reactions in which the dichloroethylene used was a mixture of the gig and Egagg isomers. It may be noted from the data in Table 6 that the greatest yield of 54 results when the cis isomer 80 of 1,2-dichloroethylene is used. Perhaps a reaction using the gig isomer and a smaller amount of aluminum chloride relative to the amount of tetrachlorocyclopropene used might afford a significantly greater yield of 24. Attempts to prepare bis(l,2-dichlorovinyl)acetylene 55 from 54 were not successful due to the small quantities of Q4 available. There seems to be no reason why 55 could not be prepared in a manner analogous to the preparation of 53’ from 52, On melting compound 54, a gas is evolved and a red liquid is formed which crystallizes on cooling. This suggests that 55 probably is formed from 54 on melting. It thus appears that the reactions of tetrachlorocyclo— propene with gig; and gigggfl,2—dichloroethylene and 1,1,2— trichloroethylene are quite analogous. A mechanism similar to that proposed for the reaction of trichloroethylene may be written for the reactions of 1,2—dichloroethylene. In all cases, it is necessary, if the proposed mechanism is to hold, that the polychloroethylene system contain at least one hydro— gen atom which may be lost as a proton. Considering this factor, it became of interest to see what course the reaction might take if no hydrogen atom were present in the polychloro— ethylene system. For this reason, the reaction of tetra— chlorocyclopropene with tetrachloroethylene in the presence of anhydrous aluminum chloride was included in this study. 81 c. The Reaction of Tetrachlorocyclgpropene with Tetra- chloroethylene The only product isolated from the reaction of tetra- chlorocyclopropene with tetrachloroethylene in the presence of anhydrous aluminum chloride was a,B—dichloroacrylic acid. This product was also isolated from the reactions in which trichloroethylene and the 1,2—dichloroethylenes were used. It is believed that the acrylic acid results from the hy— drolysis of tetrachlorocyclopropene leor trichlorocyclo- propenium tetrachloroaluminate ii. The work-up procedure for all the reactions studied includes either pouring the reaction mixture on ice and ex- tracting with ether or methylene chloride and then washing thoroughly with water or extracting the reaction mixture directly then washing the extracts with water. Using either procedure, there is ample opportunity for hydrolysis of any unreacted tetrachlorocyclopropene 52 or trichlorocyclo— propenium tetrachloroaluminate 51; The product of the aqueous hydrolysis of 50,has been reported to be a,fi—di— chloroacrylic acid (46a,54), and on pouring 5} into water, the product is reported to be 52,(44). Thus it seems likely that the acid may be coming from either one or both 52 and 2.1; A question remains as to why no product from the addi- tion of Qi to tetrachloroethylene is isolated. It would appear that electronic factors are more important than steric factors since the double bond in tetrachloroethylene 82 should not be greatly hindered. Also, since Qi’is an aromatic species, it might be unreactive toward the appar- ently quite electron poor tetrachloroethylene. This seems unlikely, however, in View of the reactivity of El as an electrophile in the previous reactions studied. One might then conclude that the reason no apparent reaction between 5i and tetrachloroethylene is observed is due rather to the unreactivity of tetrachloroethylene as an electron source. D. The Reaction of Tetrachlorocyclopropene with 1,1—Di- bromo-, igl—Dichloro—, and 1,1-Dimethylethyleneg(4§) The results of this study as