WO2024227875A1

WO2024227875A1 - Method for preparing onium fluorides

Info

Publication number: WO2024227875A1
Application number: PCT/EP2024/062117
Authority: WO
Inventors: Ruediger STUMPF; Harald Schaefer; Frederic Lucas
Original assignee: BASF SE
Current assignee: BASF SE
Priority date: 2023-05-03
Filing date: 2024-05-02
Publication date: 2024-11-07
Anticipated expiration: 2025-11-03

Abstract

The present invention relates to a method for preparing an onium fluoride (I) R₄Q⁺ F^- · n×HF (I), where the variables are as defined in the claims and the description, from a solution of the corresponding onium chloride, comprising determining the amount of hydrogen chloride contained in the solution of said onium chloride, neutralizing the same, reacting the neutralized solution with a fluoride salt and then, after having removed any precipitate formed, with anhydrous hydrogen fluoride.

Description

Method for preparing onium fluorides

The present invention relates to a method for preparing an onium fluoride (I) R₄Q⁺ F- ■ nxHF (I), where the variables are as defined below, from a solution of the corresponding onium chloride, comprising determining the amount of hydrogen chloride contained in the solution of said onium chloride, neutralizing the same, reacting the neutralized solution with a fluoride salt and then, after having removed any precipitate formed, with anhydrous hydrogen fluoride.

TECHNICAL BACKGROUND

Onium fluorides are versatile compounds that find use, for example, as catalysts for preparing non-symmetric trimeric polyisocyanates containing iminooxadiazindione groups from organic di- or polyisocyanates, or as organic soluble sources for fluoride ions. Especially tetrabutylammonium fluoride is used as promotor of reactions involving organosilyl derivatives, and in various elimination, condensation and fluorination reactions.

The onium fluorides are generally prepared by an anion exchange.

D. Landini et al. describe in Synthesis, 1988 (12), 953-955 the synthesis of onium fluorides, hydrogendifluorides and dihydrogentrifluorides by either preparative ion pair extraction from the corresponding hydrogensulfates in a benzene/saturated aqueous KF/KOH two-phase system or by neutralizing a chloroform solution of the corresponding hydrogensulfates with an aqueous potassium hydrogencarbonate solution and then reacting with stoichiometric amounts of potassium hydrogenfluoride (KHF2), or else by neutralizing a chloroform, methylene chloride or benzene solution of the corresponding onium chlorides or bromides with an aqueous potassium hydrogencarbonate solution and then reacting with a 50-fold excess of potassium hydrogenfluoride (KHF2).

Onium chlorides are more readily available than onium hydrogensulfates and therefore more interesting as starting material, especially for syntheses on an industrial scale. The use of a 50-fold excess of the fluoride source, as requested in the Landini article when onium chlorides are used as starting materials, is however not economical.

EP 0962455 A1 relates to the preparation of trimeric polyisocyanates containing at least 30 mol-% of non-symmetric iminooxadiazindione groups by trimerization of organic di- or polyiisocyanates using quaternary phophonium fluorides R₄P⁺ F' ■ n(HF) as catalyst. Tetra-n-butylphosphonium fluoride ■ n(HF) and tributyl-tetradecylphospho- nium fluoride ■ n(HF) are in turn prepared by adding a slight molar excess of KF to a methanolic solution of tetra-n-butylphosphonium chloride or tributyl-tetradecylphospho- nium chloride, stirring for 24 h at room temperature, filtering, washing the precipitate, adding another slight molar excess of KF to the unified filtrates, stirring for another 24 h at room temperature, filtering, washing the precipitate, removing the major part of the solvents and filtering again. The obtained product is then reacted with anhydrous HF.

Similarly, US 6,090,939 relates to the preparation of trimerized polyisocyanates containing at least 30 mol-% of non-symmetric iminooxadiazindione groups by trimerization of organic di- or polyiisocyanates using quaternary phophonium fluorides R₄P⁺ F’ ■ n(HF) as catalyst. Tetra-n-butylphosphonium fluoride ■ n(HF) and tributyltetradecylphosphonium fluoride ■ n(HF) are in turn prepared by by adding a slight molar excess of KF to a methanolic solution of tetra-n-butylphosphonium chloride or tributyltetradecylphosphonium chloride, stirring for 24 h at room temperature, filtering, washing the precipitate, adding another slight molar excess of KF to the unified filtrates, stirring for another 24 h at room temperature, filtering, washing the precipitate, removing the major part of the solvents and filtering again. The obtained product is then reacted with anhydrous HF.

This proceeding of EP 0962455 A1 and US 6,090,939 is tedious and still requires a rather large excess of the fluoride source, albeit not as large as in the above-mentioned method of Landini et al. Moreover, if the reaction is carried out in steel reactors, premature corrosion is observed.

It was the object of the present invention to provide a simpler and more economic method for preparing onium fluorides from the corresponding chlorides, which is less tedious and requires less fluoride source.

SUMMARY OF THE INVENTION

Onium chlorides generally contain a small amount of HOI. The present inventors found that the object of the invention is reached if the onium chloride starting material is first (quantitatively) analyzed with respect to its HOI content and the HOI present therein is neutralized before the anion exchange is carried out. This measure surprisingly allows to reduce the amount of the fluoride source significantly and to avoid repetitive additions thereof. Moreover, the neutralization of the HCI content in the starting solution allows to carry out the reaction in steel reactors, since the corrosive effect of HCI, which occurs even if the latter is present in seemingly negligible amounts, is avoided. This is a significant advantage, especially for larger scale syntheses, since due to the fluorides present in the method of the invention, reactors lined with ceramics cannot be used either.

The present invention relates thus to a method for producing an onium fluoride of the formula (I)

R₄Q⁺ F- ■ nxHF (I) where

Q is N (i.e. , the onium fluoride is an ammonium fluoride) or P (i.e. the onium fluoride is a phosphonium fluoride); each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Cs-cy- cloalkyl, Cs-Cs-cycloalkyl-Ci-C^alkyl, Ce-C -aryl, C6-C -aryl-Ci-C4-alkyl, and a 3- to 8-membered saturated, partially unsaturated or maximally unsaturated heterocyclic ring containing 1 , 2 or 3 heteroatoms or heteroatom groups selected from the group consisting of O, N, S, S(O) and S(O)2; or two R form together a C2-C2o-alkylene bridging group which may be interrupted by 1, 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1 , 2 or 3 aromatic or heteroaromatic rings; and the other two R are either as defined above or, independently, also form together a C2-C2o-alkylene bridging group which may be interrupted by 1 , 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1, 2 or 3 5- or 6-membered aromatic or heteroaromatic rings; and n is from 0.1 to 5; which method comprises

(i) providing a solution of an onium chloride of the formula (II)

R₄Q⁺ C (II) where Q and R are as defined above, in a solvent;

(ii) determining the amount of hydrogen chloride contained in said solution of the onium chloride (II);

(iii) neutralising the hydrogen chloride contained in said solution of the onium chloride of the formula (II) by reaction with a base, optionally after diluting the solution of step (i) with a solvent;

(iv) reacting the reaction mixture obtained in step (iii) with a fluoride salt of the formula (M^m+)(F )m, where M is an alkali metal or alkaline earth metal and m is the charge (valency) thereof, m being thus 1 if M is an alkali metal and 2 if M is an alkaline earth metal; (v) removing any precipitate present in the reaction mixture obtained in step (iv);

(vi) partially or completely removing the solvent; and

(vii) reacting the reaction mixture obtained in step (vi) with anhydrous hydrogen fluoride.

DETAILED DESCRIPTION OF THE INVENTION

Definitions

“Onium” as used in the present invention stands generically for ammonium and phosphonium.

The anion in the onium salt (I), depicted as F’ ■ nxHF, is statistically a polyhydrogenfluoride (a HF adduct to the fluoride salt) and could alternatively be depicted as F ■ (HF)_n or F x (HF)_n or as H_nF_n+i’.

