New peroxide method provides large rings cial synthesis—especially musk com pounds vital to the perfume industry. Several patents to be assigned to the University of Georgia have recently been applied for, and a new company, Unichem Laboratories, Inc., of Athens, Ga., has been formed to exploit this technology. The method may also have great impact on the drug indus try, since it should be applicable to synthesis of macrolide antibiotics such as erythromycin, a highly substituted C1H lactone. History. The beginning of modern macrocyclic chemistry may be dated from Leopold Ruzicka's establishment in 1926 of large ring s tinctures for the musk ketones, muscone and civetone. However, the chemistry of large carbocyclic systems is relatively unknown because of the difficulty in the past of preparing them: An expensive and generally hard-to-prepare long-chain difunctional precursor was required, and cyclizations were carried out at high dilution. In recent years, several methods have been developed to overcome these problems and to generate large rings directly from readily available monomers-by F. Sondheimer in 1956, E. J. Corey in 1967, E. Wasserman in 1968, and P. Story in 1968-but only
ORGANIC Preparation of macrocyclic compounds, once involving considerable difficulty, has been reduced almost to the ordi nary by a generally applicable syn thetic method developed by Dr. Paul R. Story and Dr. Peter Busch at the Uni versity of Georgia. Dr. Story and Dr. Busch, together with coworkers Dr. C. E. Bishop, D. D. Denson, K. Paul, and B. Lee, have now synthesized carbocyclic rings of every size from C 8 through C33—including many ring sys tems bearing one or more substituents -using their ketone peroxide fragmen tation method
[C&EN, Feb. 5, 1968,
page 20]. Since first publication of this macrocyclic synthesis two years ago, [J. Am. Chem. Soc, 90, 817 (1968)], Dr. Story has developed the method in a number of important ways, including: • New and better synthetic methods for preparation of peroxides. • Preparation of mixed peroxides. • New and safer techniques of per oxide decomposition, giving much higher yields. The Georgia chemist thus opens up macrocyclic systems to easy commer
the latter devised a completely general synthetic method, ketone peroxide fragmentation. The synthesis developed by Dr. Story uses inexpensive and readily available starting materials. The sim ple procedure involves thermal decom position of the appropriate ketone per oxide by refluxing in a hydrocarbon solvent, generating the desired macrocyclic compound directly in good yield. This thermal decomposition method gives higher yields and faster reaction times than photolysis, and is safer than pyrolysis in sealed ampoules, two methods used by Dr. Story previously. The procedure is typified by the thermal decomposition of tricyclohexylidene peroxide in refluxing decane solvent at 174° C. to give cyclopentadecane in 30 to 35% yield and 16hexadecanolide ( dihydroambrettolide ) in 10 to 15% yield, with some regen eration of cyclohexanone starting ma terial. Extreme caution is necessary in handling all of the peroxides, because of their shock sensitivity. Dihydroambrettolide is an important macrocyclic musk, and cyclopentadecane can be readily oxidized to cyclopentadecanone, also an important musk. Both of these expensive musks are obtained in a purer state and at a
Peroxide fragmentation gives cyclic compounds Thermal decomposition produces roauces musks: musks:
C
ο M
v
? ° X> / S k^l
Tficyctohexyfidene peroxide
^
Cydopentàdecane
+
On
Ο
+ Cydohexamme ύ
IS-Hexadeowolfcie (Dftydroamkrettoiide)
Mixed trimeric peroxides enable synttièsis of ail size rinôsî
?
Q00H
οN X " l^J
uS0
+ Λ £ *-
"
Cyclopentafione
U'~Dihydroperoxy-
Q
s ^-^(j-Af+C&aH
6 4 4 Trimeric peroxide
Substituted rings are easily prepared:
Q ^όοΗ
15-Pentad«canûJide OCH,
· ο 1
Cyriotetra* decane
HCi04 w -10*C.
