Macrocyclic polyether complexes and a process for the isolation of macrocyclic polyethers

Abstract

 Complexes between (1) a disulphonate of formula

(X is barium or strontium and R¹ is a bivalent group derived from an alkane by removal of two hydrogen atoms) and (2) a macrocyclic polyether compatible with a barium or a strontium ion. Also claimed is a process for isolating such macrocyclic polyethers in high yield from a solution containing them, by reaction with a disulphonate of the said formula to precipitate the said complex, separating the complex and extracting the separated complex with a solvent for the macrocyclic polyether.

Description

[0001] The invention relates to novel complexes of macrocyclic polyethers and to a process for the isolation of these macrocyclic polyethers from a mixture containing the said macrocyclic polyethers with the aid of such complexes.

[0002] Macrocyclic polyethers are defined as heterocyclic compounds containing a ring made up of carbon atoms and at least three oxygen atoms, in which each of the oxygen atoms is bound to two carbon atoms. An example of a macrocyclic polyether is 1,4,7,10,13,16-hexaoxacyclo-octadecane. Hereinafter this compound will also be referred to by its trivial name "18-crown-6". Its structural formula is shown on the Formula page (compound 1).

[0003] 18-Crown-6 can be prepared by heating tetraethylene glycol and bis(2-chloroethyl) ether in the presence of tetrahydrofuran as a solvent and potassium hydroxide. see "Synthesis" 1976, 515-516. The reaction mixture thus formed contains 18-crown-6, potassium chloride, water and by-products, i.a. macrocyclic polyethers having a ring of larger size than 18-crown-6. The solvent was evaporated from the reaction mixture to give a brown slurry to which dichloromethane was added. The potassium chloride was filtered off from the solution thus obtained, the filtrate was dried (MgSO₄), the solvent was evaporated from the dried filtrate to yield a residue of crude 18-crown-6. This residue was distilled to give a discoloured distillate containing 18-crown-6. This distillate was dissolved in acetonitrile and the solution was cooled to -45°C. The resultant precipitate of 18-crown-6--acetonitrile complex was collected by filtration. Distillation of this complex gave a distillate of pure 18-crown-6.

[0004] A disadvantage of this procedure is that the 18-crown-6 is distilled twice and that expensive measures must be taken to prevent the occurrence of powerful and destructive explosions during these distillations, see "Chemical & Engineering News", September 6, 1976, page 5 and December 13, 1976, page 5. Moreover, the 18-crown-6 is obtained in a fairly low yield and the 18-crown-6-acetonitrile complex is precipitated at a very low temperature.

[0005] The Applicants have found complexes of macrocyclic polyethers with salts of a specific group, which complexes can be precipitated from a number of solvents as explained further hereinafter.

[0006] The present invention provides complexes formed between (1) a disulphonate of the general formula

wherein X represents a barium or strontium atom and R¹ represents a bivalent group derived from an alkane by removal of two hydrogen atoms, and (2) a macrocyclic polyether compatible with the ion derived from the atom X.

[0007] R¹ has preferably fewer than ten and particularly fewer than six carbon atoms. Groups R¹ having fewer than three carbon atoms are most preferred. The group R¹ may be derived from an alkane by removal of two hydrogen atoms from one carbon atom or of one hydrogen atom from two carbon atoms. Examples of groups R1 are methylene, ethylidene, isopropylidene, dimethylene, trimethylene, tetramethylene and propylene groups. Among the groups R¹ having fewer than three carbon atoms the groups -(CH₂)_n-, n being an integer of less than 3, are preferred, i.e. methylene and dimethylene groups. Methylene groups are most preferred. The disulphonate of formula I may be water-free or may contain water of crystallization.