The organic moieties mentioned above and below are collective terms for individual listings of the individual group members. The prefix C_n-C_m indicates in each case the possible number of carbon atoms in the group.

The term "alkyl" is used herein in the usual sense and refers to saturated straightchain (linear) or branched acyclic hydrocarbon radicals having 1 or 2 ("Ci-C2-alkyl"), 1 to 4 ("Ci-C₄-alkyl"), 1 to 10 ("Ci-C -alkyl"), 2 to 10 ("C₂-Cio-alkyl") or 1 to 20 ("C1-C20- alkyl”) carbon atoms. Ci-C2-Alkyl denotes a saturated acyclic aliphatic radical with 1 or 2 carbon atoms. Examples are methyl and ethyl. Ci-C4-Alkyl denotes a saturated linear or branched acyclic aliphatic radical with 1 to 4 carbon atoms. Examples are methyl, ethyl, n-propyl, isopropyl, n-butyl, sec-butyl, isobutyl and tert-butyl. Ci-C -Alkyl denotes a saturated linear or branched acyclic aliphatic radical with 1 to 10 carbon atoms. Examples are, in addition to those mentioned for Ci-C4-alkyl, n-pentyl, 1-methylbutyl, 2- methylbutyl, 3-methylbutyl, 2,2-dimethylpropyl, 1 -ethylpropyl, 1 ,1 -dimethylpropyl, 1 ,2- dimethylpropyl, n-hexyl, 1 -methylpentyl, 2-methylpentyl, 3-methyl pentyl, 4-methylpen- tyl, 1 ,1 -dimethylbutyl, 1 ,2-dimethylbutyl, 1 ,3-dimethylbutyl, 2,2-dimethylbutyl, 2,3-dime- thylbutyl, 3,3-dimethylbutyl, 1-ethylbutyl, 2-ethylbutyl, 1 , 1 ,2-trimethylpropyl, 1 ,2,2-trime- thylpropyl, 1-ethyl-1 -methylpropyl, 1-ethyl-2-methyl propyl, n-octyl, 2-ethylhexyl, n- nonyl, n-decyl and structural isomers thereof. C2-C -Alkyl denotes a saturated linear or branched acyclic aliphatic radical with 2 to 10 carbon atoms. Examples are those mentioned above for Ci-C -alkyl, but for methyl. Ci-C2o-Alkyl denotes a saturated linear or branched acyclic aliphatic radical with 1 to 20 carbon atoms. Examples are, in addition to those mentioned for Ci-C -alkyl, n-undecyl, n-dodecyl, n-tridecyl, n-tetradecyl, n- pentadecyl, n-hexadecyl, n-heptadecyl, n-octadecyl, n-nonadecyl, n-eicosyl and structural isomers thereof.

The term "cycloalkyl" as used herein denotes in each case a mono- or bicyclic, saturated cycloaliphatic radical having usually from 3 to 6 carbon atoms (= Cs-Ce-cyclo- alkyl) or 3 to 8 carbon atoms (= Cs-Cs-cycloalkyl) as (only) ring members. Examples of monocyclic saturated cycloaliphatic radicals having 3 to 6 carbon atoms comprise cyclopropyl, cyclobutyl, cyclopentyl and cyclohexyl. Examples of monocyclic saturated cycloaliphatic radicals having 3 to 8 carbon atoms comprise cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl and cyclooctyl. Examples of bicyclic radicals having 5 to 8 carbon atoms comprise bicyclo[1.1.1]pentyl, bicyclo[2.1.1]hexyl, bicyclo[2.2.1]hep- tyl, bicyclo[3.1.1]heptyl, bicyclo[2.2.2]octyl and bicyclo[3.2.1]octyl. Preferably, cycloalkyl is monocyclic.

C3-C6-Cycloalkyl-Ci-C2-alkyl is a Ci-C2-alkyl group, as defined above, in which one hydrogen atom is replaced by a Cs-Ce-cycloalkyl ring, as defined above. Examples are cyclopropylmethyl, cyclobutylmethyl, cyclopentyl methyl, cyclohexylmethyl, 1-cyclo- propylethyl, 1 -cyclobutylethyl, 1 -cyclopentylethyl, 1 -cyclohexylethyl, 2-cyclopropylethyl, 2-cyclobutylethyl, 2-cyclopentylethyl and 2-cyclohexylethyl.

Cs-Cs-Cycloalkyl-Ci-C^alkyl is a Ci-C4-alkyl group, as defined above, in which one hydrogen atom is replaced by a Cs-Cs-cycloalkyl ring, as defined above. Examples are, in addition to those mentioned for C3-C6-cycloalkyl-Ci-C2-alkyl, cycloheptylmethyl, cyclooctylmethyl, 1 -cycloheptylethyl, 1 -cyclooctylethyl, 2-cycloheptylethyl, 2-cy- clooctylethyl, 1-cyclopropyl-1-propyl, 1-cyclobutyl-1-propyl, 1 -cyclopentyl- 1 -propyl, 1- cyclohexyl-1 -propyl, 1 -cycloheptyl- 1 -propyl, 1 -cyclooctyl- 1 -propyl, 2-cyclopropyl-1 -propyl, 2-cyclobutyl-1 -propyl, 2-cyclopentyl-1-propyl, 2-cyclohexyl-1 -propyl, 2-cycloheptyl-

1-propyl, 2-cyclooctyl-1 -propyl, 3-cyclopropyl-1-propyl, 3-cyclobutyl-1 -propyl, 3-cyclo- pentyl-1 -propyl, 3-cyclohexyl-1-propyl, 3-cycloheptyl-1-propyl, 3-cyclooctyl-1-propyl, 1- cyclopropyl-2-propyl, 1-cyclobutyl-2-propyl, 1-cyclopentyl-2-propyl, 1-cyclohexyl-2-pro- pyl, 1-cycloheptyl-2-propyl, 1-cyclooctyl-2-propyl, 2-cyclopropyl-2-propyl, 2-cyclobutyl-

2-propyl, 2-cyclopentyl-2-propyl, 2-cyclohexyl-2-propyl, 2-cycloheptyl-2-propyl, 2-cy- clooctyl-2-propyl, 1-cyclopropyl-2-propyl, 3-cyclobutyl-2-propyl, 3-cyclopentyl-2-propyl,

3-cyclohexyl-2-propyl, 3-cycloheptyl-2-propyl, 3-cyclooctyl-2-propyl and the like.

Ce-Cio-Aryl is phenyl, 1-naphthyl or 2-naphthyl.

C6-Cio-Aryl-Ci-C4-alkyl is a Ci-C4-alkyl group, as defined above, in which one hydrogen atom is replaced by a Ce-C -aryl group, as defined above. Examples are benzyl, 1 -phenylethyl, 2-phenylethyl, 1-phenylpropyl, 2-phenylpropyl, 3-phenylpropyl, 2- phenyl-2-propyl and the like.

The term "3- to 8-membered saturated, partially unsaturated or maximally unsaturated heterocyclic ring containing 1 , 2 or 3 heteroatoms or heteroatom groups selected from the group consisting of O, N, S, S(O) and S(O)2 as ring members " as used herein refers to monocyclic radicals, the monocyclic radicals being saturated, partially unsaturated or maximally unsaturated, including aromatic. The heterocyclic radical may be attached to the remainder of the molecule via a carbon ring member or via a nitrogen ring member. An unsaturated heterocycle contains at least one C-C and/or C-N and/or N-N double bond(s). Partially unsaturated heterocyclic rings contain less than the maximum number of C-C and/or C-N and/or N-N double bond(s) allowed by the ring size. A fully (or maximally) unsaturated heterocycle contains as many conjugated C-C and/or C-N and/or N-N double bonds as allowed by the size(s) of the ring(s). Maximally unsaturated 5- or 6-membered heteromonocyclic rings are generally aromatic. Exceptions are maximally unsaturated 6-membered rings containing O, S, SO and/or SO2 as ring members, such as pyran and thiopyran, which are not aromatic.