l^U'-Dihydroperoxy- 4-Methoxycyolodicyclohexyl peroxide hexanone
40 C&EN MARCH 9, 1970
>•*** Metfcoxytricydo-
Mettoxycyclopentadecane
Dr. Story (left) and Dr. Busch analyze products of peroxide fragmentation
fraction of the cost of present methods, Dr. Story says. Mixed peroxides. The most im portant new development since Dr. Story's original 1968 publication of his procedure is his discovery of methods of generating mixed trimeric peroxides in good yield. The decomposition of symmetrical dimeric and trimeric per oxides was previously accomplished by the Georgia scientist. But to make the method truly general and extend it to synthesis of all size rings, substituted and unsubstituted, it is necessary to synthesize unsymmetrical peroxides. The key intermediate in the synthe sis of mixed trimeric peroxides is 1,1'dihydroperoxydicyclohexyl peroxide, prepared by a new method giving quantitative yields. In a typical reac tion, this dihydroperoxide is stirred with cyclopentanone and anhydrous cupric sulfate at room temperature for 13 days. Addition of a large excess of water and recrystallization of the crude product from methanol give a 58% yield of the 6-6-5 trimeric peroxide. Decomposition of the mixed 6-6-5 trimer by refluxing in decane yields cyclotetradecane and 15-pentadecanolide, in 17 and 22% yields, respec tively. These rings are not otherwise available, except through cyclooctanone diperoxide—which is obtainable only in low yield. Many mixed tricycloalkylidene per oxides have been synthesized and con verted to the corresponding macrocy clic compounds. The 12-12-7 trimeric peroxide gives a C 28 hydrocarbon ring and a C 2 9 lactone on thermal decom position, in 30% and 10 to 15% yields, respectively. Some of the macrocyclic compounds obtainable from the mixed peroxides cannot be synthesized from any readily available symmetrical per oxide. The utility of the method depends considerably on the availability of the dihydroperoxide. Dr. Story and Dr. Busch have recently developed a new procedure for preparing it in very high yield. Typically, it is synthesized by stirring for five hours, at room tempera
ture in an open crystallization dish, a mixture of 90% hydrogen peroxide and cyclohexanone in 2:1 molar propor tions, together with several drops of 70% perchloric acid catalyst in acetonitrile. The mixture becomes a crys talline mass, is washed with water, and recrystallized from hexane, giving a quantitative yield of 94% pure dihy droperoxide. Besides the 6-6 dihydro peroxide, Dr. Story has also prepared 5-5, 7-7, 8-8, 12-12, and other perox ides in good yield. Substituted macrocyclics. Perhaps even more important than other appli cations, monosubstituted macrocyclic compounds can be synthesized by the ketone peroxide fragmentation method. For example, methoxycyclopentadecane can be prepared in 35% yield, by reacting l,l'-dihydroperoxydicyclohexyl peroxide with 4-methoxycyclohexanone for two to three hours at — 10° C. in propionic acid solvent, us ing a perchloric acid catalyst, and then decomposing the methoxytricyclohexylidene peroxide product. Monosubstituted macrocyclic com pounds have been prepared with alkyl, aryl, acetoxy, benzoyloxy, halogen, and amido substituents. The decomposition of the ketone peroxides is hypothesized by Dr. Story to proceed through homolysis of an oxygen-oxygen bond. The cleavage should be analogous to that of alkyl peroxides and ozonides, he believes, with a double β scission following ho molysis of the oxygen-oxygen bond, producing alkyl radicals—which un dergo a very efficient cage recombina tion to form a new carbon-carbon bond. An intermediate acyl peroxide is formed, and the cyclic hydrocarbon results from its homolytic decomposi tion, loss of carbon dioxide, and cage recombination of the resulting alkyl radicals. Preliminary data obtained on the kinetics, solvent effects, and activ ity parameters of the reaction are in support of this scheme. The ketone peroxide fragmentation procedure already constitutes the most generally useful synthesis of macrocy clic systems. Dr. Story and his col leagues are currently seeking to extend the utility of the method to synthesis of very large ring systems—up to C10o> if possible—opening the door to study of many interesting conformational problems, and possible use of C 3 0 to C 4 0 rings as selective ion-chelating agents. They are also planning to ap ply their techniques to synthesis of heterocyclic ring systems and macrolide antibiotics. A further possible ap plication is preparation of new monomeric systems for polymerization—us ing small or large rings as polymer monomers—yielding elastomeric ma terials which are elastic on a molecu lar basis because of ring rigidity.
one of a group of Evans Mercapto Acids...
THIOLACTIC ACID (2-Mercaptopropionic Acid)
HS-CH-COOH
W***
Other Evans Mercapto Acids commercially available include: Thioglycolic Acid (Mercaptoacetic Acid), 3-Mercaptopropionic Acid, Thiomalic Acid (Mercaptosuccinic Acid), Thiosaiicylic Acid (o-Mercaptobenzoic Acid). Inquiries are invited regarding your com mençai requirements for other Organic Di-Valent Sulfur Compounds. Data Sheets and Samples available on request.
CIX/AI^S
CH6fT\eTICS, IRC. 90 Tokeneke Road Darien. Connecticut 06820 Phone: 203-655-8741 Cable: EVANSCHEM TWX: 710-468-2148
MARCH 9r 1970 C&EN
41
