Method of manufacturing cyclophosphazene derivatives

Abstract

A method of manufacturing cyclophosphazenes PFPE derivatives to be used in the lubrication of magnetic recording media is herein provided. The method comprises: a) a (per)fluoropolyether (PFPE) polyol [PFPE (P.sub.pol)] comprising a fluoropolyoxyalkylene chain (R.sub.f) having two chain ends, each chain end comprising at least one hydroxy group, and b) the corresponding alkoxide of perfluoropolyether (P.sub.pol) [PFPE (P.sub.alk)] wherein the equivalent concentration of PFPE (P.sub.alk) in PFPE (P.sub.pol) is lower than 30%, preferably ranging from 5% to 15%; 2) contacting mixture (M) with a perhalocyclophosphazene (CP.sub.halo) to provide a mixture (M1) containing an equivalent ratio of PFPE (P.sub.alk)/(CP.sub.halo) of at least 1; 3) allowing mixture (M1) to react until complete disappearance of PCl groups to provide a mixture (M2); 4) submitting mixture (M2) to hydrolysis to provide a mixture (M3); 5) optionally removing (P.sub.pol) from mixture (M3) to provide a mixture (M4). A method of purifying mixture (M4) is also herein provided.

Claims

1. A method of manufacturing cyclophosphazene derivatives, the method comprising: contacting a mixture (M), the mixture (M) comprising: a) a (per)fluoropolyether (PFPE) polyol [PFPE (P.sub.pol)] comprising a fluoropolyoxyalkylene chain (R.sub.f) having two chain ends, each chain end comprising at least one hydroxy group, and b) a corresponding alkoxide of said perfluoropolyether (PFPE) polyol [PFPE (P.sub.alk)], wherein the equivalent concentration of PFPE (P.sub.alk) in PFPE (P.sub.pol) is lower than 30% with a perhalocyclophosphazene (CP.sub.halo) to provide a mixture (M1) containing an equivalent ratio of PFPE (P.sub.alk)/(CP.sub.halo) of at least 1; allowing mixture (M1) to react until complete disappearance of P.sub.halo groups to provide a mixture (M2); submitting mixture (M2) to hydrolysis to provide a mixture (M3); optionally removing PFPE (P.sub.pol) from mixture (M3) to provide a mixture (M4).

2. The method according to claim 1 wherein PFPE (P.sub.pol) complies with formula (II):
YOR.sub.f-Y(II) wherein R.sub.f is a fully or partially fluorinated polyoxyalkylene chain comprising repeating units R, said repeating units being selected from the group consisting of: (i) CFXO, wherein X is F or CF.sub.3, (ii) CFXCFXO, wherein X, equal or different at each occurrence, is F or CF.sub.3, with the provision that at least one of X is F, (iii) CF.sub.2CF.sub.2CW.sub.2O, wherein each of W, equal or different from each other, are F, Cl, or H, (iv) CF.sub.2CF.sub.2CF.sub.2CF.sub.2O, (v) (CF.sub.2).sub.jCFZ-O wherein j is an integer from 0 to 3 and Z is a group of general formula OR.sub.fT.sub.3, wherein R.sub.f is a fluoropolyoxyalkene chain comprising from 0 to 10 recurring units selected from: CFXO, CF.sub.2CFXO, CF.sub.2CF.sub.2CF.sub.2O, and CF.sub.2CF.sub.2CF.sub.2CF.sub.2O, with each of X being independently F or CF.sub.3 and T.sub.3 being a C.sub.1-C.sub.3 perfluoroalkyl group; and Y and Y, equal to or different from one another, represent a hydrocarbon group containing at least one free hydroxy group, said hydrocarbon group being optionally fluorinated and/or optionally containing one or more heteroatoms.

3. The method according to claim 2 wherein Y and Y are independently selected from: CFXCH.sub.2O(CH.sub.2CH.sub.2O).sub.nH, CFXCH.sub.2O(CH.sub.2CHCH.sub.3O).sub.nH, and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH, wherein X is F or CF.sub.3 and n ranges from 0 to 5; CFXCH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.nH and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.nH, wherein X is F or CF.sub.3 and n ranges from 1 to 3.

4. The method according to claim 2 wherein chain R.sub.f complies with formula (R.sub.f-III):
(CF.sub.2CF.sub.2O).sub.a1(CF.sub.2O).sub.a2-(R.sub.f-III) wherein a1 and a2 are integers>0 such that the number average molecular weight is between 400 and 10,000, with the ratio a2/a1 being comprised between 0.1 and 10; and Y and Y, equal to or different from one another, are selected from: CFXCH.sub.2O(CH.sub.2CH.sub.2O).sub.nH and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH, wherein X is F or CF.sub.3 and n ranges from 0 to 5; CFXCH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.nH and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.nH, wherein X is F or CF.sub.3 and n ranges from 1 to 3.

5. The method according to claim 4 wherein PFPE (P.sub.pol) is a PFPE diol (P.sub.diol) (IIA) wherein both Y and Y comply with formula CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH wherein n ranges from 0 to 2.