[0008] The macrocyclic polyether present in the novel complexes should be compatible with a barium and/or a strontium ion, that is such an ion and the macrocyclic polyether must be stereo- chemically compatible with regard to the size of the hole in the polyether ring and the diameter of the ion. Hence, the hole in the polyether ring should be capable of accommodating a barium or a strontium ion. The closer the fit between the diameter of the ion and the size of the hole, the more stable the complex will be. Accordingly, preferred macrocyclic polyethers are those consisting of 4 to 10

units where Y represents the group

each of _R² and _R³ representing a hydrogen atom or an alkyl group having from one to four carbon atoms. To reduce steric hindrance methyl and ethyl groups are preferred among the alkyl groups _R² and _R³ in formula III may represent. Preferably, both R² and _R³ represent a hydrogen atom. Preferred complexes are those between barium methanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecane, between barium methanedisulphonate and 1,4,7,10,13-pentaoxacyclopentadecane, between barium 1,2-ethanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecanc, and between strontium methanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecane.

[0009] Another preferred group of macrocyclic

of compounds in which the polyether ring in the macrocyclic polyether contains 4 to 10 oxygen atoms and in which

(a) each oxygen atom of the polyether ring is separated from. the next oxygen atom by two carbon atoms,

(b) at least one pair of vicinal carbon atoms of the polyether ring also forms part of an aromatic ring or of a ring obtained by saturation of such an aromatic ring, and

(c) each of the carbon atoms of the polyether ring only forming part of the polyether ring is bonded to (1) a hydrogen atom and (2) a hydrogen atom or an alkyl group having from one to four carbon atoms.

[0010] The aromatic ring present in the macrocyclic polyether may be a carbocyclic or a hetero-aromatic ring. The preferred carbocyclic aromatic ring is the o-phenylene ring. The carbocyclic aromatic rings may be fused, as is the case in, for example, 1,2- and 2,3-naphthylene groups and 1,2- and 2,3-an- thrylene groups. Hetero-aromatic rings are derived from hetero-aromatic compounds which are defined as in Kirk-Othmer, "Encyclopedia of Chemical Technology ", Second Edition, Volume 2 (1963), page 702: obtained by replacement of one or more carbon atoms of a carbocyclic aromatic compound by a hetero-atom -for example pyridine, pyrimidine, pyrazine, quinoline and isoquinoline - and also include those heterocyclic compounds having five-membered rings which show aromatic characteristics and are mentioned on page 703 of the said volume, for example thiophene, pyrrole, furan, indole and benzothiophene. Examples of hetero-aromatic rings are 2,3-furylene, 2,3-thienylene, 3,4-furylene and 3,4-thienylene rings. The aromatic ring or rings and the rings obtained by saturation of such aromatic rings may carry substituents, for example halogen atoms or alkyl, nitro or cyano groups.

[0011] Preferably only one pair of vicinal carbon atoms of the polyether ring also forms part of an aromatic ring or of a ring obtained by saturation of such an aromatic ring. The preferred ring obtained by saturation of an aromatic ring is the 1,2-cyclohexylene ring. Methyl and ethyl groups are preferred among the alkyl groups bound to the carbon atoms of the polyether ring only forming part of the polyether ring. Preferably, each of the latter carbon atoms is bonded to two hydrogen atoms. Preferred complexes are those between barium methanedisulphonate and 2,3-benzo-1,4,7,10,13,16-hexaoxacyclo- octadeca-2-ene, between barium methanedisulphonate and 2,3-ben- zo-1,4,7,10,13,16,19-heptaoxacycloheneicosa-2-ene and between barium methanedisulphonate and 2,5,8,15,18,21-hexaoxatricy- c_lo/20.4.0.0⁹'¹⁴ 7-_hexacosane.

[0012] The complexes according to the invention are very suitable for use in a process for the isolation of macrocyclic polyethers from a mixture containing them. The macrocyclic polyethers can thus be obtained in high yield and need not be distilled, thus avoiding the explosions mentioned hereinbefore. Moreover, the use of very low temperatures is avoided.