described in the Abstracts of the Division of Organic Chemistry from the 156th National Meeting of The American Chemical Society, September 1968, are analogous to those from our study. Divinyl cyclopropenones are reported to be isolated in yields of 20 to 60 per cent. The ultraviolet Spectra are reported to be characteristic of triene systems with maxima near 300 mu. The ultraviolet spectra of bis(trichlorovinyl)cyclo— propenone 221 bis(trichlorovinyl)acetylene 53/ and bis(1,2- dichlorovinyl)cyclopropenone 54 show evidence for extended conjugation with maxima above 300 mu and extinction coeffici- ents of approximately 20,000. The report of Tobey, Whittemore, and Lourandos (45) does not make reference to any of the acetylenes which might result from loss of carbon monoxide from the cyclopropenones such as bis(trichlorovinyl)acetylene 53 prepared from bis— (trichlorovinyl)cyclopropenone inby us. 83 E. Conclusions The results of this study provide no information con- cerning the mechanism of the reaction between hexadhloro— cyclopentadiene and trichloroethylene. Some unusual di— vinyl cyclopropenones did,however, result. Also, the scope of the participation of the trichlorocyclopropenium ion 5i in electrophilic substitution reactions was extended. As noted previously, electrophilic substitution reac- tions between 5i and aryl systems are well known (49,50). This work and that of Tobey, Whittemore, and Lourandos (45) show that the trichlorocyclopropenium ion is capable of reacting as an electrophile with systems which might be considered to be rather poor systems to be involved in electrophilic substitution reactions as the electron rich member. This may not in fact be too surprising since ii is after all a cationic Species. It seems, however, that the limit to the electrophilic reactivity of 51 with elec- tron poor ethylenes is reached in the reaction with tetra— chloroethylene since no addition or substitution product is observed. EXPERIMENTAL A. The Reaction of Tetrachlorocyclopropene with Trichloro— ethylene 1. The Reaction Using a 10 to 3 Ratio of Tetrachloro— cyglopiopene and Aluminum Chloride. To a stirred suspension of 0.4 g (3 mmol) of anhydrous aluminum chloride in 10.0 g of trichloroethylene was added 1.8 g (10 mmol) of tetrachlorocyclopropene. After stirring at room temperature for one hour and at reflux for five hours under a blanket of nitrogen, the reaction mixture, which contained a red oil, was allowed to cool to room temperature. On cooling, a tan solid formed. The reaction mixture was extracted with methylene chloride and the ex— tracts washed thoroughly with water and with saturated sodium chloride solution. The extracts were then filtered through magnesium sulfate and concentrated. The yellow oil which resulted was chromatographed on a column of 40 g of silicic acid using chloroform as the eluent. The chromatography yielded, in addition to 1.44 g of yellow—orange oil, 1.40 g (47%) of white crystalline bis(trichlorovinyl)cyclopropenone 52” Recrystallization from chloroform-pentane afforded an analytical sample as colorless plates: mp 151—1520; ir (KBr) Figure 13; uv (heptane) Figure 16; mass spectrum (70 eV) 84 1% 85 + mge (relative intensity) M‘, 316 (1.0), 314 (2.0), 312 (2.5), + 310 (1.5); (M-28)’, 290 (9.2), 288 (33.7), 286 (78.7), 284 (97.5), 282 (50.2); (M-63)+, 255 (4.0), 253 (21.2), 251 (63.7), 249 (100.0), 247 (63.0). Anal. Calcd for 070016: c, 26.88; H, 0.00; Cl, 68.00. Found: C, 26.84; H, trace; Cl, 67.92. 