6. The method according to claim 4 wherein PFPE (P.sub.pol) is a mixture of: PFPE diol (P.sub.diol) (IIA), wherein both Y and Y comply with formula CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH wherein n ranges from 0 to 2; PFPE tetraol (P.sub.tetraol)-(IIB), wherein both Y and Y comply with formula CF.sub.2CH.sub.2OCH.sub.2CHOHCH.sub.2OH and PFPE triol (P.sub.triol)(IIC), wherein one of Y and Y is a group of formula CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH wherein n ranges from 0 to 2 and the other of Y and Y is a group of formula CF.sub.2CH.sub.2OCH.sub.2CHOHCH.sub.2OH said mixture of PFPE (P.sub.pol) (IIA)-(IIC) being optionally in the protected form.

7. The method according to claim 1 wherein perhalocyclophosphazene (CP.sub.halo) is selected from one or more of perhalocyclophosphazene (CP.sub.halo) complying with formula (I-A) or (I-B): ##STR00014## wherein Hal is a halogen selected from fluorine, chlorine, bromine and iodine.

8. The method according to claim 1 wherein the equivalent ratio of (PFPE-P.sub.alk)/(CP.sub.halo) ranges from 1.1 to 2.5.

9. The method according to claim 8 wherein the equivalent ratio of (PFPE-P.sub.alk)/(CP.sub.halo) is 2 and wherein the equivalent concentration of (PFPE-P.sub.alk) in (CP.sub.halo) is between 5% and 15%.

10. The method according to claim 1 further comprising submitting mixture (M4) to fractionation with a supercritical fluid.

11. A mixture of: (A) cyclophosphazene derivatives complying with formulae (CP-1)-(CP-4): ##STR00015## wherein R.sub.f is a fluoropolyoxyalkylene chain; z is 3 or 4, w is selected from 0, 1 or 2; and T and T, equal to or different from one another, represent a hydrocarbon group which is optionally fluorinated and which optionally contains one or more heteroatoms and/or one or more hydroxy groups, and (B) a PFPE (P.sub.pol) of formula (II) as defined in claim 2.

12. The mixture of cyclophosphazene derivatives according to claim 11, wherein the mixture comprises the cyclophosphazene complying with formulae (CP-1)-(CP-3), characterised by a molar content of cyclophosphazene derivative (CP-1) of at least 40%.

13. A lubricant composition comprising a mixture of cyclophosphazene derivatives according to claim 11 in admixture with further ingredients or additives.

14. A method of lubrifying magnetic recording media comprising contacting the media with a lubricant composition as defined in claim 13.

15. The method according to claim 4 wherein a1 and a2 are integers>0 such that the number average molecular weight is between 400 and 5,000, with the ratio a2/a1 being comprised between 0.2 and 5.

Description

DETAILED DESCRIPTION OF THE PROCESS OF THE INVENTION

(1) The process of the invention is typically carried out as described herein below.

(2) Step 1Preparation of Mixture (M)

(3) Mixtures (M) are typically prepared by treating a PFPE (P.sub.pol) with a base, usually a strong inorganic base, preferably NaOH or KOH, more preferably KOH, or an organic base, preferably potassium tert-butylate, in an equivalent amount ranging lower than 30%, preferably ranging from 5 to 15%, with respect to PFPE (P.sub.pol). Typically, a water solution of inorganic base having a concentration of about 50% wt is added to PFPE (P.sub.pol) and the resulting mixture is heated until complete elimination of water. Usually, the temperature is of about 70 C., but it can be adjusted by the person skilled in the art according to the selected PFPE (P.sub.pol) and base. In the present description, the expression an equivalent amount of base means the equivalents of the base referred to the total equivalents of hydroxy groups of the PFPE (P.sub.pol).

(4) Step 2Preparation of Mixture (M1)

(5) A perhalocyclophosphazene (CP.sub.halo) (I-A) and/or (I-B) is dissolved in a fluorinated aprotic polar solvent, which is typically selected from hydrofluoroethers (HFEs), like 3M Novec HFEs, hydrofluorocarbons (HFCs) and hexafluoroxylene, the preferred solvent being hexafluoroxylene. The kind and amount of solvent will be selected by the skilled person according to the selected (CP.sub.halo); however, the amount of solvent is typically adjusted in such a way as that the concentration of (CP.sub.halo) ranges from 1 to 10% w/w.

(6) For the sake of clarity, it is pointed out that the term equivalent referred to perhalocyclophosphazenes (CP.sub.halo) is referred to PCl groups therein contained. Thus, 1 mol of perhalocyclophosphazenes (CP.sub.halo)-(IA) contains six equivalents of PCl groups, while 1 mol of perhalocyclophosphazenes (CP.sub.halo)-(.sub.IB) contains 8 equivalents of PCl groups. In order to obtain cyclophosphazene derivatives wherein each P atom in the phosphazene ring bears two PFPE substitutents, the equivalent ratio between (PFPE-P.sub.alk) and (CP.sub.halo) must be of at least of 1; this means that if (CP.sub.halo) (IA) is used, the molar ratio between the PFPE-P.sub.alk and (CP.sub.halo) (IA) must be at least 6; if (CP.sub.halo) (IB) is used, the molar ratio between the PFPE-P.sub.alk and (CP.sub.halo) (IA) must be at least 8. However, it has been observed that, in order to optimise the reaction rate, it is preferred that the equivalent ratio of (PFPE-P.sub.alk)/(CP.sub.halo) ranges from 1.1 to 2.5; an equivalent ratio of about 2 is particularly preferred. Indeed, it has been observed that when the process of the invention is carried out using an equivalent ratio of (PFPE-P.sub.alk)/(CP.sub.halo) equal to 2 and a mixture (M) wherein the equivalent concentration of (PFPE-P.sub.alk) in PFPE (P.sub.pol) is between 10 and 15%, mixtures (M4) with an overall content of PFPE (CP-1), (CP2) and (CP3) of about 80% wt can be obtained in less than 10 hours.