[0013] Accordingly, the present invention also provides a prcccss for the isolation of a macrocyclic polyether compatible with a barium or strontium ion from a solution containing the said macrocyclic polyether, which process comprises contacting the said solution with a disulphonate of the general formula

wherein X represents a barium or strontium atom and R¹ represents a bivalent group derived from an alkane by removal of two hydrogen atoms, to form a solid complex of the said disulphonate and the said macrocyclic polyether separating the solid complex from the mixture containing the complex and extracting the separated complex with a solvent for the macrocyclic polyether.

[0014] The complex formed between a disulphonate of formula I and a compatible macrocyclic polyether contains one molecule of the disulphonate per molecule of macrocyclic polyether. Accordingly, the starting solution containing the macrocyclic polyether is preferably contacted with the disulphonate cf formula I using a molar ratio of disulphonate to macrocyclicpolyether of at least 1. The reaction mixture formed Uy contacting the starting solution with such a disulphonate contains as solid materials the complex and any stoichiometric excess of disulphonate. To reduce the excess of disulphonate in the reaction mixture the molar ratio of disulphonate to macrocyclic polyether is preferably not higher than 2. Molar ratios of disulphonate to macrocyclic polyether higher than 2, for example up to 5, may be used, if desired, but usually will not be of further advantage. The use of a molar ratio of disulphonate to macrocyclic polyether of less than 1 gives a complex containing only very little non-complexed disulphonate, if any at all. In view of these considerations this molar ratio is preferably near to 1.0, for example between 1.0 and 1.1.

[0015] The solution containing the macrocyclic polyether may be contacted with the disulphonate of formula I by adding the disulphonate to the solution and stirring the suspension thus formed for, for example, 0.5 to 10 hours, to form the solid complex. The macrocyclic polyether present in the mixture obtained upon this reaction partly resides in the complex and the balance is dissolved non-complexed in the solution obtained. The proportion of the macrocyclic polyether residing in the complex is particularly high, for example more than 97%, when the macrocyclic polyether is 18-crown-6. This proportion is considerably higher than in the case of the 18-crown-6-acetonitrile complex; hence, the 18-crown-6 can be isolated in a correspondingly higher yield. Furthermore, the macrocyclic polyether need not be reacted with the disulphonate of formula I at a very low temperature. The solution containing the macrocyclic polyether is preferably contacted with the disulphonate at at a temperature in the range of from -30^oC to +50°C and particularly in the range of from 15°C to 30°C. Ambient temperature is very suitable.

[0016] The Applicants have modified the process according to the invention by replacing the disulphonate of formula I by potassium hexafluorophosphate, barium acetate, barium ethanesulphonate, barium p-toluenesulphonate, barium succinate and barium 2-methylene-1,3-propanedisulphonate, but these salts did not form precipitates with dissolved 18-crown-6.

[0017] A wide variety of solvents may be present in the starting solution containing the macrocyclic polyether to be isolated. Very suitable solvents are alkanols, nitroalkanes, dialkyl sulphoxides, dialkyl ketones, alkanenitriles and dialkyl ethers; all of these solvents preferably have fewer than six carbon atoms per molecule. Particularly preferred are methanol, nitromethane, dimethyl sulphoxide, acetone and acetonitrile. Other examples of suitable solvents are tetrahydrothiophene 1,1-dioxide and N-methylpyrrolidone.

[0018] The macrocyclic polyether in the starting mixture may have been formed by any known process. Very good results have been obtained with 18-crown-6 formed by reacting tetraethylene glycol with a bis(2-haloethyl) ether, halo representing chloro, bromo or iodo, in the presence of an alkali metal hydroxide, as described in "Synthesis" 1976, 515-516. If desired, alkali metal halide formed may be removed from the reaction mixture obtained, leaving the 18-crown-6-containing starting solution.

[0019] The disulphonates of formula I are selective in that, when contacted with a mixture containing 18-crown-6 and one or more other compatible macrocyclic polyethers, they preferentially form complexes with 18-crewn-6. Hence, 18-crown-6-containing starting mixtures containing one or nore other compatible macrocyclic polyethers are very suitable starting materials, particularly those obtained by catalytic oligomerization of ethylene oxide, as described in German )ffenlegungsschrift 2,401,126.