2. The Reaction Using a 10.0 to 7.5 Ratio of Tetra— chlorocyclopropene and Aluminum Chloride A mixture of 1.0 g (7.5 mmol) of anhydrous aluminum chloride, 1.8 g (10 mmol) of tetrachlorocyclopropene and 10.0 g of trichloroethylene was stirred at room temperature for one hour and refluxed for six hours. The system was allowed to react under a blanket of nitrogen. A tan solid formed as the reaction mixture cooled. Work-up as above yielded, after chromatography, 2.82 g of yellow-orange oil and 0.05 g (1.5%) of the cyclopropenone 523 In addition, on using a more polar eluent of carbon tetrachloride and ethyl acetate in a ratio of ten to three, 0.03 g (2%) of c.8-dichloroacrylic acid was isolated. B. The Thermolysis of Bis(trichlorovinyl)cyclopropenone 52 In an evacuated sealed tube, 300 mg (1 mmol) of his- (trichlorovinyl)cyclopropenone 52 was heated to 155-160° for 15 minutes. Chromatography of the red-orange solid, which formed when the tube was allowed to cool, on a column of 10 g of silicic acid using chloroform as the eluent gave 257 mg (94%) of bis(trichlorovinyl)acetylene 53 as red-orange 86 crystals. Recrystallization from pentane at —15° gave an analytical sample of 53 as fine orange-pink needles: mp 52.0-52.5°; ir (KBr) Figure 14; uv (heptane) Figure 17; mass spectrum (70 ev) gig (relative intensity) MT, 290 (3.8), 288 (13.3), 286 (30.0), 284 (38.1), 282 (19.5);(M—35)+, 255 (3.3), 253 (14.3), 251 (41.4), 249 (68.5), 247 (41.8); 142 (100.0). 3.23;. Calcd for 06016; c, 25.31; H, 0.00; Cl, 74.69. Found: C, 25.31; H, O ; Cl, 74.66. C. The Reactions of TetrachlorocycloprOpene with 1,2-Di- chloroethylenes 1. The Reaction Using a Mixture of cis— and trans—1,2- Dichloroethykxmeand a 10.0 to 7.5 Ratio Of Tetra— chlorocyclopropene and Aluminum Chloride. A suspension of 1.0 g (7.5 mmol) of anhydrous aluminum chloride in 1.8 g (10 mmol) of tetrachlorocyclopropene and 10 ml of 1,2—dichloroethylene (a mixture of gig and EEEEE. isomers) was stirred at room temperature under nitrogen for three hours and at reflux for 13.5 hours. The excess solvent was evaporated by passing a stream of air over the reaction mixture. The light brown solid which resulted was extracted with methylene Chloride and chloroform. The extracts were combined, washed thoroughly with water and with a saturated sodium chloride solution, filtered through magnesium sulfate, and concentrated. The yellow-brown oil which resulted was chromatographed on a column of 40 g of silicic acid. Elution |IIIIIIIIIIIlll---:::——————_______l 87 with chloroform gave 2.03 g of yellow—brown oil. Further elution with the more polar carbon tetrachloride-ethyl acetate (10:3) solvent system afforded 0.30 g (12%) of rust colored solid. Recrystallization two times from carbon tetrachloride-pentane provided an analytical sample of his- (1,2~dichlorovinyl)cyclopropenone Qg'as fine very light tan needles: mp 126.5-127.5°; ir (KBr) Figure 15; uv (heptane) Figure 18; mass spectrum (70 ev) m1g_(relative intensity), + )0 Molecular ion not present in spectrum. (M-28 , 220 (10.4), I )+, 185 (4.9), 218 (36.8), 216 (77.8), 214 (63.2); (M—63 183 (31.2), 181 (100.0), 179 (100.0). Anal. Calcd for C7H20Cl4: C, 34.47; H, 0.83; Cl, 58.14. Found: C, 34.45; H, 0.85; Cl, 58.23. 