(7) According to a preferred embodiment (herein after procedure A), mixture (M) is stirred and heated at a temperature ranging from 40 C. to 90 C., then slowly added with the solution of (CP.sub.halo), preferably (I-A) and/or (I-B), typically in about 2 to 6 hours.

(8) According to another embodiment, (herein after procedure B), a solution of (CP.sub.halo), preferably (I-A) and/or (I-B), is stirred and heated at a temperature ranging from 40 C. to 90 C., then slowly added with mixture (M).

(9) According to another embodiment (herein after procedure C), mixture (M) and the solution of (CP.sub.halo), preferably (I-A) and/or (I-B), are rapidly mixed together at room temperature, to provide a mixture (M1) which is then heated to a temperature ranging from 40 C. to 90 C.

(10) Among procedures A-C, procedure A is preferred.

(11) Step 3Preparation of Mixture (M2)

(12) After obtainment of mixture (M1) [i.e. once mixture (M) is contacted with all the solution of (CP.sub.halo)], the fluorinated aprotic polar solvent is optionally removed, typically by evaporation under vacuum, then the mixture is stirred and heated until complete conversion of the PCl groups of (CP.sub.halo) into POCH.sub.2 groups. Typically, conversion is checked by withdrawing samples and by submitting them to .sup.31P-NMR spectroscopy; complete conversion (99% conversion) is confirmed by the appearance of a singlet at 17 ppm.

(13) Step 4Preparation of Mixture (M3)

(14) Once complete conversion is achieved, the resulting mixture (M2) is submitted to hydrolysis, namely acid hydrolysis. Typically, hydrolysis is accomplished by addition of aqueous HCl and an aliphatic alcohol, typically isobutyl alcohol. The aqueous phase is then separated to provide an organic phase which, after drying and removal of solvent(s), affords mixture (M3). Mixture (M3) comprises PFPE cyclophosphazene derivatives (CP-1)-(CP-4) as defined above in admixture with unreacted PFPE (P.sub.pol), preferably a PFPE (P.sub.pol) of formula (II) as defined above, in an amount typically ranging from 50 to 90% wt with respect to the weight of the mixture.

(15) If a mixture of PFPE (P.sub.pol) (IIA)-(IIC) is used in the protected form, the protective groups are completely removed according to known methods.

(16) Mixtures (M3) obtainable according to the above steps 1)-4) are also part of the present invention. These mixtures can be used in cases where the lubrication of MRM is satisfactorily achieved at low concentrations of the PFPE cyclophosphazene derivatives of the invention, which still contain a certain amount of unreacted PFPE (P.sub.pol) (IIA)-(IIC) ranging from 50 to 90% wt. Mixtures (M3) can be either used as such or they can be used in the preparation of further lubricant compositions.

(17) Step 5) Preparation of Mixture (M4)

(18) Mixture (M3) can optionally be submitted to distillation in order to remove the excess of PFPE (P.sub.pol) to provide mixtures (M4), which comprises PFPE cyclophosphazene derivatives (CP-1)-(CP-4) as defined above in admixture with unreacted PFPE (P.sub.pol), preferably a PFPE (P.sub.pol) of formula (II) as defined above, said PFPE (P.sub.pol) being in a lower amount than in mixture (M3); typically, in mixture (M4) the PFPE (P.sub.pol) is in an amount ranging from 1 to 30% with respect to the weight of the mixture.

(19) Mixtures (M4) obtainable through steps 1)-5) as defined above are also part of the present invention.

(20) According to a preferred embodiment, mixtures (M3) and (M4) in accordance with the present invention are those obtainable from PFPE diols (II) as defined above and phosphazenes (CP.sub.halo)-(IA) and/or (IB) as defined above.

(21) Such mixtures thus comprise dangling cyclic phosphazenes (CP-1) complying with the formula here below:

(22) ##STR00012##
wherein R.sub.f is a fluoropolyoxyalkylene chain as defined above, z is 3 or 4 and T and T, equal to or different from one another, represent a hydrocarbon group which is optionally fluorinated and which optionally contains one or more heteroatoms and/or one or more hydroxy groups. Preferably, T and T, equal to or different from one another, are chosen is such a way as T-O and T-O are any one of the followings:
CFXCH.sub.2O(CH.sub.2CH.sub.2O).sub.n, CFXCH.sub.2O(CH.sub.2CHCH.sub.3O).sub.n and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.n, wherein X is F or CF.sub.3 and n ranges from 0 to 5;
CFXCH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.n and CF.sub.2CF.sub.2CH.sub.2O(CH.sub.2CHOHCH.sub.2O).sub.n, wherein X is F or CF.sub.3 and n ranges from 1 to 3.

(23) Particularly preferred are mixtures (M3) and (M4) obtainable by using (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, with a PFPE diol (IIA) or with a mixture of PFPE (P.sub.pol) (IIA)-(IIC) as defined above. Thus, these mixtures contain a dangling cyclic phosphazenes of formula (CP-1) wherein n is 3 and groups T-O and T-O, equal to or different from one another, are selected from CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.n and CF.sub.2CH.sub.2OCH.sub.2CHOHCH.sub.2O, wherein n is as defined above; preferably, n ranges from 0 to 2.