[0020] Other examples of processes for the preparation of 18-crown-6 are:

(1) elimination of hydrogen chloride from 17-chloro-3,6,9,1?,15--pentaoxaheptadecanol, followed by ring closure, in the presence of potassium tert-butoxide, see British patent specification 1,285,367 and

(2) reaction of triethylene glycol with 3,6-dioxa-1,8-dichloro- octane in the presence of potassium hydroxide and 10% aqueous tetrahydrofuran, as described in J.Org.Chem. 39 (1974)2445-2446.

[0021] The suspended complex can easily be separated from the solution obtained, for example by filtration, centrifugation or decantation.

[0022] As the solid complexes between a disulphonate of formula I and a macrocyclic polyether in the presence of a solvent for the macrocyclic polyether are in equilibrium with the solid disulphonate and dissolved macrocyclic polyether the complex can simply be extracted with such a solvent, thus giving a solution of the macrocyclic polyether and leaving the solid disulphonate. The temperature at which this extraction-is carried out is not critical and is preferably in the range of from 50°C to 125⁰C. Suitable solvents are, for example, nitromethane, acetonitrile and tetrahydrofuran. Evaporation of the solvent from the solution of the macrocyclic polyether at sub-atmospheric pressure leaves the macrocyclic polyether. This macrocyclic polyether is very pure and need not be distilled. When nitromethane is used as the solvent and 18-crown-6 as the macrocyclic polyether, a residue of the 18-crown-6-nitromethane complex is first formed. The latter complex is described in British patent application 26333/77. When acetonitrile is used as the solvent and 18-crown-6 as the macrocyclic polyether, a residue of the 18-crown-6-acetonitrile complex is first formed. Upon further-heating the 18-crown-6-nitromethane complex is decomposed into 18-crown-6 and nitromethane and the 18-crown-6-acetonitrile complex into 18-crown-6 and acetonitrile. The disulphonate of formula I, left after the extraction with the solvent for the macrocyclic polyether, may be reused for complexing further quantities of compatible macrocyclic polyether.

[0023] The invention will now be illustrated by reference to the following Examples.

[0024] The barium methanedisulphonate used was water-free.

Example I

[0025] A suspension obtained by addition of barium methanedisulphonate (1 mmol) to a 0.1 M solution (20 ml) of 18-crown-6 in methanol was stirred for six hours at 20°_c, the molar ratio of barium methanedisulphonate to 18-crown-6 being 0.5. Then, analysis by NMR spectroscopy of the filtrate obtained by filtration showed that the barium methanedisulphonate had removed 1 mmol of the 18-crown-6. Hence, the filtered material was an equimolar complex between barium methanedisulphonate and 18-crown-6. An elemental analysis of this complex gave the results presented in Table I. These results are compared with the percentages calculated for the equimolar complex.

Examples II-VI and Comparative Experiment A

[0026] A suspension obtained by addition of barium methanedisulphonate (1.5 mmol) to a 0.1 M solution (10 ml) of a macrocyclic polyether in methanol was stirred for one hour at 25°C. Then, analysis by NMR spectroscopy of the filtrate obtained by filtration of the reaction mixture gave the percentage of macrocyclic polyether removed from the starting solution, presented in Table II. Six macrocyclic polyethers were tested in this manner; their structural formulas are shown on the formula page. Each macrocyclic polyether has been given a number which is mentioned in Table II and on the formula page. Table II shows that barium methanedisulphonate removes macrocyclic polyethers 1 to 5 and hardly removes macrocyclic polyether 6, if at all. Hence, macrocyclic polyethers 1 to 5 are compatible and macrocyclic polyether 6 is not compatible with barium ions.

Examples VII and VIII and Comparative Experiments B-1

[0027] Ten experiments were carried out as described in Example II, the difference being that the barium methanedisulphonate was replaced by 1.5 mmol of another salt. Table III states the ten salts used and presents the results.