2. The Reaction Using a Mixture of cis— and trans—1,2— Dichloroethylene and a 4 to 1 Ratio of Tetrachloro— cyclopropene and Aluminum Chloride. A reaction mixture consisting of 1.1 g (6.2 mmol) of tetrachlorocyclopropene and 0.2 g (1.5 mmol) of anhydrous aluminum chloride in 6.0 g of a mixture of gigf and EEEEE‘ 1,2—dichloroethylene was stirred at room temperature under nitrogen for one hour and at reflux for ten hours. The clear brown reaction mixture was allowed to cool and poured into a separatory funnel to which 30 ml of methylene chloride was then added. After shaking thoroughly, 15 ml of water was added to the separatory funnel, the system thoroughly mixed, and the two phases separated. The aqueous solution was extracted once more with methylene chloride and two times 88 with ether. The extracts were combined and washed with a saturated sodium chloride solution, dried over magnesium sulfate, and concentrated. Chromatography of the oil which remained, as in reaction 1 above, afforded 0.25 g (16%) of cyclopropenone 54, 3. The Reaction Using trans-1,2—Dichloroethylene. A suspension of 1.8 g (10 mmol) of tetrachlorocyclo— propene and 1.0 g (7.5 mmol) of anhydrous aluminum chloride in 8 ml of trans—1,2-dichloroethylene (bp 49-520), which was distilled from a commercial mixture of gig: and EEEEE‘ 1,2—dichloroethylene, was stirred under nitrogen for two and one—half hours at room temperature and at reflux for eight hours. The reaction mixture was then stirred at room temperature for 12 hours. The reaction mixture was poured onto ice, the aqueous system extracted with methylene chlor— ide, chloroform, and ether, the extracts washed two times with saturated sodium chloride solution, dried over magnesium sulfate, and concentrated. The golden yellow oil which re- sulted was chromatographed on a column of 40 g of silicic acid. Elution with chloroform gave 1.81 g of yellow-red oil. Elution with a carbon tetrachloride-ethyl acetate solution (10:3) gave 0.13 g (5%) of a rust colored solid shown to be bis(1,2—dichlorovinyl)cyclopropenone 55, Further elution with the same solvent mixture afforded 0.10 g (7%) of q,5— dichloroacrylic acid identified by comparing its infrared spectrum and melting point with those given in the literature 89 (see the experimental section for the reaction of tetra- chlorocyclopropene with tetrachloroethylene). 4. The Reaction Using cis-1,2-Dichloroethylene. To a stirred suspension of 1.0 g (7.5 mmol) of anhydrous aluminum chloride in 8 ml of gig-1,2-dichloroethylene (bp 58-600), which had been distilled from a commercial mixture of Eléf and trans-1,2—dichloroethylene, was added 1.8 g (10 mmol) of tetrachlorocyclopropene. The reaction mixture was stirred at room temperature under nitrogen for four hours, at reflux for five hours, and again at room tempera— ture for 12 hours. The clear red—brown solution was then poured onto ice. The aqueous system was then extracted two times each with ether and chloroform and the extracts washed with saturated sodium chloride solution. The extracts were dried over magnesium sulfate, concentrated, and the yellow- brown oil which resulted chromatographed on a column of 40 g of silicic acid. Elution with chloroform afforded 1.94 g of yellow—red oil and on further elution with chloroform 0.32 g (13%) of bis(l,2-dichlorovinyl)cyclopropenone 5g, Elution with the carbon tetrachloride-ethyl acetate (10:3) solvent system gave an additional 0.09 g (4%) of 53 and 0.16 g (11%) of a,B—dichloroacrylic acid. 90 D. The Reaction of Tetrachlorocyclopropene with Tetra- chloroethylene A reaction mixture consisting of 1.8 g (10 mmol) of tetrachlorocyclOpropene, 1.0 g (7.5 mmol) of anhydrous alum— inum chloride, and 10.0 g of tetrachloroethylene was stirred at room temperature under nitrogen for two hours and at approximately 80° for three hours. The reaction mixture was poured on ice. The aqueous system was extracted three time with ether, the extracts washed one time with saturated sodium chloride solution, dried over magnesium sulfate, and concentrated by distilling off the ether at atmospheric pressure and the excess tetrachloroethylene and unreacted tetrachlorocyclopropene