Examples for 3-, 4-, 5-, 6-, 7- or 8-membered saturated, partly unsaturated, fully unsaturated or aromatic heterocyclic rings are:

3-, 4-, 5-, 6-, 7- or 8-membered saturated heterocyclic rings: e.g., oxiranyl, aziridinyl, azetidinyl, 2 tetrahydrofuranyl, 3- tetrahydrofuranyl, 2 tetrahydrothienyl, 3 tetrahydrothienyl, 2-pyrrolidinyl, 3-pyrrolidinyl, 3 pyrazolidinyl, 4 pyrazolidinyl, 5- pyrazolidinyl, 2 imidazolidinyl, 4 imidazolidinyl, 2-oxazolidinyl, 4-oxazolidinyl, 5 oxazolidinyl, 3-isoxazolidinyl, 4 isoxazolidinyl, 5 isoxazolidinyl, 2 thiazolidinyl, 4- thiazolidinyl, 5-thiazolidinyl, 3 isothiazolidinyl, 4-isothiazolidinyl, 5 isothiazolidinyl, 1 ,2,4- oxadiazolidin-3-yl, 1 ,2,4 oxadiazolidin 5 yl, 1 ,2,4-thiadiazolidin-3-yl, 1 ,2,4 thiadiazolidin- 5-yl, 1,2,4 triazolidin-3-yl, 1,3,4-oxadiazolidin-2-yl, 1 ,3,4 thiadiazolidin-2-yl, 1,3,4 triazolidin-2-yl, 2-tetrahydropyranyl, 4 tetrahydropyranyl, 1 ,3-dioxan-5-yl, 1,4-dioxan-2- yl, 2-piperidinyl, 3-piperidinyl, 4-piperidinyl, 3-hexahydropyridazinyl, 4 hexahydropyridazinyl, 2-hexahydropyrimidinyl, 4-hexahydropyrimidinyl, 5 hexahydropyrimidinyl, 2-piperazinyl, 1 ,3,5-hexahydrotriazin-2-yl and 1,2,4 hexahydrotriazin-3-yl, 2-morpholinyl, 3-morpholinyl, 2-thiomorpholinyl, 3- thiomorpholinyl, 1-oxothiomorpholin-2-yl, 1-oxothiomorpholin-3-yl, 1,1- dioxothiomorpholin-2-yl, 1 ,1-dioxothiomorpholin-3-yl, hexahydroazepin-1-, -2-, -3- or -4- yl, hexahydrooxepinyl, hexahydro-1, 3-diazepinyl, hexahydro-1 , 4-diazepinyl, hexahydro- 1,3-oxazepinyl, hexahydro-1 , 4-oxazepinyl, hexahydro-1, 3-dioxepinyl, hexahydro- 1,4- dioxepinyl and the like. Examples of 3-, 4-, 5-, 6- or 7-membered partially unsaturated heterocyclic rings include: 2,3-dihydrofur-2-yl, 2,3-dihydrofur-3-yl, 2,5-dihydrofur-2-yl,

2.5-dihydrofur-3-yl, 2,3-dihydrothien-2-yl, 2,3 dihydrothien-3-yl, 2,5 dihydrothien-2-yl,

2.5-dihydrothien-3-yl, 2-pyrrolin-2-yl, 2-pyrrolin-3-yl, 3 pyrrolin-2-yl, 3-pyrrolin-3-yl, 2- isoxazolin-3-yl, 3-isoxazolin-3-yl, 4 isoxazolin 3 yl, 2-isoxazolin-4-yl, 3-isoxazolin-4-yl, 4-isoxazolin-4-yl, 2 isoxazolin-5-yl, 3-isoxazolin-5-yl, 4-isoxazolin-5-yl, 2-isothiazolin-3- yl, 3 isothiazolin-3-yl, 4-isothiazolin-3-yl, 2-isothiazolin-4-yl, 3-isothiazolin-4-yl, 4 isothiazolin-4-yl, 2-isothiazolin-5-yl, 3-isothiazolin-5-yl, 4-isothiazolin-5-yl, 2,3 dihydropyrazol-1-yl, 2,3-dihydropyrazol-2-yl, 2,3-dihydropyrazol-3-yl, 2,3 dihydropyrazol-4-yl, 2,3-dihydropyrazol-5-yl, 3,4-dihydropyrazol-1-yl, 3,4 dihydropyrazol-3-yl, 3,4-dihydropyrazol-4-yl, 3,4-dihydropyrazol-5-yl, 4,5 dihydropyrazol-1-yl, 4,5-dihydropyrazol-3-yl, 4,5-dihydropyrazol-4-yl, 4,5 dihydropyrazol-5-yl, 2,3-dihydrooxazol-2-yl, 2,3-dihydrooxazol-3-yl, 2,3 dihydrooxazol- 4-yl, 2,3-dihydrooxazol-5-yl, 3,4-dihydrooxazol-2-yl, 3,4 dihydrooxazol-3-yl, 3,4- dihydrooxazol-4-yl, 3,4-dihydrooxazol-5-yl, 3,4 dihydrooxazol-2-yl, 3,4-dihydrooxazol-3- yl, 3,4-dihydrooxazol-4-yl, 2-, 3-, 4-, 5- or 6-di- or tetrahydropyridinyl, 3-di- or tetrahydropyridazinyl, 4 di- or tetrahydropyridazinyl, 2-di- or tetrahydropyrimidinyl, 4-di- or tetrahydropyrimidinyl, 5 di- or tetrahydropyrimidinyl, di- or tetrahydropyrazinyl, 1 ,3,5- di- or tetrahydrotriazin-2-yl, 1 ,2,4-di- or tetrahydrotriazin-3-yl, 2, 3,4,5- tetrahydro[1 H]azepin-1-, -2-, -3-, -4-, -5-, -6- or -7-yl, 3,4,5,6-tetrahydro[2H]azepin-2-, -

3-, -4-, -5-, -6- or -7-yl, 2, 3,4, 7 tetrahydro[1 H]azepin-1-, -2-, -3-, -4-, -5-, -6- or -7-yl, 2, 3, 6, 7 tetrahydro[1 H]azepin-1-, -2-, -3-, -4-, -5-, -6- or -7-yl, tetrahydrooxepinyl, such as 2,3,4,5-tetrahydro[1 H]oxepin-2-, -3-, -4-, -5-, -6- or -7-yl, 2, 3, 4, 7 tetrahydro[1 H]oxepin-2-, -3-, -4-, -5-, -6- or -7-yl, 2, 3, 6, 7 tetrahydro[1 H]oxepin-2-, -3-, -

4-, -5-, -6- or -7-yl, tetrahydro-1 , 3-diazepinyl, tetrahydro-1 , 4-diazepinyl, tetra hydro- 1 ,3- oxazepinyl, tetrahydro-1 , 4-oxazepinyl, tetrahydro-1 , 3-dioxepinyl and tetrahydro- 1 ,4- dioxepinyl;

5-, 6-, 7- or 8-membered monocyclic partially unsaturated heterocycles: e.g., 2,3- dihydrofuran-2-yl, 2,3-dihydrofuran-3-yl, 2,5-dihydrofuran-2-yl, 2,5-dihydrofuran-3-yl,