Example IX-XV

[0028] A suspension obtained by addition of barium methanedisulphonate (0.3 mmol) to a 0.1 M solution (2 ml) of 18-crown-6 was stirred for four hours at 20°C. Then, analysis by NMR spectroscopy of the filtrate obtained by filtration of the reaction mixture gave the percentage of 18-crown-6 removed from the starting solution. Seven solvents were tested in this manner. Table IV presents the results.

Example XVI

[0029] A suspension obtained by addition of barium methanedisulphonate (0.104 mmol) to a solution in methanol (1.5 ml) of 0.10 mmol of 18-crown-6 and 0.10 mmol of macrocyclic polyether 3 of the formula page was stirred for six hours at 20⁰C. The molar ratio of macrocyclic polyether 3 to 18-crown-6 in the filtrate obtained by filtration of the reaction mixture was 7:1, indicating that 18-crown-6 is preferentially complexed.

Examples XVII-XIX

[0030] A suspension obtained by addition of a solvent (20 ml) to the .18-crown-6-barium methanedisulphonate complex (0.35 mmol) was heated under reflux for 16 hours. Then the suspended solid material was filtered off at reflux temperature and the filtrate obtained was boiled down at a pressure of 1.9 Pa, leaving a residue of 18-crown-6. Three solvents were tested in this manner. Table V presents the yield of 18-crown-6, calculated on 18-crown-6 in the starting complex.

Examples XX and XXI

[0031] The 18-crown-6-barium methanedisulphonate complex (1 mmol) was continuously extracted for 16 hours in a Soxhlet apparatus. At the end of this period the percentage of the 18-crown-6 extracted from the complex was determined. Two solvents were tested. Table VI presents the results.

[0032] 

Preparation of crude 18-crown-6

[0033] A 3 1 three-necked round-bottomed flask, fitted with a mechanical stirrer, a reflux condenser and a 250-ml dropping funnel was charged with potassium hydroxide pellets (416 g. containing 6.3 mol KOH), tetraethylene glycol (1.25 mol) and tetrahydrofuran (1000 ml). The reaction vessel was placed in a heating mantle and gently heated. After 15 minutes a solution of bis (2-chloroethyl) ether (3.125 mol) in tetrahydrofurann (150ml) was added in one stream from the dropping funnel to the vigorously stirred reactants.The reaction mixture was then heated under reflux, with stirring for 18 hours. Subsequently, the reaction mixture was cooled and deiled down under a pressure of 1.9 kPa to give a brown slurry to wich dichloromethane (750 ml) was added. The resulting suspension of potassium chloride was filtered and the potassium chloride riltered off was washed with dichloromethane (100 ml). The combined filtrate and washings were dried over anhydrous magnesium sulphate and the solvent was evaporated at a pressure of

kPa to give a residue of crude 18-crown-6 (396 g) containing 0.531 mol of 18-crown-6 (yield 42.5%, calculated on starting tetraethylene glycol); potassium chloride and at least eight other compounds among which 1,4,7,10,13,16,19,22,25,28,31,34-dodeca- oxacyclohexatriacontane, hexaethylene glycol and unreacted tetraethylene glycol.

This crude 18-crown-6 was used as described hereinafter.

Comparative Experiment J

[0034] Crude 18-crown-6 (39.6 g), prepared as described above, was distilled at a pressure of 20 Pa to give a distillate (21.3 g) boiling at 140-210°C. This distillate was mixed with acetonitrile (53 ml) at 20°C and the mixture formed was cooled to -45°C. The resultant 18-Crown-6-acetonitrile complex was filtered off and subjected to distillation at a pressure of 2 Pa to give a distillate of 18-crown-6 in a yield of 25% calculated on starting tetraethylene glycol, or.59% calculated on 18-crown-6 in the crude 18-crown-6.