under vacuum. The oil which re— mained was chromatographed on a column of 40 g of silicic acid. Elution with chloroform gave only a small amount of oil. Elution with carbon tetrachloride—ethyl acetate (10:3) afforded 0.32 g (22%) of crystalline d,6—dichloroacrylic acid. Two recrystallizations from pentane provided and analytical sample: mp 85—860, lit. 86° (55). The infrared spectrum of the analytical sample was identical to that of an authentic sample of the acid as shown in the "Sadtler Index of Standard Infrared Spectra" (56). Aggi. Calcd for C3H202C12: C, 25.56; H, 1.43. Found: C, 25.51; H, 1.48. BIBLIOGRAPHY 10. BIBLIOGRAPHY R. West and D. L. Powell, J. Amer. Chem. Soc., 85,2577 (1963). R. West, "Squaric Acid and the Aromatic Oxocarbons", Aldrichimica Acta, i(2), 3 (1968). S. Cohen, J. R. Lacher, and J. D. Park, J. Amer. Chem. Soc., 5i, 3480 (1959). R West, . Y. Niu, D. L. Powell, and M. v. Evans, ibid., 82, 6204 (1960). R. West and M. Ito, 133$” 55, 2580 (1963). R. West and H. Y. Niu, igig., §§/ 2586 (1963) R. West and H. Y. Niu, igig., 55/ 2589 (1963). R. West and H. Y. Niu, igig., 54” 1324 (1962) a. E. T. McBee and J. S. Newcomer, igig., Zi, 952 (1949). b. A. Roedig and L. Hornig, Ann. Chem., 595’, 208 (1956). a. N. Prileschajew, Chem. Ber., 55/ 194 (1926). I b. M Mousseron, F. Winternitz, and R. Jacquier, Bull. Soc. Chim. Fr. , 81 (1947). c. M. Mousseron and R. Jacquier, ibid., 698 (1950). d. R. N. McDonald and P. A. Schwab, J. Amer. Chem. Soc., 55/ 4004 (1963). e. R. N. McDonald and T. E. Tabor, ibid., 55, 6573 (1967) f. R. N. McDonald and T. E. Tabor, J. Org. Chem., 55/ 2934 (1968) and references therein. g. R. N. McDonald and R. N. Steppel, J. Amer. Chem. Soc., gi/ 782 (1969). 91 92 11. H. 0. House, "Modern Synthetic Reactions," W. A. Benja- min, Inc., New York, 1965, p. 116. 12. M. Igarashi and H. Midorikawa, J. Org. Chem., 52” 3399 (1967) and references therein. 13. E. LeGoff and C. Bauer, unpublished observations. 14. K. Scherer, Jr., and T. Meyers, J. Amer. Chem. Soc., 52” 6253 (1968). 15. G. Moahs and P. Hegenberg, Angew. Chem. Intern. Ed., 5, 888 (1966). 16. F. W. McLafferty, "Interpretation of Mass Spectra," W. A. Benjamin, Inc., New York, 1967, p. 26. 17. F. Feigl and V. Anger, "Spot Tests in Organic Chemistry," 7th ed, Elsevier Publishing Co., New York, 1966, p. 66. 18. H. 0. House, "Modern Synthetic Reactions," W. A. Benja- min, Inc., New York, 1965, PP. 84—85. 19. F. Feigl and V. Anger, "Spot Tests in Organic Chemistry," 7th ed., Elsevier Publishing Co., New York, 1966, p. 178. 20. G. M. Strunz, A. D. Court, and J. Komlossy, Tet. Letters, 3613 (1969). 21. F. A. Bovey, "NMR Data Tables for Organic Compounds," Vol. 1, Interscience Publishers, New York, 1967. 22. H. 0. House, "Modern Synthetic Reactions," W. A. Benjamin, Inc., New York, 1965, p. 82. 23. A. Roedig and G. Markl, Ann. Chem., 636, 1 (1960). 24. A. Roedig, R. Helm, R. West, and R. M. Smith, Tet. Letters, 2137 (1967). 25. J. R. Dyer, "Applications of Absorption Spectroscopy of Organic Compounds," Prentice-Hall, Inc., Englewood Cliffs, N.J., 1965, pp. 34-38. 26. F. B. Kipping and J. J. Wren, J. Chem. Soc., 1733 (1957). 27. A. Roedig, F. Bischoff, H. Burkhard, and G. Markl, Ann. Chem., 670, 8 (1963). 28. J. Brandmuller and E. Ziegler, 2. Anal. Chem., 200, 299(1964). 29. H. C. Stevens and A. J. Kaman, J. Amer. Chem. Soc., 51, 734 (1965). lllhllll‘ll‘lllllllll Id||l1||l ll! Illilill 93 30. R. E. Parker and N. S. Isaccs, Chem. Rev., 55, 735 (1959). 31. A. Rowosky in "Heterocyclic Compounds with Three— and Four-membered Rings," A. Weissberger, Ed., Interscience Publishers, Inc., New York, 1964, Part I, p. 1. 