2.3-dihydrothien-2-yl, 2,3-dihydrothien-3-yl, 2,5-dihydrothien-2-yl, 2,5-dihydrothien-3-yl, 2-pyrrolin-2-yl, 2-pyrrolin-3-yl, 3-pyrrolin-2-yl, 3-pyrrolin-3-yl, 2-isoxazolin-3-yl, 3-isoxa- zolin-3-yl, 4-isoxazolin-3-yl, 2-isoxazolin-4-yl, 3-isoxazolin-4-yl, 4-isoxazolin-4-yl, 2- isoxazolin-5-yl, 3-isoxazolin-5-yl, 4-isoxazolin-5-yl, 2-isothiazolin-3-yl, 3-isothiazolin-3- yl, 4-isothiazolin-3-yl, 2-isothiazolin-4-yl, 3-isothiazolin-4-yl, 4-isothiazolin-4-yl, 2-isothi- azolin-5-yl, 3-isothiazolin-5-yl, 4-isothiazolin-5-yl, 2,3-dihydropyrazol-1-yl, 2,3-dihydro- pyrazol-2-yl, 2,3-dihydropyrazol-3-yl, 2,3-dihydropyrazol-4-yl, 2,3-dihydropyrazol-5-yl,

3.4-dihydropyrazol-1-yl, 3,4-dihydropyrazol-3-yl, 3,4-dihydropyrazol-4-yl, 3,4-dihydropy- razol-5-yl, 4,5-dihydropyrazol-1-yl, 4,5-dihydropyrazol-3-yl, 4,5-dihydropyrazol-4-yl, 4,5- dihydropyrazol-5-yl, 2,3-dihydrooxazol-2-yl, 2,3-dihydrooxazol-3-yl, 2,3-dihydrooxazol- 4-yl, 2,3-dihydrooxazol-5-yl, 3,4-dihydrooxazol-2-yl, 3,4-dihydrooxazol-3-yl, 3,4-dihy- drooxazol-4-yl, 3,4-dihydrooxazol-5-yl, 3,4-dihydrooxazol-2-yl, 3,4-dihydrooxazol-3-yl,

3.4-dihydrooxazol-4-yl, 3,6-dihydro-2H-pyran-2-, -3-, -4-, -5- or 6-yl, 3,4-dihydro-2H-py- ran-2-, -3-, -4-, -5- or 6-yl, 3,6-dihydro-2H-thiopyran-2-, -3-, -4-, -5- or 6-yl, 3,4-dihydro- 2H-thiopyran-2-, -3-, -4-, -5- or 6-yl, 2-, 3-, 4-, 5- or 6-di- or tetrahydropyridinyl, 3-di- or tetrahydropyridazinyl, 4-di- or tetrahydropyridazinyl, 2-di- or tetrahydropyrimidinyl, 4-di- or tetrahydropyrimidinyl, 5-di- or tetrahydropyrimidinyl, di- or tetrahydropyrazinyl;

5-, 6-, 7- or 8-membered monocyclic fully unsaturated (including aromatic) heterocyclic ring: e.g., 2-furyl, 3-furyl, 2-thienyl, 3-thienyl, 1-pyrrolyl, 2-pyrrolyl, 3-pyrrolyl, 1- pyrazolyl, 3-pyrazolyl, 4-pyrazolyl, 5-pyrazolyl, 2-oxazolyl, 4-oxazolyl, 5-oxazolyl, 2-thi- azolyl, 4-thiazolyl, 5-thiazolyl, 1-imidazolyl, 2-imidazolyl, 4-imidazolyl, 5-imidazolyl, 2- pyridinyl, 3-pyridinyl, 4-pyridinyl, 3-pyridazinyl, 4-pyridazinyl, 2-pyrimidinyl, 4-pyrimidi- nyl, 5-pyrimidinyl, 2-pyrazinyl; pyran-2-yl, pyran-3-yl, pyran-4-yl, thiopyran-2-yl, thiopy- ran-3-yl and thiopyran-4-yl. Among these, all heterocycles, but for the pyran and thiopyran rings, are aromatic. If not specified otherwise, an aromatic ring in terms of the present invention is a benzene or a naphthalene ring.

If not specified otherwise, a heteroaromatic ring in terms of the present invention is a 5- or 6-membered heteroaromatic ring containing 1 , 2 or 3 heteroatoms selected from N, O and S as ring members. Examples are furan, thiophen, pyrrole, pyrazole, imidazole, triazole, oxazole, thiazole, isoxazole, isothiazole, oxadiazole, thiadiazole, pyridine, pyridazine, pyrimidine, pyrazine or triazine rings.

Alkylene is a linear or branched divalent alkanediyl radical. C2-C2o-Alkylene (or C2-C2o-alkyanediyl) is a linear or branched divalent alkyl radical having 2 to 20 carbon atoms. Examples are -CH₂CH₂-, -CH(CH₃)-, -CH₂CH₂CH₂-, -CH(CH₃)CH₂-, -CH₂CH(CH₃)-, -C(CH₃)₂-, -CH2CH2CH2CH2-, -CH(CH₃)CH₂CH₂-, -CH₂CH₂CH(CH₃)-, -C(CH₃) ₂CH₂-, -CH₂C(CH₃)₂-, -(CH₂)₅-, -(CH₂)6-, -(CH₂)7-, -(CH₂)₈-, -(CH₂)₉-, -(CH₂)IO-, -(CH₂)H-, -(CH₂)12-, -(CH₂)I₃-, -(CH₂)14-, -(CH₂)15-, -(CH₂)l6-, -(CH₂)17-, -(CH₂)18-, - (CH2) -, -(CH2)2O- and positional isomers thereof. C4-C6-Alkylene (or C4-Ce-alkyanediyl) is a linear or branched divalent alkyl radical having 4 to 6 carbon atoms. Examples are -CH2CH2-, -CH(CH₃)-, -CH2CH2CH2-, -CH(CH₃)CH₂-, -CH₂CH(CH₃)-, -C(CH₃)₂-, -CH2CH2CH2CH2-, -CH(CH₃)CH₂CH₂-, -CH₂CH₂CH(CH₃)-, -C(CH₃) ₂CH₂-, -CH₂C(CH₃)₂- , -(CH2)S-, -(CH2)e- and positional isomers thereof.

Embodiments (E.x) of the invention

General and preferred embodiments E.x are summarized in the following, non-exhaus- tive list. Further preferred embodiments become apparent from the paragraphs following this list.

E.1 . A method for producing an onium fluoride of the formula (I) R₄Q⁺ F- ■ nxHF (I) where

Q is N or P; each R is independently selected from the group consisting of Ci-C2o-alkyl, C₃- Cs-cycloalkyl, C₃-C8-cycloalkyl-Ci-C4-alkyl, Ce-C -aryl, Ce-Cio-aryl-Ci-C4- alkyl, and a 3- to 8-membered saturated, partially unsaturated or maximally unsaturated heterocyclic ring containing 1 , 2 or 3 heteroatoms or heteroatom groups selected from the group consisting of O, N, S, S(O) and S(O)2 as ring members; or two R form together a C2-C2o-alkylene bridging group which may be interrupted by 1 , 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1 , 2 or 3 aromatic or heteroaromatic rings; and the other two R are either as defined above or, independently, also form together a C2-C2o-alkylene bridging group which may be interrupted by 1 , 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1 , 2 or 3 5- or 6-membered aromatic or heteroaromatic rings; and n is from 0.1 to 5; which method comprises

(i) providing a solution of an onium chloride of the formula (II)

R₄Q⁺ C (II) where Q and R are as defined above, in a solvent;

(iv) reacting the reaction mixture obtained in step (iii) with a fluoride salt of the formula (M^m+)(F')_m, where M is an alkali metal or alkaline earth metal and m is the charge thereof, m being thus 1 if M is an alkali metal and 2 if M is an alkaline earth metal;

(v) removing any precipitate present in the reaction mixture obtained in step (iv);

(vi) partially or completely removing the solvent; and

E.2. The method according to embodiment E.1 , where Q is P.