(24) Most particularly preferred are mixtures (M3) and (M4) obtainable using a (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, and a PFPE diol (IIA) as defined above, wherein both Y and Y are CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH, wherein n is as defined above. Thus, these mixtures contain a dangling cyclic phosphazenes (CP-1) wherein z is 3 and groups T-O and T-O are both CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.n, wherein n is as defined above.

(25) Most particularly preferred are also mixtures (M3) and (M4) obtainable using a (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, and a protected mixture of PFPE (P.sub.pol) (IIA)-(IIC) as defined above. These mixtures (M3) and (M4) contain a dangling cyclic phosphazenes (CP-1) wherein z is 3 and groups T-O are independently selected from CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.n wherein n is as defined above, and CF.sub.2CH.sub.2OCH.sub.2CHOHCH.sub.2O, while groups T-O are CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.n wherein n is as defined above.

(26) Mixtures (M3) and (M4) typically contain, in addition to the desired dangling PFPE (CP-1), also spiro-, ansa- and bridged hydroxy-PFPE cyclophosphazenes [herein after also respectively referred to as PFPE (CP-2), PFPE (CP-3) and PFPE (CP-4)], complying with the formulae reported here below:

(27) ##STR00013##

(28) wherein R.sub.f, z, w, T and T are as defined above.

(29) As explained in step 4) above, mixtures (M3) contains an amount of PFPE (P.sub.pol), preferably a PFPE (P.sub.pol)-(II) typically ranging from 50% to 90% wt. Mixtures (M4) still contain a residual amount of PFPE (P.sub.pol), preferably a PFPE (P.sub.pol)-(II), typically ranging from 1% to 30% wt. Therefore, mixtures (M3) and (M4) both comprise PFPE (CP-1)-(CP-4) in admixture with unreacted PFPE (P.sub.pol), preferably a PFPE (P.sub.pol)-(II). Mixtures (M4) can be used as such in the lubrication of MRM or for the preparation of lubricant compositions or they can be submitted to fractionation, including, but not limited to, fractionation with chromatographic techniques, solvent extraction techniques, and fractionation with a supercritical fluid, as described in detail herein below.

(30) In particular, the applicant observed that any residual amount of PFPE (P.sub.pol)-(II) can be removed by fractionation of a mixture (M4) with a supercritical fluid, namely supercritical CO.sub.2 (scCO.sub.2) and that this technique allows separating bridged PFPE (CP-4) from PFPE (CP1), (CP2) and (CP3), thereby obtaining a mixture [mixture (M5)] with reduced polydispersity.

(31) Therefore, a further aspect of the present invention is a method for purifying a mixture (M4) containing a PFPE cyclophosphazene derivative [PFPE (CP-1)] wherein each phosphorus atom of the phosphazene ring bears two PFPE chains, each PFPE chain bearing at least one hydroxy group, said method comprising submitting mixture (M4) to fractionation, preferably to fractionation with a supercritical fluid, more preferably to fractionation with scCO.sub.2.

(32) In particular, a further aspect of the invention is a method comprising, preferably consisting of, steps 1)-5) as defined above, followed by a step 6) comprising, preferably consisting of, the fractionation of mixture (M4), preferably fractionation with a supercritical fluid, more preferably fractionation with scCO.sub.2.

(33) Fractionation with scCO.sub.2 is typically carried out according to conventional methods under isothermal conditions, progressively increasing the pressure. Typically, temperature is set at a value ranging from 40 C. to 150 C., while pressure is progressively increased from 8 to 50 MPa.

(34) Fractionation with scCO.sub.2 allows to completely remove from mixture (M4) residual PFPE (P.sub.pol) and also to obtain a mixture [mixture (M5)] with an increased amount of dangling, spiro and ansa PFPE (CP-1), (CP-2) and (CP-3) with respect to the corresponding bridged PFPE (CP-4). As it will be clearer from the results reported in the experimental section, any residual PFPE (P.sub.pol) is eluted first, followed by fractions containing PFPE (CP-1), (CP-2) and (CP-3) (herein after intermediate fractions); PFPE (CP-4) is eluted last. Among the intermediate fractions, those which are eluted first contain a higher amount of PFPE (CP-2) and (CP-3), while those eluted last contain a higher amount of dangling PFPE (CP-1). All intermediate fractions can be pooled together and used as such; otherwise, only intermediate fractions containing a higher amount of dangling PFPE (CP-1) can be pooled together and, optionally, be re-submitted to fractionation with supercritical fluid in order to further increase purity; this process can be repeated as many times as desired in order to increase purity according to the intended use. Bridged PFPE (CP-4), which can be isolated by fractionation of mixture (M4), is also encompassed in the scope of the present invention.

(35) Mixtures (M5), which contain dangling PFPE (CP-1) together with PFPE spiro- and ansa-PFPE (CP-2) and (CP-3), obtainable by means of the above purification method, in particular mixtures obtainable through a process comprising steps 1)-6) as defined above, are also within the scope of the present invention. These mixtures are characterised by a molar content of PFPE (CP-1) of at least 40%.

(36) Preferred are mixtures (M5) obtainable from PFPE diols (II) as defined above and phosphazenes (CP.sub.halo)-(IA) and/or (IB) as defined above.

(37) Particularly preferred are mixtures (M5) obtainable from (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, and a PFPE diol (IIA) or with a mixture of PFPE (P.sub.pol) (IIA)-(IIC) as defined above.