Comparative Experiment K

[0035] Crude 18-crown-6 (3.021 g), prepared as described above, was dissolved in acetonitrile (4 ml), the solution formed was cooled to -20°C and the 18-Crown-6-acetonitrile complex was filtered off. The complex was kept for 30 minutes at 70°C and a pressure of 13 Pa, leaving 18-crown-6 in a yield of 27% calculated on starting tetraethylene glycol, or 63% calculated on 18-crown-6 in the crude 18-crown-6.

Example XXII

[0036] Crude 18-crown-6 (0.991 g), prepared as described above, was dissolved in methanol (38 ml), barium methanedisulphonate (5.7 mmol) was added to the solution obtained and the suspension formed was stirred for six hours at 20°C. Then, the suspended material (1.82g) was filtered off and extracted with nitromethane (180 ml) in a Soxhlet apparatus. The extract phase obtained was boiled down as a pressure of 1.6 kPa to yield a crystalline residue of the 18-crown-6-nitromethane complex. The complex was kept for 30 minutes at 70°C and a pressure of 13 Pa, leaving 18-crown-6 in a yield of 34%,calculated on starting tetraethylene glycol, or 79%, calculated on 18-crown-6 in the crude 18-crown-6.

Example XXIII

[0037] A solution of crude 18-crown-6 (1.202 g) prepared as described above in methanol (30 ml) was cooled to -25°C. Barium methanedisulphonate (6.75 mmol) was added to the cooled solution and the suspension formed was stirred for five hours at -25°C. Then, the suspended material (2.26 g) was filtered off and extracted with nitromethane (200 ml) in a Soxhlet apparatus. The extract phase obtained was boiled down at a pressure of 1.6 kPa to yield a crystalline residue of the 18-crown-6-nitromethane complex. The complex was kept for 30 minutes at 70°C and a pressure of 13 Pa, leaving 18-crown-6 in a yield of 30% calculated on starting tetraethylene glycol, or 72% calculated on 18-crown-6 in the crude 18-crown-6.

Claims

1. Complexes formed between (1) a disulphonate of the general formula

wherein X represents a barium or strontium atom and R¹ represents a bivalent group derived from an alkane by removal of two hydrogen atoms, and (2) a macrocyclic polyether compatible with the ion derived from the atom X.

2. Complexes as claimed in claim 1, in which the group R in formula I has fewer than six.carbon atoms.

3. Complexes as claimed in claim 2, in which the group R¹ in formula I has fewer than three carbon atoms.

4. Complexes as claimed in claim 3, in which the group R represents a group -(CH₂)_n-, n being an integer of less than 3.

5. Complexes as claimed in any one of the preceding claims, in which the macrocyclic polyether consists of 4 to 10

units, where Y represents the group

each of R² and R³ representing a hydrogen atom or an alkyl group having from one to four carbon atoms.

6. Complexes as claimed in claim 5, in which R and _R³ in group III both represent a hydrogen atom.

7. The complex formed between barium methanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecane.

8. The complex formed between barium 1,2-ethanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecane.

9. The complex formed between strontium methanedisulphonate and 1,4,7,10,13,16-hexaoxacyclooctadecane.

10. The complex formed between barium methanedisulphonate and 1,4,7,10,13-pentaoxacyclopentadecane.

11. Complexes as claimed in any one of claims 1 to 4, in which the polyether ring in the macrocyclic polyether contains 4 to 10 oxygen atoms and in which

(a) each oxygan atom of the polyether ring is separated from the next oxygen atom by two carbon atoms,

(b) at least one pair of vicinal carbon atoms of the polyether ring also forms part of an aromatic ring or of a ring obtai.ned by saturation of such an aromatic ring, and

(c) each of the carbon atoms of the polyether ring only forming part of the polyether ring is bonded to (1) a hydrogen atom and

(2) a hydrogen atom or an alkyl group having from one to four carbon atoms.

12. Complexes as claimed in claim 11, in which the aromatic ring mentioned in (b) is an o-phenylene ring.

13. Complexes as claimed in claim 11 or 12, in which only one pair of vicinal carbon atoms of the polyether ring also forms part of an aromatic ring or of a ring obtained by saturation of such an aromatic ring.