E.3. The method according to any of embodiments E.1 or E.2, where each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl; or two R form together a C4-C6-alkylene bridging group and the other two R are either independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cy- cloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl, or, independently, also form together a C4-Ce-alkylene bridging group. E.4. The method according to embodiment E.1 , where each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and Cs-Ce-cyclo- alkyl-Ci-C2-alkyl.

E.5. The method according to embodiment E.4, where each R is independently Ci- C2o-alkyl.

E.6. The method according to embodiment E.5, where each R is independently C2- Cw-alkyl.

E.7. The method according to embodiment E.6, where each R is independently C2-C6- alkyl.

E.8. The method according to embodiment E.7, where each R is independently C3-C5- alkyl.

E.9. The method according to embodiment E.8, where each R is n-butyl.

E.10. The method according to any of the preceding embodiments, where n is 0.5 to 2. E.11. The method according to embodiment E.10, where n is approximately 1.

E.12. The method according to any of the preceding embodiments, where the solvents of steps (i) and (iii) (if applicable) are independently selected from the group consisting of Ci-Cs-alkanols, C2-Cs-alkanediols and mixtures thereof.

E.13. The method according to embodiment E.12, where the solvents are independently selected from Ci-Cs-alkanols.

E.14. The method according to embodiment E.13, where the solvents are independently selected from Ci-Cs-alkanols and 2-ethylhexanol.

E.15. The method according to embodiment E.14, where the solvents are independently selected from methanol, isopropanol, 2-ethylhexanol and mixtures thereof.

E.16. The method according to embodiment E.14, where the solvents are independently selected from Ci-Cs-alkanols

E.17. The method according to embodiment E.16, where the solvents are independently selected from methanol, isopropanol and mixtures thereof.

E.18. The method according to embodiment E.17, where the solvents are independently methanol, optionally in admixture with isopropanol.

E.19. The method according to embodiment E.18, where the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, and is specifically isopropanol; and the solvent of step (iii) is methanol.

E.20. The method according to any of the preceding embodiments, where the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal hydroxides, alkali metal or alkaline earth metal carbonates, alkali metal or alkaline earth metal hydrogen carbonates and alkali metal or alkaline earth metal Ci-Cs-alkanolates.

E.21. The method according to embodiment E.20, where the base of step (iii) is selected from the group consisting of alkali metal Ci-Cs-alkanolates. E.22. The method according to embodiment E.21 , where the base of step (iii) is selected from alkali metal Ci-C4-alkanolates.

E.23. The method according to embodiment E.22, where the base of step (iii) is selected from alkali metal Ci-Cs-alkanolates.

E.24. The method according to embodiment E.22, where the base of step (iii) is selected from potassium Ci-C4-alkanolates.

E.25. The method according to embodiment E.24, where the base of step (iii) is selected from potassium Ci-Cs-alkanolates.

E.26. The method according to embodiment E.25, where the base of step (iii) is potassium methanolate.

E.27. The method according to any of the preceding embodiments, where the fluoride salt of step (iv) is of the formula MF, where M is an alkali metal.

E.28. The method according to embodiment E.27, where M is sodium or potassium E.29. The method according to embodiment E.28, where M is potassium (and MF is KF).

E.30. The method according to any of the preceding embodiments, where in step (iii) the solution of step (i) is diluted with a solvent and then hydrogen chloride is neutralized with a base.

E.31 . The method according to any of the preceding embodiments, where the solvents of steps (i) and (iii) (if applicable) are independently a Ci-Cs-alkanol or a mixture of different Ci-Cs-alkanols and the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal Ci-Cs-alkanolates, where the Ci- Cs-alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), and the alkali metal or alkaline earth metal of the base corresponds to M of the fluoride salt used in step (iv).

E.32. The method according to embodiment E.31 , where the solvents of steps (i) and (iii) (if applicable) are independently a Ci-Cs-alkanol or a mixture of different Ci- Cs-alkanols and the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal Ci-Cs-alkanolates, where the Ci-Cs-alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), and the alkali metal or alkaline earth metal of the base corresponds to M of the fluoride salt used in step (iv).

E.33. The method according to embodiment E.32, where the solvent of steps (i) and (iii) (if applicable) is independently methanol, isopropanol or a mixture of methanol and isopropanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

E.34. The method according to embodiment E.33, where the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, specifically isopropanol; the solvent of step (iii) is methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF. E.35. The method according to any of the preceding embodiments, where Q is P; R is C2-Cio-alkyl, preferably C2-Ce-alkyl, more preferably Cs-Cs-alkyl, specifically n-bu- tyl; n is 0.5 to 2; the solvent of steps (i) and (iii) (if applicable) is independently methanol, isopropanol or a mixture of methanol and isopropanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

E.36. The method according to embodiment E.35, where Q is P; R is C2-Cio-alkyl, preferably C2-Ce-alkyl, more preferably Cs-Cs-alkyl, specifically n-butyl; n is 0.5 to 2; the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, specifically isopropanol; the solvent of step (iii) is methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

E.37. The method according to embodiment E.36, where Q is P; R is n-butyl; n is 0.5 to 2, preferably approximately 1 ; the solvent of step (i) is isopropanol; the solvent of step (iii) is methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

E.38. The method according to any of the preceding embodiments, where the reaction mixture to be reacted in step (iv) (i.e. the reaction mixture before the fluoride salt is added) contains the onium chloride of the formula (II) in an amount of from 0.5 to 3 mol per kg of solvent.

E.39. The method according to embodiment E.38, where the reaction mixture to be reacted in step (iv) contains the onium chloride of the formula (II) in an amount of from 1 to 2.5 mol per kg of solvent.

E.40. The method according to embodiment E.39, where the reaction mixture to be reacted in step (iv) contains the onium chloride of the formula (II) in an amount of from 1.5 to 2.2 mol per kg of solvent, e.g. of from 1.5 to 2.0 mol per kg of solvent.

E.41. The method according to any of the preceding embodiments, where the fluoride salt of the formula (M^m+)(F')_m is used in an amount of from 1 to 2 mol, per mol of the onium chloride of the formula (II) provided in step (i), where the amount of the fluoride salt relates to the amount of F contained therein.

E.42. The method according to embodiment E.41 , where the fluoride salt of the formula (M^m+)(F )m is used in an amount of from 1 .05 to 1 .5 mol per mol of the onium chloride of the formula (II) provided in step (i), where the amount of the fluoride salt relates to the amount of F contained therein.

E.43. The method according to embodiment E.42 where the fluoride salt of the formula (M^m+)(F )m is used in an amount of from 1.1 to 1.2 mol per mol of the onium chloride of the formula (II) provided in step (i), where the amount of the fluoride salt relates to the amount of F contained therein.

E.44. The method according to any of the preceding embodiments, where step (iv) is carried out at a temperature of from 0 to 50°C.

E.45. The method according to embodiment E.44, where step (iv) is carried out at a temperature of from 10 to 40°C E.46. The method according to embodiment E.45, where step (iv) is carried out at a temperature of from 20 to 35°C.

E.47. The method according to any of the preceding embodiments, where in step (vi) partially or completely removing the solvent is carried out under reduced pressure and at a temperature of from 25 to 50°C, preferably from 25 to 45°C.

E.48. The method according to any of the preceding embodiments, where in step (vi) the solvent is removed only partially.

E.49. The method according to any of the preceding embodiments, where anhydrous hydrogen fluoride is used in amount of from 0.1 to 5 mol per mol of the onium chloride of the formula (II) provided in step (i).

E.50. The method according to embodiment E.49, where anhydrous hydrogen fluoride is used in amount of from 0.5 to 2 mol per mol of the onium chloride of the formula (II) provided in step (i).

E.51. The method according to embodiment E.50, where anhydrous hydrogen fluoride is used in amount of approximately 1 mol per mol of the onium chloride of the formula (II) provided in step (i).