(38) More particularly preferred are mixtures (M5) obtainable from (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, and a PFPE diol (IIA) PFPE diol (IIA) wherein both Y and Y are CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH, wherein n is as defined above.

(39) More particularly preferred are also mixtures (M5) obtainable from (CP.sub.halo)-(IA), preferably hexachlorocyclophosphazene, and a protected mixture of PFPE (P.sub.pol) (IIA)-(IIC) as defined above.

(40) Mixtures (M3)-(M5) according to the present invention can be used as such in the lubrication of MRM or they can be added with further ingredients and/or additives to provide further lubricant compositions for MRM. Thus, a further object of the present invention is a method for lubrifying MRM comprising using any one of mixtures (M3)-(M5) as defined above, alone or in the form of compositions containing further ingredients and/or additives. Particularly preferred is the use of mixtures (M4) and (M5), the use of mixtures (M5) being particularly preferred.

(41) The invention will be illustrated in greater detail in the following experimental section.

(42) Should the disclosure of any patents, patent applications, and publications which are incorporated herein by reference conflict with the description of the present application to the extent that it may render a term unclear, the present description shall take precedence.

EXPERIMENTAL SECTION

Material and Methods

(43) Materials

(44) PFPE (P.sub.pol)-(IIA) used in example 1 was obtained by multiple thin layer distillations under vacuum of a commercial grade of Fomblin Z-DOL PFPEs, until obtainment of a PFPE (P.sub.pol) having EW=541, a1/a2=1.0, a1 and a2 being selected in such a way as to obtain Mn=1040, and Mw/Mn=1.10.

(45) PFPE (P.sub.pol)-(IIA) used in example 2, was obtained by multiple thin layer distillation under vacuum of commercial Fluorolink E10H PFPE, until obtainment of a PFPE (P.sub.pol) having EW=621, a1/a2=0.9, a1 and a2 being selected in such a way as to obtain Mn=1224 and Mw/Mn=1.10.

(46) The mixture of PFPE polyols (P.sub.pol) (IIA)-(IIC) used in example 3 was prepared from the PFPE (P.sub.pol)-(IIA) used in example 1 following the procedure described in example 1 of EP 2197939, with the difference that conversion was limited to 30%, so as to obtain a mixture of diol (IIA), ketal-protected PFPE triol (IIB) and ketal-protected tetraol (IIC) having a hydroxyl equivalent weight of 773 g/eq.

(47) Hexachlocyclotriphosphazene (HCP) was purchased from Strem

(48) Chemicals Inc. with a 98.5% purity.

(49) 1,3-hexafluoroxylene was obtained from Miteni SpA.

(50) HCl, NaOH 50% and isobutyl alcohol were reagent grade chemicals and they were used as received.

(51) Analytical Methods

(52) NMR Spectroscopy

(53) All NMR experiments were recorded on an Agilent System 500 operating at 499.86 MHz for .sup.1H and 470.30 MHz for .sup.19F and equipped with a 5-mm triple resonance .sup.1H, .sup.19F{.sup.13C,.sup.31P} PFG Agilent probe with a single axis (Z) gradient coil. Samples have been acquired either neat either dissolved in a mixture 3:1 v/v CFC113/Methanol-d4 (CD.sub.3OD) 99.9 atom % D at about 10% w/w. Fluorine spectra have been referenced against external CFCl.sub.3, phosphorous spectra against external H.sub.3PO.sub.4, whereas proton and carbon spectra has been referenced with respect to the residual solvent signal (Methanol-d4) at 3.3 ppm and 49.0 ppm respectively.

(54) 19F-NMR. The following fluorine acquisition parameters were applied on the neat samples: sample temperature of 25 C., sample spinning rate of 20 Hz, relaxation delay of 6.0 s, 90 flip angle corresponding to a pulse duration of 9.2 ms, at least 256 transients, and 65536 complex free induction decay (FID) data points acquired over a spectral width of 96153 Hz (acquisition time 0.6 s). Prior to Fourier transformation, all time domain data were processed with an exponential window function using a line broadening factor of 2 Hz.

(55) 1H-NMR. The following proton acquisition parameters were applied on the dissolved samples: sample temperature of 25 C., sample spinning rate of 20 Hz, relaxation delay of 20.0 s, 90 flip angle corresponding to a pulse duration of 9.0 ms, at least 64 transients, and 16384 complex free induction decay (FID) data points acquired over a spectral width of 8013 Hz (acquisition time 2.05 s). No weighting functions were applied.

(56) 31P-NMR. The following phosphorous acquisition parameters were applied on the neat samples: sample temperature of 25 C., sample spinning rate of 20 Hz, relaxation delay of 15.0 s, 90 flip angle corresponding to a pulse duration of 14.8 ms, at least 64 transients, and 16384 complex free induction decay (FID) data points acquired over a spectral width of 19841 Hz (acquisition time 0.83 s). Proton decoupling (WALTZ-16 scheme) was also applied during acquisition to cut out all possible .sup.1H-.sup.31P coupling constant. Prior to Fourier transformation, all time domain data were processed with an exponential window function using a line broadening factor of 2 Hz.