32. D. L. Garin, J. Org. Chem., 52, 2355 (1969). 33. A. Padawa and R. Hartman, J. Amer. Chem. Soc., 55, 1518 1966 . 34. W; J. Linn, o. W. Webster, and R. E. Benson, ibid., 51, 3651 (1965). 35. P. Yates and G. H. Stout, ibid., Z5, 5110 (1954). 36. C. H. Hassall in "Organic Reactions," Vox. IX, Roger Adams, Ed., John Wiley and Sons, Inc., New York, 1957, Chapter 3. 37. R. West and K. Kusuda, J. Amer. Chem. Soc., 55, 7354 (1968). 38. R. M. Smith and R. West, Tet. Letters, 2141 (1969). 39. "Sadtler Index of Standard Infrared Spectra," published by Sadtler Research Laboratories, Inc., Spectrum No. 15453. 40. J. A. Krynitsk and R. W. Best, J. Amer. Chem. Soc., 55, 1918 (1947 . . Schafer, Chem. Comm., 1622 (1968). 41. L 42. R. Breslow, H. W. Chang, R. Hill, and E. Wasserman, J. Amer. Chem. Soc., 55, 1112 (1967). 43. L. Schafer and H. P. Fritz, J. Organometal. chem., i, 31 (1964). 44. R. West, A. Sade, and S. W. Tobey, J. Amer. Chem. Soc., 55, 2488 (1966). 45. S. W. Tobey, K. S. Whittemore, and M. Z. Lourandos, Abstracts, Division of Organic Chemistry, 156th National Meeting of the American Chemical Society, Atlantic City, N.J., Sept. 1968, no. 168. 46. a. R. West, J. Chickos, and E. Osawa, J. Amer. Chem. Soc., 55, 3885 (1968). b. E. Osawa, K. Kitamura, and Z. Yoshida, ibid., 55, 3814 (1967). c. A. W. Krebs, Angew. Chem. Intern. Ed., 5, 10 (1965). _¥—k d... 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 94 a. R. B. WOodward and R. HOffman, J. Amer. Chem. Soc., 51, 2146 (1965) . b. R. Hoffman and R. B. Woodward, Accounts Chem. Res., i, 17 (1968). c. D. G. Farnum, J. Chickos, and P. E. Thurston,:ii Amer. Chem. Soc., 55, 3075 (1966). W. S. Richardson and J. H. Goldstein, J. Chem. Phys., i5, 1314 (1950) . S. W. Tobey and R. West, J. Amer. Chem. Soc., 55, 4215 (1964). R. West and D. C. Zecher, ibid., 55, 152 (1967). . West and K. Kusuda, ibid., 55, 7354 (1968). . J. Prins, Rec. Trav. Chim., 55, 779 (1937). 22G :11 70 . J. Prins, ibid., 51, 659 (1938). S. W. Tobey and R. West, Tetrahedron Letters, 1179 (1963). w. H. Melville, Amer. chem. J., :1, 174 (1882). "Sadtler Index of Standard Infrared Spectre," published by Sadtler Research Laboratories, Inc., Spectrum No. 26553. APPENDIX 95 Table 7. Mass spectrum of compound 13 Rel. Rel. Rel. Rel. Rel. m/e Inten- m/e Inten— m/e Inten- m/e Inten— m/e Inten— Sity Sity sity sity sity 350 0.3 269 2.8 207 2.1 145 1.9 77 2.5 348 0.4 268 1.9 205 4.3 143 4.9 76 2.5 346 0.5 267 5.0 203 7.5 141 5.0 344 0.3 266 2.5 201 7.2 63 1.5 265 5.0 133 2.5 61 2.5 332 0.4 264 1.8 199 4.6 131 2.9 330 0.8 263 2.9 197 5.1 56 1.9 328 0.9 261 1.8 120 4.1 326 0.5 193 3.7 118 5.4 49 2.9 239 3.8 192 6.6 48 2.9 314 0.9 237 6.5 191 6.3 111 2.4 47 3.1 312 2.3 235 8 4 190 14.5 310 3.2 234 3.4 189 3.7 109 2.1 45 1.6 6 5 JA 308 2.1 233 . 188 10.6 107 3.2 44 14.0 .9 43 6. 232 3 297 0.6 231 5.0 183 2.4 103 3.2 295 1.3 230 5.0 181 2.8 38 41.2 293 1.9 98 1.6 37 4.1 291 1.3 229 7.6 171 4.0 96 3.7 36 100.0 227 13.2 169 7.1 35 10.0 278 1.6 225 12.8 167 5.4 89 276 3.5 87 5.0 28 12.0 274 2.8 223 1.8 157 . 221 4.6 155 4.0 85 4.3 18 13.2 219 9.0 153 4.0 83 7.1 17 3.8 217 6.8 96 Table 8. .Mass spectrum of compound 11, Rel. Rel. Rel. Rel. Rel. m/e Inten— m/e Inten— m/e Inten— m/e Inten— m/e Inten— Sity sity sity sity sity 394 0.3 268 0.3 207 0.3 122 0.7 65 0.4 392 1.0 267 0.4 205 0.9 120 3.5 63 0.8 390 2.1 266 0.4 203 2.0 118 5.4 388 2.6 265 0.5 201 1.4 61 0.7 386 1.5 264 0.3 113 0.6 60 0.9 263 0.4 194 2.1 111 1.6 332 0.3 192 8.5 50 331 0.3 261 0.2 190 17.9 110 0.3 49 0.6 330 0.5 259 0.3 188 13.6 109 0.5 48 329 0.4 257 0.2 108 0.8 47 328 0.7 185 .4 107 0.8 327 0.2 242 0.3 183 1.0 106 0.6 45 1.3 326 0.4 241 0.3 181 1.1 44 8.9 240 0.7 98 0.4 43 100.0 307 0.4 239 0.9 157 1.4 96 1.2 42 2.1 305 1.2 238 1.1 155 4.1 94 0.9 41 0.5 303 2.8 237 1.3 153 4.5 301 3.1 236 0.7 89 0.5 38 299 1.7 235 0.9 145 0.4 87 1.5 36 143 0.9 85 295 0.4 220 0.3 141 0.9 83 5.9 29 293 0.6 218 28 7.1 291 0.4 216 0.4 133 0.6 78 131 0.9 76 1.1 18 4.4 17 0.9 73 . 