E.52. The method according to any of the preceding embodiments, where step (vii) is carried out at a temperature of from 0 to 40°C.

E.53. The method according to embodiment E.52, where step (vii) is carried out at a temperature of from 5 to 30°C.

E.54. The method according to embodiment E.53, where step (vii) is carried out at a temperature of from 10 to <30°C.

E.55. The use of an onium fluoride of the formula (I) as defined in any of embodiments E.1 to E.11 or as obtained in the process of any of embodiments E.1 to E.54 as a catalyst for the preparation of isocyanate trimers containing iminooxadiazinedione groups.

E.56. A process for preparing isocyanate trimers containing iminooxadiazinedione groups, comprising reacting at least one (cyclo)aliphatic diisocyanate in the presence of an onium fluoride catalyst of the formula (I) as defined in any of embodiments E.1 to E.11 or as obtained in the process of any of embodiments E.1 to E.54, and when the reaction has reached a predetermined degree of conversion of the (cyclo)aliphatic diisocyanates, stopping the reaction by addition of at least one catalyst poison for the catalyst, and preferably separating off unreacted (cycloaliphatic diisocyanate.

E.57. The process according to embodiment E.56, where the (cyclo)aliphatic diisocyanates are selected from the group consisting of hexamethylene diisocyanate (HDI), pentamethylene diisocyanate (PDI), 5-isocyanato-1-(isocyanatomethyl)- 1 ,3,3-trimethylcyclohexane (I PDI), 2-methyl pentane -1 ,5-diisocyanate, 2,4,4-tri- methyl-1 ,6-hexane diisocyanate, 2,2,4-trimethyl-1 ,6-hexane diisocyanate and 4- isocyanatomethyl-1 ,8-octane. E.58. The process according to embodiment E.57, where the (cyclo)aliphatic diisocyanates are selected from the group consisting of hexamethylene diisocyanate (HDI) and pentamethylene diisocyanate (PDI).

E.59. The process according to embodiment E.58, where the (cyclo)aliphatic diisocyanate is hexamethylene diisocyanate.

E.60. The process according to any of embodiments E.56 to E.59, where the catalyst is used in amount of from 20 ppm to 500 ppm, based on the weight of the (cycloaliphatic diisocyanate.

E.61. The process according to embodiment E.60, where the catalyst is used in amount of from 50 ppm to 300 ppm, based on the weight of the (cyclo)aliphatic diisocyanate.

E.62. The process according to embodiment E.61, where the catalyst is used in amount of from 50 ppm to 150 ppm, based on the weight of the (cyclo)aliphatic diisocyanate.

E.63. The process according to any of embodiments E.56 to E.62, comprising dissolving the catalyst in a solvent and adding the obtained solution to the (cycloaliphatic diisocyanate.

E.64. The process according to any of embodiments E.56 to E.5638, where the catalyst poison is selected from organic and inorganic acids and acid derivatives, such as sulfonic acids, e.g., p-toluene sulfonic acid or dodecyl benzene sulfonic acid; benzoic acid, benzoyl chloride, phosphoric acid, acidic esters thereof, phosphorous acid and acidic esters thereof.

E.65. A process for producing polyurethane coatings, comprising reacting the trimeric isocyanate composition obtained according to the process of embodiments E.56 to E.64 with at least one binder selected from the groups consisting of polyacrylate polyols, polyester polyols, polyether polyols, polyurethane polyols, polyurea polyols, polyetherols, polycarbonates, polyester polyacrylate polyols, polyester polyurethane polyols, polyurethane polyacrylate polyols, polyurethane-modified alkyd resins, fatty acid-modified polyester polyurethane polyols, copolymers with allyl ethers and copolymers or graft polymers thereof.

E.66. The use of the trimeric isocyanate composition obtained according to the process of embodiments E.56 to E.64 as a curing agent, preferably in a material selected from the group consisting of coating materials in primers, primer surfacers, pigmented topcoats, basecoats and clearcoats in the sectors of refinishing, automotive refinishing, large vehicle finishing and wood, plastic, and OEM finishing, in utility vehicles in the agricultural and construction sector and as curing agent in adhesives and sealants.

The reaction (anion exchange) taking place in step (iv) can be depicted as follows: m R₄Q⁺ O + (M^m+)(F-)_m m R₄Q⁺ F + (M^m+)(C|-)_m Q in formulae (I) and (II) is N or P. If Q is N, the onium salt is an ammonium salt, and if Q is P, the onium salt is a phosphonium salt. Preferably, Q is P; i.e. , the onium salts are preferably phosphonium salts.

In compounds (I) and (II), two R may form together a bridging group as defined above and the other two R are independently Ci-C2o-alkyl, Cs-Ce-cycloalkyl or Cs-Ce-cycloal- kyl-Ci-C2-alkyl or also form together a bridging group as defined above. In the latter case, the compounds (I) and (II) are spiro-form with Q being the spiro center.

If just two R form together a bridging group (and the other two R are Ci-C2o-alkyl, C3- Ce-cycloalkyl or C3-C6-cycloalkyl-Ci-C2-alkyl), the compounds (I) and (II) are a heterocyclic ring containing a nitrogen (if Q is N) or phosphorous (if Q is P) atom as ring member, in which said N or P atom is quaternized by carrying two R groups which are independently Ci-C2o-alkyl, Cs-Ce-cycloalkyl or C3-C6-cycloalkyl-Ci-C2-alkyl.

Examples for rings where two R form together an alkylene bridge not interrupted by any heteroatoms or rings, are: If Q is N: the quaternized forms (i.e. the ring nitrogen atom carries two R groups which are independently Ci-C2o-alkyl, Cs-Ce-cycloalkyl or C3-C6-cycloalkyl-Ci-C2-alkyl) of aziridin, azetidin, pyrrolidin, piperidin and the higher homologues; and if Q is P: the quaternized forms (i.e. the ring nitrogen atom carries two R groups which are independently Ci-C2o-alkyl, Cs-Ce-cycloalkyl or Cs-Ce-cycloal- kyl-Ci-C2-alkyl) of phosphirane, phosphethane, phospholane, phosphinane and the higher homologues.

Preferably, each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl; or two R form together a C4-Ce-al- kylene bridging group and the other two R are either independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl, or independently also form together a C4-C6-alkylene bridging group. More preferably, each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cy- cloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl and is even more preferably Ci-C2o-alkyl. In particular, each R is independently C2-Cio-alkyl, more particularly C2-Ce-alkyl, even more particularly Cs-Cs-alkyl and is specifically n-butyl.

The values given for n are in general statistical averages which, for a specific product, may be an integer or a fraction on average.

In compounds (I), n is preferably 0.5 to 2, and is more preferably approximately 1. “Approximately” in this context means to include minor deviations from this figure due, for example, to metering errors or impurities present in hydrogen fluoride. The deviation is generally at most 10%, preferably at most 5%. Compounds (II) are generally commercially available or can be prepared by standard methods of organic or inorganic chemistry. Compounds (II) where two R may form together a bridging group, if not commercially available, can for example be prepared as or in analogy to the method described in WO 2015/124504.

The solvents used in steps (i) and (iii), if applicable, are expediently chosen thusly that they show good solvation properties for the chloride (II) and the fluoride salt (M^m+)(F')m as well as the onium fluoride R4Q⁺ F’ formed therefrom, and distinctly poorer solvation properties for the chloride salt (M^m+)(CI')_m. Choosing the solvents thusly allows to remove (M^m+)(CI )_m (formed as by-product in the anion exchange step (iv); see above) rather easily, since this precipitates readily, if necessary after concentrating the reaction mixture. Preferably, the solvents of steps (i) and (iii), if applicable, are selected from the group consisting of Ci-Cs-alkanols, C2-Cs-alkanediols and mixtures thereof. Preferably, said solvents are either essentially anhydrous or contain water in a defined quantity, so that the above desired solvation properties are met. “Essentially anhydrous” means a water content of at most 2% by weight, preferably at most 1 % by weight, more preferably at most 0.5% by weight, even more preferably at most 0.2% by weight, e.g. at most 0.1% by weight, relative to the total weight of the solvent. If the solvents contain water in a defined quantity range, this is preferably rather low in order to avoid a higher solubility for the chloride salt (M^m+)(CI')m and/or a negative impact during the further use of the desired onium fluoride, and is for example at most 10% by weight of water, preferably at most 5% by weight of water, more preferably at most 3% by weight of water, relative to the total weight of the solvent. More preferably, the solvents of steps (i) and (iii), if applicable, are essentially anhydrous.