(57) 13C-NMR. The following carbon acquisition parameters were applied on the dissolved samples: sample temperature of 25 C., sample spinning rate of 20 Hz, relaxation delay of 0.1 s, 45 flip angle corresponding to a pulse duration of 5.95 ms, at least 10000 transients, and 32768 complex free induction decay (FID) data points acquired over a spectral width of 31250 Hz (acquisition time 1.05 s). No weighting functions were applied. Proton decoupling (WALTZ-16 scheme) was also applied during the whole sequence acquisition to cut out all .sup.1H-.sup.13C coupling constant and increase peak intensity due to de nOe. Prior to Fourier transformation, all time domain data was processed with an exponential window function using a line broadening factor of 2 Hz.

(58) Definition and Determination of the Ratio R

(59) R is defined as the ratio between POCH.sub.2 groups (P is the phosphorus atom in the cyclophosphazene ring) and the overall amount of functional and non-functional end groups. In pure dangling PFPE (CP-1) R is 1, while in spiro-, ansa- and bridged-PFPE (CP-2), (CP-3) and (CP-4), the ratio is higher than 1.

(60) The estimation of the ratio R has been performed by using proton, carbon and optionally fluorine spectra which show distinct peaks for POCH.sub.2 and free hydroxy groups.

(61) 4.4.2 Gel Permeation Chromatography (GPC)

(62) Molecular weight distribution, Mn and Mw averages and polydispersity were determined by Gel Permeation Chromatography (GPC).

(63) The GPC system was equipped with a Waters HPLC 515 pump, three PL-Gel columns (one Mixed-D and two Mixed-E) and a Waters 2414 refractive index detector. The columns and detector were thermostated at 35 C.

(64) The mobile phase was a mixture of 1,3-bis(trifluoromethyl)benzene and isopropanol (80/20 vol.), fluxed at 1.0 ml/min. Samples were dissolved at 1% wt/vol concentration in the mobile phase under stirring at room temperature until complete dissolution (about 1 hour). For the analysis 200 ml of the solution were injected.

(65) The calibration curve was obtained by using seven Fomblin Z DOL PFPE narrow fractions with molecular weights known from NMR analysis and falling in the range 460-9200. Acquisition and the calculations were performed using Waters Empower software.

EXAMPLES

Example 1

Manufacture of a Mixture (M4) From a PFPE Diol (IIA) and Hexachlorocyclophosphazene and Fractionation by scCO2

(66) Step 1Mixture (M)

(67) 540 g of PFPE diol (IIA) of formula:

(68) HOCH.sub.2CF.sub.2O(CF.sub.2CF.sub.2O).sub.a1(CF.sub.2O).sub.a2CF.sub.2CH.sub.2OH (EW 541 g/eq; 998.15 meq.) was charged into a 1 l round-bottomed flask equipped with mechanical stirrer, dropping funnel, thermometer and refrigerant, then added with 14.84 g KOH (132.26 meq.; 50% solution in water). The mixture was heated and maintained at 80 C. under stirring, then vacuum was applied by means of a mechanical pump until complete elimination of water (about 2 hours at P=10 Pa), thereby obtaining a clear solution.

(69) Step 2)Mixture (M1)

(70) In a separated flask 3.50 g hexachlorocyclotriphosphazene (HCP, 60.40 meq.) was dissolved under nitrogen atmosphere in 108 g 1,3-hexafluoroxylene (HFX); the solution was poured into the dropping funnel and slowly added to the solution from step 1) under stirring at 80 C. during 5 hours.

(71) Step 3)Mixture (M2)

(72) HFX was then distilled under vacuum and the reaction mixture from step 2) was maintained at 80 C. under stirring, controlling the conversion from time to time by .sup.31P-NMR analysis. After about 2 hours the conversion was quantitative (singlet in the .sup.31P-NMR at 17 ppm) and the reaction was stopped.

(73) Step 4)Mixture (M3)

(74) After cooling, mixture (M2) was added with 140 g distilled water, 16 g HCl 37% w/w water solution and 23 g isobutyl alcohol. The resulting two phases were vigorously stirred at 50 C. for 30 minutes and, after separation, the lower organic layer was collected. The solvents (isobutanol and traces of water) were removed by distillation at 80 C. under reduced pressure to afford 534 g crude product, containing a large amount of unreacted PFPE diol (IIA).

(75) Step 5)

(76) Most of diol (IIA) was then removed by molecular distillation under a residual pressure of 1.8 Pa (two stages at 120 C. and 150 C. respectively), obtaining two low-viscous fractions (61% and 25% by weight, respectively) of substantially pure PFPE diol (IIA), as confirmed by the absence of signals in the .sup.31P-NMR spectrum. The high boiling residue (74.8 g) was characterized by .sup.19F-NMR, .sup.1H-NMR, .sup.31P-NMR and GPC analysis.

(77) The GPC chromatogram shows three main components having a peak molecular weight of 1836, 6539 and 10995 dalton respectively. The first component corresponds to residual PFPE diol (IIA), the second component is attributed to dangling-, spiro- and ansa-PFPE (CP-1), (CP-2) and (CP-3), while the last component is most likely attributed to a bridged PFPE (CP-4).