14. Complexes as claimed in any one of claims 11 to 13, in which each of the carbon atoms of the polyether ring only forming part of the polyether ring is bonded to two hydrogen atoms.

15. The complex formed between barium methanedisulphonate and 2_;3-benze-1,4,7,10,13,16-hexaoxacyclooctadeca-2-ene.

16. The complex formed between barium methanedisulph

2,3-benzo-1,4,7,10,13,16,19-heptaoxacycloheneicosa-2-ene,

17. The complex formed between barium methanesulphonate and 2,5,8,15,18,21-hexaoxatricyclo/ 20.4.0.0^9'14/ 7 h_ex_ac_osane.

18. Complexes as claimed in claim 1 substantially as hereinbefore described.

19. Process for the isolation of a macrocyclic polyether compatible with a barium or strontium ion from a-solution containing the said macrocyclic polyether, which process comprises contacting the said solution with a disulphonate of the general formula

wherein X represents a barium or strontium atom and R represents a bivalent group derived from an alkane by removal of two hydrogen atoms, to form a solid complex as claimed in claim 1, separating the solid complex from the mixture containing the complex and extracting the separated complex with a solvent for the macrocyclic polyether.

20. Process as claimed in claim 19, in which the starting solution containing the macrocyclic polyether is contacted with the disulphonate of formula I at a temperature in the range of from -30°C to +50°_C.

21. Process as claimed in claim 20, in which the starting solution containing the macrocyclic polyether is contacted with the disulphonate of formula I at a temperature in the range of from 15°C to 30°C.

22.. Process as claimed in any one of claims 19 to 21, in which the solvent of the starting solution containing the macrocyclic polyether is an alkanol with fewer than six carbon atoms per molecule.

23. Process as claimed in claim 22, in which the alkanol is methanol.

24. Process as claimed in any one of claims 19 to 21, in which th- solvent of the starting solution containing the macrocyclic polyether is a dialkyl sulphoxide with fewer than six carbon atoms per molecule.

25. Process as claimed in claim 24, in which the dialkyl sulphoxide is dimethyl sulphoxide.

26. Process as claimed in any one of claims 19 to 21, in which the solvent of the starting solution containing the macrocyclic polyether is a dialkyl ketone with fewer than six carbon atoms per molecule.

27. Process as claimed in claim 26, in which the dialkyl ketone is acetone.

28. Process as claimed in any one of claims 19 to 21, in which the solvent of the starting solution containing the macrocyclic polyether is an alkanenitrile with fewer than six carbon atoms per molecule.

29. Process as claimed in claim 28, in which the alkanenitrile is acetonitrile.

30. Process as claimed in any one of claims 19 to 29, in which the starting solution containing the macrocyclic polyether is contacted with the disulphonate of formula I using a molar ratio of disulphonate to macrocyclic polyether in the range of from 1 to 2.

31. Process as claimed in any one of claims 19 to 30, in which the macrocyclic polyether is 1,4,7,10,13,16-hexaoxacyclo- octadecane which has been obtained by reacting tetraethylene glycoi with a bis(2-haloethyl) ether in the presence of an alkali metal hydroxide.

32. Process as claimed in any one of claims 19 to 30, in which the starting mixture contains 1,4,7,10,13,16-hexaoxacy- clooctadectadecane and one or more other macrocyclic polyethers.

33. Process as claimed in any one of claims 19 to 32, in which the separated complex is extracted with nitromethane.

34. Process as claimed in any one of claims 19 to 32, in which the separated complex is extracted with acetonitrile.

35. Process as claimed in any one of claims 19 to 34, in which the separated complex is extracted with the solvent for the macrocyclic polyether at a temperature in the range of from 50°C to 125°C.

36. Process as claimed in claim 19, substantially as hereinbefore described with reference to Examples XXII and XXIII.

37. Macrocyclic polyethers whenever isolated by a process as claimed in any one of claims 19 to 36.

Drawing