Лако-красочные материалы — производство

The problem in the synthesis of macrocyclic musks lies with entropy. The simplest approach to building a large ring is to make a long chain with functionality at each end such that the two ends of the chain can react to close the ring through the formation of a new carbon-carbon bond. However, entropy dictates that the likelihood of the two ends of the chain meeting is lower than that of one end reacting with the end of another chain, repetition of which leads to polymerization. Thus, for example, if 15-hydroxypentadecanoic acid is subjected to esterification conditions, the product is a polyester rather than the musk cyclopenta — decanolide. In the 1930s, Stoll solved this problem through the use of extremely high dilution, which reduces the chance of intermolecular reaction and therefore gives the functional groups at each end of the reagent molecule time to find each other and undergo intramolecular reaction. The high dilution principle works well on a laboratory scale, but it is not satisfactory as a manufacturing process because of the poor reactor utilization and the need to recover and recycle very large volumes of solvent. Both of these features lead to unacceptably high process costs.

Many methods have been developed in an attempt to overcome the problem posed by entropy in the formation of large rings. Inclusion of unsaturation in the substrate reduces the number of degrees of freedom in it and consequently increases the possibility of the two ends of the chain meeting. In the acyloin reaction, electrostatic attraction is used to restrict movement of the alkyl chain. A dicarboxylic acid is added to a suspension of sodium (or another alkali metal) particles in an inert solvent. The acid functions react with the metal to form carboxylate anions which are held to the positive surface of the metal by electro­static attraction. This means that the chain becomes a loop, ‘fastened’ to the metal particle at each end until the two carboxylate groups approach close enough to allow the acyloin reaction to take place between them.

Another method was developed in the 1930s by Carothers. He began by preparing polyesters from hydroxy acids or from mixtures of dicarboxylic acids and diols. He then depolymerized these to give monocyclic lactones. The lactones are much more volatile than the oligomers and polymers in the mixture and so can be separated from it by distillation. Obviously, this steady removal of the lactones also helps to force the equilibrium in the desired direction. The dicarboxylic acids are available by oxidation of unsaturated fatty acids or cyclic olefins or ketones. For example, ozonolysis of erucic acid, the major fatty acid of oilseed rape, gives brassylic acid (tridecanedioic acid). His first method of depolymerization was to heat the polymer to a high temperature and allow the chains to bite back on themselves, and the volatile lactones were removed by distillation from the mixture. He later developed the useful technique of using high-boiling alcoholic solvents to achieve the depolymerization.

The main alcohol used was glycerol. Glycerol serves two purposes in the depolymerization. Firstly, it provides hydroxyl groups to help keep the interesterification equilibrium reactions in progress. Secondly, the boiling point of glycerol is in the same range as those of lactones in the 15-18 carbon range. By maintaining the system under reflux of glycerol, removal of the lactones by distillation is made more efficient. Furthermore, the lactones are only poorly soluble in liquid glycerol and the distillate readily separates into two phases, making removal of the lactones easy through use of a Dean-Stark trap. Calcium oxide or hydroxide is usually employed as the catalyst in these reactions. Unfortunately, glycerol is not very stable under the conditions and much of it is lost, adding to costs in terms of both materials used and waste disposal. Ethylene brassylate is prepared by this type of process from brassylic acid and ethylene glycol. The low cost of the starting materials and the simplicity of the process make this the least expensive of macrocylic musks at present, and therefore the one which is used in the highest tonnage.

The second largest macrocylic musk, in tonnage terms, is cyclopenta- decanolide. Its synthesis is shown in Scheme 4.48. The key starting

material is methyl undecylenate, which is obtained from castor oil by pyrolysis and esterification. Exposure of tetrahydrofuran to a radical initiator generates the radical at a carbon next to the oxygen atom. This radical adds to the terminal double bond of methyl undecylenate and the radical produced then abstracts a hydrogen atom from another tetrahydrofuran and so propagates the radical chain reaction. The methyl ll-(2′-tetrahydrofuryl)undecanoate thus produced is subjected to elimination, hydrogenation and hydrolysis to give 15-hydroxypenta — decanoic acid. Polymerization-depolymerization gives cyclopenta — decanolide.

-н2о

Cyclopentadecanolide

Scheme 4.48

In 1959, Wilke discovered that butadiene could be trimerized round a metal template to give cyclododecatriene. This could be converted into the mono-olefin, the ketone, the alcohol, etc., by obvious means. The ready availability of an inexpensive supply of cyclododecane deriva­tives set in train a new direction in musk research. The 12-carbon ring could be broken open, with or without addition of a side chain, to provide new linear precursors for macrocyclization. Furthermore, the ring could be enlarged by fusing a second ring to it, breaking the bridgehead bond to produce a larger ring. This latter option offers an elegant means of overcoming the entropic problem of macrocycliza­tion. An example of this type of approach is shown in Scheme 4.49.

Eschenmoser’s synthesis of muscone (Scheme 4.49) uses cyclodode — canone as the starting material and employs the Eschenmoser frag­mentation reaction in the key bridgehead-breaking step. Firstly, a methacrylate unit is fused to the 12-membered ring using conventional anionic chemistry. This leads to the bicyclic ketone which can be converted into the epoxide with alkaline hydrogen peroxide. Addition of methane sulfonylhydrazine gives the hydrazone. When treated with base, the hydrogen attached to the nitrogen atom is lost and the resultant negative charge flows through the molecule to spring open the epoxide ring. The negative charge can then flow back across the bridgehead by a different route and the molecule fragments, losing nitrogen. It is the energy gain in forming free nitrogen that drives this reaction. The acetylenic ketone produced is easily hydrogenated to produce muscone. Routes such as this are very elegant examples of chemical synthesis, but they are multi-step and therefore attract high process costs when operated on an industrial scale. At present, no macrocyclic musks are available in the same cost bracket as the polycyclic musks. Consequently, the fragrance industry still carries out a great deal of research into methods of producing macrocyclic ketones and lactones at lower cost.