Step 6Fractionation of Mixture (M4) With scCO2Obtainment of Mixture (M5)

(78) Mixture (M4) obtained from step 5 was charged into a 300 ml SFT-150 Supercritical CO.sub.2 Extraction System and heated at 100 C. Through a step-by-step pressure increase (from 18 to 30 MPa) and operating at a CO.sub.2 flow rate of 4 NI/min, 13 fractions were collected. Each fraction was characterized by .sup.31P-NMR, .sup.19F-NMR, .sup.1H-NMR, .sup.13C-NMR and GPC. The GPC analysis of the fractions shows that residual PFPE diol (IIA) and bridged PFPE (CP-4) were selectively removed at lower and higher pressures respectively. Fractions 3 to 9 (33.9 g), containing only PFPE (CP-1), (CP-2) and (CP-3), were separately washed three times with water/isobutyl alcohol and after phase separation residual solvents were carefully removed. The GPC analyses of fractions 3 to 9 showed in all cases a single peak having average molecular weight Mn of about 5900 and confirm the absence of PFPE diol (IIA) and bridged PFPE (CP-4).

(79) NMR analyses confirmed the structure of the PFPE (CP-1)-(CP-3);

(80) particularly significant are the signals corresponding to POCH.sub.2CF.sub.2O moieties: .sup.19F: 78.8, 80.8 ppm; .sup.1H: 4.31 ppm; .sup.13C: 65.5 ppm (CH.sub.2) and those corresponding to CF.sub.2CH.sub.2OH moieties: .sup.19F: 81.1, 83.1 ppm; .sup.1H: 3.80 ppm; .sup.13C: 63.0 ppm (CH.sub.2). The ratio R between the POCH.sub.2 CF.sub.2O and the OCF.sub.2X end groups (X=CH.sub.2OH, F or H, measured by .sup.19F-NMR, .sup.1H-NMR and .sup.13C-NMR) was found to be higher than 1 in all fractions, indicating that the product is a mixture (M4) of dangling PFPE (CP-1) and spiro- and ansa-PFPE (CP-2) and (CP-3).

(81) From the R ratio it is possible to calculate the composition of each fraction, which is reported in Table 1.

(82) TABLE-US-00001 TABLE 1 Fraction Molar composition (%) number R (CP-1) (CP-2)/(CP-3) 3 1.24 43 57 4 1.20 50 50 5 1.18 54 46 6 1.14 63 37 7 1.13 66 34 8 1.10 73 27 9 1.07 79 21

(83) The above data indicate that it is possible to increase the amount of PFPE (CP-1) with respect to PFPE (CP-2) and (CP-3) by fractionation with scCO.sub.2. An even further increase can be achieved by collecting the fractions with a higher content of (CP-1) and submitting the same to further scCO.sub.2 fractionation cycles.

Example 2

Manufacture of a Mixture (M4) From a PFPE Diol (IIA) and Hexachlorocyclophosphazene and of a Mixture (M5) by Fractionation with scCO2

(84) Step 1Mixture (M)

(85) 188 g of an ethoxylated PFPE diol of formula:
H(OCH.sub.2CH.sub.2).sub.nOCH.sub.2CF.sub.2O(CF.sub.2CF.sub.2O).sub.a1(CF.sub.2O).sub.a2CF.sub.2CH.sub.2O(CH.sub.2CH.sub.2O).sub.nH (EW 621 g/eq; 302.7 meq; n=1.5).
were charged into a 0.5 l round-bottomed flask equipped with a mechanical stirrer, a dropping funnel, a thermometer and a refrigerant, and then added with 8.85 g KOH (78.9 meq; 50% solution in water). The mixture was heated and maintained at 80 C. under stirring, then vacuum was applied to the reactor by means of a mechanical pump until complete elimination of water (about 30 minutes at P=4 Pa), to obtain homogeneous, slightly opalescent solution.

(86) Step 2Mixture (M1)

(87) In a separated flask 2.11 g hexachlorocyclotriphosphazene (HCP, 36.4 meq.) was dissolved in 53 g 1,3-hexafluoroxylene (HFX); the resulting solution was poured into the dropping funnel and slowly added to mixture (M) from step 1), under stirring at 80 C. during 2 hours.

(88) Step 3Mixture (M2)

(89) HFX was then distilled off under vacuum and the reaction mixture was maintained at 80 C. under stirring controlling the conversion from time to time by .sup.31P-NMR analysis. After 30 minutes the conversion was quantitative (singlet in the .sup.31P-NMR spectrum at 17 ppm) and the reaction was stopped.

(90) Step 4)Mixture (M3)

(91) After cooling at room temperature, the mixture was added with 170 g distilled water, 11 g HCl 37% w/w solution in water and 34 g isobutyl alcohol. The two phases were vigorously shaken and, after separation, the lower organic layer was collected and the solvents were removed by distillation at 80 C. under reduced pressure to afford 182 g crude product, which contains unreacted ethoxylated PFPE diol and a mixture of dangling, spiro, ansa and bridged PFPE complying with formulae (CP-1)-(CP-4) wherein R.sub.f is (CF.sub.2CF.sub.2O).sub.a1(CF.sub.2O).sub.a2, wherein p and q are as defined above and T-O and T-O are both O(CH.sub.2CH.sub.2O).sub.n with n=1.5.

(92) Step 5Mixture (M4)

(93) Most of the unreacted ethoxylated PFPE diol was removed in two passages by molecular distillation under a residual pressure of 2.2 Pa at 160 C. and 190 C., respectively. Two low-viscous fractions (54% and 19% by weight, respectively) of only ethoxylated PFPE diol, as confirmed by the absence of signals in the .sup.31P-NMR spectrum, were removed, leaving 49 g of a high boiling, low volatility residue [mixture (M4)], which was characterized by .sup.19F-NMR, .sup.1H-NMR and .sup.31P-NMR.