Ci-Cs-alkanols are compounds R-OH, where R is linear or branched Ci-Cs-alkyl. Examples are methanol, ethanol, n-propanol, isopropanol, n-butanol, sec-butanol, isobutanol, tert-butanol, 1-pentanol, 1-hexanol, 1-heptanol, 1-octanol, 2-ethylhexanol and (other) structural isomers of the four last-mentioned 1 -alkanols.

C2-Cs-alkanediols are compounds HO-A-OH, where A is linear or branched C2-Cs-al- kanediyl (or C2-Cs-alkylene), where the two OH groups are not geminally bound (i.e. , are not bound to the same carbon atom). Examples are ethylene glycol (1 ,2-ethane- diol), propylene glycol (1 ,2-propanediol), 1 ,3-propanediol, 1 ,2-butanediol, 1 ,4-butane- diol, 1 ,2-pentanediol, 1 ,5-pentanediol, 1 ,2-hexanediol, 1 ,6-hexanediol, 1 ,2-heptanediol, 1 ,2-octanediol, 2-ethylhexane-1 ,3 diol, 2,2,4-trimethyl-1 ,3-pentanol and the like.

More preferably, the solvent is selected from Ci-Cs-alkanols, such as methanol, ethanol, n-propanol, isopropanol, n-butanol, sec-butanol, isobutanol, tert-butanol, 1-penta- nol, 1-hexanol, 1-heptanol, 1-octanol, 2-ethylhexanol and mixtures thereof. Among the Ci-Cs-alkanols, preference is given to Ci-Cs-alkanols, 2-ethylhexanol and mixtures thereof, i.e. to methanol, ethanol, n-propanol, isopropanol, 2-ethylhexanol and mixtures thereof, and more preference to methanol, isopropanol, 2-ethylhexanol and mixtures thereof.

In particular, the solvent is selected from Ci-C4-alkanols, more particularly from C1-C3- alkanols, such a methanol, ethanol, n-propanol, isopropanol and mixtures thereof. Among the Ci-Cs-alkanols, preference is given to methanol, isopropanol and mixtures thereof and specifically to methanol and mixtures of methanol and isopropanol.

If in step (iii) a solvent is added, the solvent of step (i) and of step (iii) can be the same or different.

In a specific embodiment, the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, and is specifically isopropanol, and the solvent of step (iii) is methanol.

The solution provided in step (i) can either be a concentrate, a concentrate being a solution which contains the onium chloride in a higher concentration than intended for the further steps and especially for the anion exchange step (iv), and for example containing the onium chloride (II) in a concentration of from 40 to 80% by weight or from 50 to 80% by weight or from 60 to 80% by weight or from 65 to 75% by weight, relative to the total weight of the solution (compounds (II) are often commercially available in the form of such concentrates); or can be a diluted solution containing the onium chloride expediently in the concentration desired for the ensuing reaction steps, and especially for the anion exchange step (iv). If the solution provided in step (i) is to be a diluted solution containing the onium chloride (II) in the concentration desired for the ensuing reaction steps, and especially for the anion exchange step (iv), this solution contains the onium chloride of the formula (II) in an amount of preferably from 0.5 to 3 mol, more preferably from 1 to 2.5 mol and in particular from 1.5 to 2.2 mol per kg of solvent.

Since the amount of HCI in the onium chloride is generally not high (commercial onium chloride concentrates, for example, containing the onium chloride (II) in an amount of ca. 60 to 80% by weight, generally contain ca. 1 to 2% by weight, mostly ca. 1% by weight of HCI, relative to the total weight of the solution), it is more expedient to provide in step (i) a concentrated solution to allow a more precise determination of the HCI content in step (ii).

Thus, the solution provided in step (i) is preferably a concentrate containing the onium chloride of the formula (II) in an amount of preferably from 40 to 80% by weight, more preferably from 50 to 80% by weight, even more preferably from 60 to 80% by weight and in particular from 65 to 75% by weight, relative to the total weight of said solution. It is however also possible to provide a diluted solution in step (i). In this case, to ensure high precision of the quantitative HCI analysis in step (ii), it might be preferable to determine the amount of the hydrogen chloride contained therein by taking a sample of the diluted solution, concentrating the same to a definite concentration by partially removing a defined amount of the one or more solvents and subjecting this concentrated sample to a quantitative analysis of the hydrogen chloride content.

Out of practical reasons, the first variant (providing in step (i) a concentrated solution) is however preferred. In this case, preferably, step (iii) comprises diluting the solution of step (i) with a solvent and neutralising the hydrogen chloride contained in the diluted solution by reaction with a base. In a specific embodiment of this case, the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, and the solvent of step (iii) is methanol. In a more specific embodiment of this case, the solvent of step (i) is isopropanol, and the solvent of step (iii) is methanol.

The solution of step (i) can be provided by usual means. If, for example, the onium chloride (II) is available in substance, this can be dissolved in the desired solvent(s) to the desired concentration. Alternatively, a commercially available concentrate of the onium salt (II) can be used if this is in the desired solvent(s); this concentrate can then be further diluted with one or more solvents if a (more) diluted solution is desired. If the commercially available concentrate is not in the desired solvent(s), the solvents contained therein can be removed, e.g. distillatively, expediently under reduced pressure, and the desired solvent(s) can be added up to the desired concentration.

Quantitative determination in step (ii) of the hydrogen chloride content in the solution of the onium chloride (II) provided in step (i) is carried out by usual means, such as acidbase titration, a typical proceeding being for example potentiometric titration with 1 N aqueous NaOH solution. For this purpose, generally a sample of the solution provided in step (i) is taken, analysed and the result is extrapolated to the complete solution of step (i) to allow calculation of the amount necessary for neutralization.

In step (iii), the hydrogen chloride determined to be present in the solution provided in step (i) is neutralised by reaction with a base. Generally, a base is added to the solution of (II), expediently under thorough mixing. If the solution provided in step (i) is a concentrate, it is expedient to dilute the solution with a solvent before neutralising. Suitable and preferred solvents have already been listed above. The solvent is added in such amounts that after dilution, the solution contains the onium chloride of the formula (II) in an amount of preferably from 0.5 to 3 mol, more preferably from 1 to 2.5 mol and in particular from 1.5 to 2.2 mol per kg of solvent. The base can be added in pure form or as solution, e.g., in (one of) the solvents of step (i) or used in step (iii) (if applicable). Preferably, the base is added in solution, the solvent being preferably the or one of the solvents used in step (i) or (iii).

Since the amount of HCI is generally not high (as said above, commercial onium chloride concentrates, containing the onium chloride in an amount of ca. 60 to 80% by weight, generally contain ca. 1 to 2% by weight, mostly approximately 1% by weight of HCI, relative to the total weight of the solution; diluted solutions containing HCI in even lower relative amounts), neutralization does not lead to a notable heat formation, and it is thus generally not required to take any measures for cooling the reaction.

The type of base is not critical; it is however desirable to use a base which does not introduce matter which needs to be separated off the desired product (as would be the case, for example, for amine bases, basic heterocycles and the like). Suitable bases are for example alkali metal or alkaline earth metal hydroxides, alkali metal or alkaline earth metal carbonates, alkali metal or alkaline earth metal hydrogen carbonates and alkali metal or alkaline earth metal Ci-Cs-alkanolates.