71 1.1 15 14 97 Table 9. Mass spectrum of compound 18. Rel. Rel. Rel. Rel. Rel. m/e Inten— m/e Inten— m/e Inten- m/e Inten- m/e Inten- Sity sity sity sity sity 358 2.4 251 11.7 187 7.5 123 4.8 73 6.3 356 7.5 249 47.6 185 12.9 121 10.5 71 11.1 354 11.1 247 100.0 119 15.6 352 6.9 245 88.6 173 6.9 65 4.5 243 50.6 171 18.9 113 6.0 63 9.6 323 3.9 241 50.6 169 18.6 111 9.9 61 6. 321 13.5 319 27.0 233 6.0 159 5.7 110 3.0 45 .2 317 21.1 231 11.4 158 4.2 109 6.9 44 .5 229 9.3 157 9.3 108 13.2 43 13.5 295 5.1 156 11.4 107 9.6 42 3.0 293 17.1 221 10.8 155 8.1 106 19.5 291 32.3 219 21.8 154 12.0 38 5.4 289 27.3 217 15.9 153 9.0 96 9.6 37 .1 36 13.8 286 21.8 205 10.8 151 4.2 89 3.6 35 5.4 284 65.3 203 22.4 149 12.6 87 11.7 282 91.6 201 21.8 147 13.8 86 5.4 29 9.0 280 55.5 199 8.4 85 14.7 28 11.4 145 10.2 84 14.7 27 .5 270 3.6 193 10.5 143 26.6 83 17.7 26 .3 268 9.9 191 20.2 141 26.3 266 15.0 189 15.9 78 .9 18 4-5 264 10.2 136 6.3 77 5.4 17 1. 135 6.3 76 .9 134 8.4 15 4.5 133 14.4 131 14.4 MM scouglcmflo h.®.m.m.m.fi mo Esnuommm omHMHmCH ue.mumummr_o.m.mLoHosoenouoeromxmrn . H mufimflm In. .UZwDOwou >u >u1330wa. 00w 01 .u - oce_ 000m cows 000m _ . OOOV _ . , N u h . . . _ flag 1 Jo _ 7 7 r . M.. m . A _ M I . m H - .;J77 77. :7 .if7 )5 .WW 7 ,. _ 1 I U .o 1.))-.-” --7... ON awoo_ 00m— oon m :3. 225 . - - l 11.; (-1- 71-7)--7illiii7:77w- :77 7- . m 7 8 m .w 4 8 hatch n 8 o 9 I._- l I t l O 37 (~74. JDNVlHWSNval O 0 0 co I‘ll'. ll7 l'l. lllli.’"‘ I‘ll" I 1|)7.' Iu.iil|llitll‘lllill I ii in 1I|| h . . ON. 0.: 0.9 293.2 or on 205:2 o... as. )Loe 99 .mm 064-603-973.56; IHo.N.MHOH0%OHQOHOHQUMHG0Q15.©.m.m.HI%xOHU>£IN mo Ednuommm UmumuwcH .N wusoflm .mluDanu A. .,/..; A; a 00m cool cemi recei comm . oom_ ooom OOWN 000m 00mm ooow 36:... .35 Go: 0.9 73.5 0: 0.01 mZOmEE Om OK 0.0 ow mZOmYE o.v 9m QDNVUIW‘TNVHL KWM scouwicwflolo . NI sumo: IHo.N.muoHomoaQOHoHSUmucmmlh.w.m.m.HtmxosumEIN mo Ednuoomm UmHmHmcH .m musmem “ _ to. Clo-40“.}. .10 0230:: 00m 000— OON— 00v— V 000m OOn m “#0:!— 5V l 7 w W ow 7 2. m. 1 M n w oo 3 n .‘/ A I \ om. oo. , . . . . . 0.3 oN— 0.: 0.0— . 5._..1,_.-_. 8---. | mzoeoi cs or as on mzoeoi one he or” as 101 beam 06027627503993 IOHOhOHmeo;mlmxonomfilmlonoanomanlw.b.m.v.N.a mo Eduuommm owHMHMCH .v wHSUHm .10 >Uzw30m3 1o yuzwaouf 0 00m 000. com. 003 002 009 ooom comm 000m 8mm 08v . . 7 . e _ l 7 1. 36:2 35 I. a v 9. ow w 7-..-. M u v w oo 06.: mm Om O 00 ..... 02) i 11..-. 111111111111) Illil) 00. md 0.0. 0&— o.: 0.9 mZOmuzz Qm OK 0.0 _ On mZOmU_<< oé Wm QM 102 umm UCDOQEOO mo.Mm mumumum who 00 ESuuoomm UmumHmcH O 000 000— 009 003 002 com, 000m . .7 . - 3 0m 0w 00 I n O i b l vvvvv m' on mzoeoal 04 m .m mudmflm s 73:: .55 00...... ' :13de 1 HWSNVdL 103 CV 00 «««««« 021,05... oov_ QR Q0 xmm Hmumx mo Eduuommm CommumcH _ 0n0m 00mm 000w N, AUW WvLflVNfiU_<< C V W.M .0 musmflm 00mm Aaouuom mev o F \i\ I 000v 2:!-11!Jo lUWSNVHl 1W 0/ )f‘ 104 av. 00 00 00— 0.2. 000 1.5150 use .wm. 3088 t cos wo Edmpoomm UmnmnmcH . 02030 :3 .b musmflm v. | I I -44.. | v :1. .. , U -2.-...4 .4..._..__. «2‘!2 ‘Tr‘ I .) ' J ”1,“ :, . I. . . 2 1 . v 1 I --v.—-.-...T 1 A u... .1... . .. -1- I..Il|. |I.II f . .l‘oll Q . . . Ill 2|! 2. . . m . 