(94) Step 6Preparation of Mixture (M5) by Fractionation of Mixture (M4) With scCO2

(95) Mixture (M4) from step 5) was charged into a 300 ml SFT-150 scCO.sub.2 Extraction System and heated at 100 C. Through a step-by-step increase of pressure (from 20 to 35 MPa) and operating at a CO.sub.2 flow rate of 4 NI/min, dangling, spiro and ansa PFPE (CP1)-(CP3) were isolated. Any residual unreacted ethoxylated PFPE diol was easily removed at scCO.sub.2 low pressure, while bridged PFPE (CP-4) was selectively collected at high pressure. Each fraction was characterized by .sup.31P-NMR, .sup.19F-NMR, .sup.1H-NMR, .sup.13C-NMR and GPC. Fractions containing only PFPE (CP1)-(CP3) were pooled (overall yield: 4 g). Ratio R, measured by .sup.19F-NMR, .sup.1H-NMR and .sup.13C-NMR, was found to be 1.22, corresponding to a molar percent composition of 46% PFPE (CP-1) and 54% PFPE (CP-2)+(CP-3).

Example 3

Manufacture of a Mixture (M4) from a Protected Mixture of PFPE (Ppol) (IIA)-(IIC) Mixture and Hexachlorocyclophosphazene and Manufacture of a Mixture (M5) by Fractionation with scCO2

(96) Step 1Mixture (M)

(97) 635 g of ketal-protected mixture of PFPE (P.sub.pol) (IIA)-(IIC) (EW=773, 821.5 meq) was charged into a 1 liter round bottom flask equipped with a mechanical stirrer, a dropping funnel, a thermometer and a refrigerant, and 22.36 g KOH (50% wt solution in water, 199.3 meq.) was added. The mixture was stirred and heated with an external bath to 80 C., then vacuum was applied by means of a mechanical pump until complete elimination of water (about half an hour).

(98) Step 2Mixture (M1)

(99) In a separated flask 5.26 g hexachlorocyclotriphosphazene (90.77 meq) was dissolved in 164 g HFX; the solution was poured into the dropping funnel and slowly added to mixture (M) from step 1) under stirring at 80 C. during 3.5 hours.

(100) Step 3Mixture (M2)

(101) The reaction mixture was maintained at 80 C. under stirring controlling the conversion from time to time by .sup.31P-NMR analysis. After 30 minutes the conversion was quantitative (singlet in the .sup.31P-NMR spectrum at 17 ppm) and the reaction was stopped.

(102) Step 4Hydrolysis and Deprotection to Obtain Mixture (M3)

(103) The mixture was then added with 140 g distilled water, 21 g HCl 37% w/w water solution and 21 g isobutyl alcohol. The two phases were vigorously shaken for 1 h at 40 C. and, after separation, the lower organic layer was collected. The solvents (HFX and isobutyl alcohol) were removed by distillation at 80 C. under reduced pressure to afford 612 g of crude product.

(104) The crude product was then added with 200 g methanol, 78 g distilled water and 37 g HCl 37% w/w water solution, and subsequently heated at 70 C. and stirred during 3 hours, in order to completely remove the protective groups. After phase separation, the lower organic layer was collected and the solvent was removed by distillation at 80 C. under reduced pressure, to afford 590 g crude product which was characterized by .sup.31P-NMR, .sup.19F-NMR and .sup.1H-NMR.

(105) Step 5Mixture (M4)

(106) Most of the unreacted precursors PFPE (P.sub.pol) (IIA)-(IIC) were removed in two passages by molecular distillation under a residual pressure of 1.6 Pa at 180 C. and 200 C., respectively. Two fractions, corresponding to 81% by weight of only unreacted PFPE (P.sub.pol) (IIA)-(IIC), as confirmed by the absence of signals in the .sup.31P-NMR spectrum, were removed, leaving 112 g of a high boiling, low volatility residue, which was characterized by .sup.19F-NMR, .sup.1H-NMR, .sup.31P-NMR and GPC.

(107) Step 6Preparation of Mixture (M5) by Fractionation of Mixture (M4) With scCO2

(108) Mixture (M4) from step 5) was charged into a 300 ml SFT-150 scCO.sub.2 Extraction System and heated at 100 C. Through a step-by-step increase of pressure (from 19.5 to 30 MPa) and operating at a CO.sub.2 flow rate of 4 NI/min, the dangling, spiro and ansa PFPE (CP1)-(CP3) were isolated. Any residual unreacted PFPE (P.sub.pol) (IIA)-(IIC) and tetraol (IIB) were easily removed at scCO.sub.2 low pressure, while bridged PFPE (CP4) was selectively collected at high pressure. Each fraction was characterized by .sup.31P-NMR, .sup.19F-NMR, .sup.1H-NMR, .sup.13C-NMR and GPC. Fractions containing only PFPE (CP1)-(CP3) were pooled together (9.2 g). The ratio R between the POCH.sub.2CF.sub.2O and the OCF.sub.2X end groups (X=CH.sub.2OH, CH.sub.2OCH.sub.2CH(OH)CH.sub.2OH, F or H, measured by .sup.19F-NMR, .sup.1H-NMR and .sup.13C-NMR) was found to be 1.21, corresponding to a molar percent composition of 48% (CP-1) and 52% (CP-2)+(CP-3). The ratio between CH.sub.2OH and CH.sub.2OCH.sub.2CH(OH)CH.sub.2OH end-groups resulted to be 1.34.