Six years ago [2] we discussed some of the principles involved in the synthesis of catena compounds, a new class of substances consisting of rings, which are not chemically combined but which are held together like the links of a chain (I), and began some experimental work on such systems [2,3].As thereactions we envisaged were of s t a t i s t i c a l character, only very poor yields were to be expected. Our experimental results [3] were even more disappointing than expected although they gave valuable information about the behavior and morphology of long-chain and macrocyclic compounds in solution.On the other hand, Wussevmarz [4] claimed to have obtained yields of approximately 1 % (relative to the entire mixture isolated) using one of these methods. Later on he modified this statement to a claim [5] of having obtained by chromatography yields that were several powers of ten less than those previously reported, namely a few milligrams of an oil which was thought to be a catenane. The new principle of the d i r e c t e d synthesis which we have now applied proved successful. By steric control it avoids the extra-annular ring closures which always occur predominantly in statistical cyclization methods. [6] was converted into the dicarboxylic ester (3) by Wittig reaction with an excess of 10-methoxycarbonyldecylidenetriphenylphosphorane [7]. Hydrolysis gave the crystalline dicarboxylic acid (4) (m.p. 67-71 OC, after sintering) in 52 -58 % yield based on (2). Catalytic hydrogenation and esterification of ( 4 ) gave (5). Surprisingly, this ester could not be cyclized by the acyloin ring closure method of Hansley, Pvrlog, and StoN, but remained unchanged, whereas the dicarboxylic ester (6) of a hydroquinone ether was cyclized readily. 4,5-Dimethoxyisophthalaldehyde (2) ____ 0~1 1~The diester (5) was therefore reduced to the diol (7) with LiAIH4, and (7) was converted into the dinitrile (9) via the dibromide (8). The cyclic P-imino nitrile obtained from (

Langkettige Dithiole unterschiedlicher Kettenlange mit verschieden substituierten, mittelstandigen aromatischen Ringen lassen sich zu makrocyclischen Disulfiden (Ringgliederzahl 18, 20, 24 und 28) mit Ansa-Struktur dehydrocyclisieren. Alle untersuchten Dithiole geben EinschluRverbindungen mit a-bzw. 8-Cyclodextrin. Die Dehydrocyclisation nach EinschluD sollte zu einem neuartigen Verbindungstyp X bzw. XXIX fuhren, bei dem sich zwei Ringe wie Kettenglieder umfassen, ohne chemisch aneinander gebunden zu sein. Es wird gezeigt, daR die Cyclodextrin-Umhullung auch in verdiinnter waRriger Losung noch so stark ist, daR die Dithiol-Dehydrierung bei der gewahlten Kettenlange verhindert wird.Molekule aus mindestens zwei Ringen, die einander umfassen, ohne in irgendeiner chemischen Bindung miteinander zu stehen (Bautyp XXIX einer echten Gliederkette), sind bislang weder aus natiirlichem noch aus synthetischem Material isoliert worden. Ihre Existenz in der belebten Natur erscheint keineswegs ausgeschlossen. Polymerisate und Polykondensate wie rohes Terylenl), Nylon oder Perlon2) enthalten neben linearen Molekulen auch Makroringe, und nach den Gesetzen der Statistik sollte sich bei derartigen Polymerisationen auch ein geringer Anteil von gliederkettenartigen Doppelringen bilden. PROBLEMSTELLUNGUns interessieren seit langer Zeit die Bildungsmoglichkeiten und Eigenschaften von Verbindungen des Typs X bzw. XXIX. Von den vielen fur die Synthese in Frage kom-*) Vorgetragen von H. PRINZBACH im Eroffnungskolloquium zum Neubau des Chem.

No abstract

No abstract

No abstract