00v— 000v 0m 0 V O 0 V m)33NVLUWSNV81 b I 00 ON. 0;— Q0. mzoeoi om QR GAO on mzoeoi or. 105 \ka Uflomflw mo \Wm mwfluvwncm 93 mo 8350QO UmumumcH .w mndmflm I l Agoflflom umxv Q N . :/ ‘7. W m H \ 1..., 0 m V _ _ . M _ . , , \_.|«l: 1.1:. l ---z-iii - in- ifa! I -l - x ..i. Li+ - ,“_\ L 1 m: . < . V . u ST .: 112: lL- - r, ; i {51--. z 9 w I , ‘x . - _ ,. 106 00m 000_ .me UHUMHU m0 WNW Hmuwm HhQDGEHU GSU m0 Eduvommm UwHMHmCH .m mudmflm 1 Nu V OVN g M H V N can .. / Owr. W - f 1 a - 3.23.83 om m _ oxwuu _ r . fl ~ 00— m4 “ . ..»z n . . . 11: 1 - ‘ - . -i. oo_ 06. A9m_o;_ 06— mZOmU§gnvw Om 90 On mZOmuE< O? md Qm Wm 107 .Uflom UdameQMMUHUIN.lecmusnoaomolmflo mo Eduuommm UmymumcH 1U .. ”39...“... 000— 003 now w _ M. w JW .OH mHSmHm :02 no.3 r33 1?. 9.0m ( qlu. -I- 1 -i‘llal )1 l ’5 I..- -llJ .II III I I -ll - l + , _ n y . m p . - _ :37: La; 1.. - - Ln. . . u ._ p k . ..1 V v .. 1‘. .r u . g é§63}u ” . _ . n . J, I 00m. 000. KWW wflomfiv mo kummlmams .3:qu mzu wo 5:930me UmHMgwcH .HH mudmflm ., mi in. ., 15:. I»! @ONLI. ‘ ‘ylwmmliil “1002 002 000m OOwN 000m 000m %) aDNvuIWSNvm ( ‘ .. , I. ow , L 4%» W ‘ W ‘‘‘‘ r m\ _ ,. fl .. , , m m _ . w _ _ 4 v i I + ‘ r W a . . ., . . . H U ._ , ,___‘.: o: 00— mZOmEE Ow OK 00 _ on mZOmU=2 0v 109 .NM 9:69:00 mo WM- pamm EdHUOm msu mo Eduuommm UmumuwcH .NH mudmam ,- .1..M . mm 0 00m 000. OON_ 003 000— 8 a. M 3.. W 0v W m w 00 , .w .T. haoaaom umx“ w .u 0m .4 a .4 on _.q .9. u 00— . . . - tu-i- -2-. ll-l- :2 f: -, - - ,.2 0.0. 0.9 o.: 0.9 mZOmEE Qm OK 0.0 2 . Om mZOaUE ow mm 06 m.- 110 .mm mcocmmoumoH0>oAHth>OHoH£oHHuvaQ mo Eduuowmm UmumumcH .MH musmam -I ,.. a! -umawr---.wmmw. ,-,-mmmw coo. cow. oooN comm coon comm 000v lil'lll'] I‘ll-D J 2- ol |I|II l-[If' r III-[Illi-g O i _ cm i a V ch S M I... l V N can - 2 I l- - -$ - u 2 II- AJOAaoA hav , «Sunfloolouolfioouows om /\ .ll-llwl pl 0 I M _ o ,2-1.F- - 22-42-2- .— 0 0021le l--- .12 - - - - - - - 2 . J. o I - , .lh ' .' .fi.|l. I1 I - OOF o~.o__ 09 mzoqu ow QR ed on mzoggz oe ad on _ 2 _ mm Kmm ocmamumomAHmzflyouoanoflnuvmflfl mo Eduflowmm UmumumcH .vH musmflm ,Eu >U7w30wuu Xv .ngficwu‘ OON_ 00: 000— 009 OnON DOWN Gown 000v m . . m m _ U _ . 2 . A 2 ,2 O a m 2 2 om I. a V - 0v N S M l l V N con - 2 - - - - . - - . - A‘Hnlm Lav - - . f , N- - W. cm . vflwuulo HulRuul-U ,0 09 ------ I2|I,!-?2-i2 - 2 I 2 2 2 -,I-22-.: .I- . , - - 2 - 2WJIL- - - 09 0.2 2- 0.3 0.: 0.9 mzomui 3 OK 0.0 3. 220632 3 3 om 2 .wm wcocwmonmoaohoA 00w 000— CON. Q~_o;_ od_ mzomgz oa 9m OOV_ ahcfl>onoanoflvlm~fivmfln mo com. oooN fl ooo_ Ennuowmm UmHmHMCH .11.)wa comm 000m ow asayz Qv ma HHHGHAUOIOHO truer-$30 / \ .mH mnsmflm oonn ill] Act—Alb .35 = O l, 2 2, 0m ON 0 V O 0 (%L33Nv1uw9Nvm 113 me waocmmonmoaomoAchfl>ogoH£0flHpvaQ mo Eduuommm umaofl>muuab transmittance OOH om ow ON T I r I l V T I I {Til-J .®H mmsmflm mconoHEHHHflE Qumawam>m3 omv oom mum 0mm mum oom _ $3 3:3 an in a... L 3“; uwuoav \I man 055.7: .- - _ . _ _ # «Hoondoolouolaoouomflo /\ o = o eoueqxosqe 114 transmittance amm mamamumowAchfl>ouoaaofluuvmflfl mo Esuuommm pmaofl>mnuab .bH musmflm mGOHoHEHHHHE sumcwam>m3 Omv 00¢ 0mm mNm 00m DEN 0mm mNN CON 2: m p a _2,. / \- ow . 2 om . / > >//\K . S n / \ //\ L m 2 , 2, x _ - H :/ \ a 2 < K MMMHH wwwmw Mu “mm gag... _ _ m . 2 icon-Sohomolsmuoms A oo aoueqlosqe 115 transmittance xwm mcoqmmonmoaoho Iflawqfl>omoa£oflclmtavmflfl mo Ednuommm umHOH>MHuHD .wH wudmflm mQOHOflEHHHflE mecmam>m3 0 0mm mum 0mm mam com oon5 7lllllf- -\\\2, 1!! H // \\\\\\\l r // 2\/ t om_- /(\l\ ow. 2r \ om. w OH ”I w H o M - - 1 _ a a _ $3 3 3c \- RN i . _ . _ gun-J; N - _ HozoflaoolO/Hvo lacuna-.330 o = o v.0 N.H ®.H eoueqxosqe IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 2222|22L222222122222922L222L22L222L|2L22L22|2L222L22L2|2L2222|22