Ci-Cs-Alkanolates (also termed Ci-Cs-alkoxides) are the salts of Ci-Cs-alkanols (and could be depicted as R-O' M⁺, where M⁺ is a cation equivalent and R is Ci-Cs-alkyl), examples being methanolate (also termed methoxide), ethanolate (ethoxide), n-propano- late (n-propoxide), isopropanolate (isopropoxide), n-butanolate (n-butoxide), sec-buta- nolate (sec-butoxide), isobutanolate (isobutoxide), tert-butanolate (tert-butoxide), 1- pentanolate, 1-hexanolate, 1 -heptanolate, 1-octanolate, 2-ethylhexanolate and other structural isomers thereof.

Suitable alkali metal cations in the above-listed bases are lithium, sodium, potassium and caesium cations. Suitable alkaline earth metal cations in the above-listed bases are magnesium and calcium cations.

Considering the above-listed suitable and preferred solvents, seeing their better solubility in the latter, preference is given to the alkali metal hydroxides, carbonates, hydrogen carbonates and alkanolates. More preference is given to the potassium hydroxides, carbonates, hydrogen carbonates and alkanolates.

Expediently, a strong base is used for neutralization, so that the required amount can be easily calculated without having to take account of incomplete dissociation, buffer effects and the like. Strong bases are such with a pKb of at most 2, preferably of at most 1. Thus, among the above-listed bases, preference is given to alkali metal or al- kaline earth metal hydroxides and alkali metal or alkaline earth metal Ci-Cs-alka- nolates, more preference being given to alkali metal hydroxides and alkali metal Ci-Cs- alkanolates and especially to the potassium hydroxides and Ci-Cs-alkanolates.

To avoid the formation of water (as would occur if hydroxides, carbonates or hydrogen carbonates are used as a base) and thus a change in the solubility product of the various salts, it is more expedient to use a base which does not form water in the neutralization reaction. Thus, among the above bases, more preference is given to alkali metal or alkaline earth metal Ci-Cs-alkanolates, even more preference to alkali metal Ci-Cs- alkanolates and particular preference to potassium Ci-Cs-alkanolates. In particular, alkali metal Ci-C4-alkanolates, especially potassium Ci-C4-alkanolates, and more particularly alkali metal Ci-Cs-alkanolates, especially potassium Ci-Cs-alkanolates, are used.

Out of practical reasons, if the solvent of step (i) or (iii) is a Ci-Cs-alkanol or a mixture of different Ci-Cs-alkanols, the base of step (iii) is expediently selected from the group consisting of alkali metal or alkaline earth metal Ci-Cs-alkanolates, where the Ci-Cs- alkanolate is derived from that Ci-Cs-alkanol or one of those Ci-Cs-alkanols which are used as solvent in step (i) or (iii). For instance, if the solvent of step (i) or (iii) is or comprises methanol, the base is expediently an alkali metal or alkaline earth metal methanolate, if the solvent of step (i) or (iii) is or comprises ethanol, the base is expediently an alkali metal or alkaline earth metal ethanolate, etc.

Specifically, the solvent of step (i) or (iii) is or comprises methanol and the base is potassium methanolate.

The base is used in such an amount that it neutralizes the amount of HCI determined in step (ii). Preferably, the base is used in an amount of 0.9 to 1.2 mol, more preferably 0.9 to 1.1 mol, even more preferably 0.95 to 1.05 mol and specifically approximately 1 mol per mol of HCI present in the solution of step (i). “Approximately” in this context means to include minor deviations due, for example, to weighing or metering errors or impurities present in the base. The deviation is generally at most 2%, preferably at most 1 %.

If in step (i) a concentrated solution has been provided and in step (iii) no further solvent has been added, it is expedient to dilute the neutralised solution with a solvent before adding in step (iv) the fluoride salt (M^m+)(F')_m. Suitable and preferred solvents are those listed above. The solvent is added in such amounts that after dilution, the solution contains the onium chloride of the formula (II) in an amount of preferably from 0.5 to 3 mol, more preferably from 1 to 2.5 mol and in particular from 1.5 to 2.2 mol per kg of solvent. Preferably however, either in step (i) a diluted solution is provided and in step (iii) no additional solvent is added, or in step (i) a concentrated solution is provided and in step (iii) the concentrated solution is diluted with a solvent and then neutralization is carried out, so that no further dilution after neutralization step (iii) is necessary.

In the fluoride salt of the formula (M^m+)(F')_m added in step (iv), M is an alkali metal or alkaline earth metal. If M is an alkali metal, the salt has the formula MF, and if M is an alkaline earth metal, the salt has the formula MF2. Suitable alkali metal cations in the fluoride salt bases are lithium, sodium, potassium and caesium cations. Suitable alkaline earth metal cations in the fluoride salt are magnesium and calcium cations.

Considering the above-listed suitable and preferred solvents and the solubility of the fluorides, preference is given to the alkali metal fluorides, and more preference to sodium or potassium fluorides. Particular preference is given to potassium fluoride (KF).

In a particular embodiment, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-Cs-alkanols and mixtures of different Ci-Cs-alkanols, the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal Ci-Cs-alkanolates, where the Ci-Cs-alkanolate is derived from the Ci-Cs- alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), if applicable, and the alkali metal or alkaline earth metal of the base corresponds to M of the fluoride salt used in step (iv). More particularly, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-Cs-alkanols and mixtures of different Ci-Cs- alkanols, the base of step (iii) is an alkali metal Ci-Cs-alkanolate, where the Ci-Cs-alka- nolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), and the alkali metal of the base corresponds to M of the fluoride salt used in step (iv). Even more particularly, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-Cs-alkanols and mixtures of different Ci-Cs-alkanols, the base of step (iii) is a potassium Ci-Cs-alkanolate, where the Ci-Cs- alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), if applicable, and the fluoride salt used in step (iv) is KF. Specifically, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-C4-alkanols and mixtures of different Ci-C4-alkanols, the base of step (iii) is an alkali metal Ci-C4-alkanolate, where the Ci-C4-alkanolate is derived from the Ci-C4-alkanol or one of the Ci-C4-alkanols used as solvent in step (i) and/or (iii), if applicable, and the fluoride salt used in step (iv) is KF. More specifically, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-C4-alka- nols and mixtures of different Ci-C4-alkanols, the base of step (iii) is a potassium C1- C4-alkanolate, where the Ci-C4-alkanolate is derived from the Ci-C4-alkanol or one of the Ci-C4-alkanols used as solvent in step (i) and/or (iii), if applicable, and the fluoride salt used in step (iv) is KF. Specifically, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-Cs-alkanols and mixtures of different Ci-Cs-alkanols, the base of step (iii) is an alkali metal Ci-Cs-alkanolate, where the Ci- Cs-alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), if applicable, and the fluoride salt used in step (iv) is KF. More specifically, the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of Ci-Cs-alkanols and mixtures of different Ci-Cs-alkanols, the base of step (iii) is a potassium Ci-Cs-alkanolate, where the Ci-Cs-alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols used as solvent in step (i) and/or (iii), and the fluoride salt used in step (iv) is KF. Even more specifically the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of methanol, isopropanol and mixtures of methanol and isopropanol, where at least one of the solvents of step (i) and step (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF. Even more specifically, the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, the solvent of step (iii), if applicable, is methanol, where at least one of the solvents of step (i) and step (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF. Particularly specifically, the solvent of step (i) is isopropanol, the solvent of step (iii) is methanol, the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

In a specific embodiment, Q is P; R is C2-Cio-alkyl, preferably C2-Ce-alkyl, more preferably Cs-Cs-alkyl and specifically n-butyl; n is 0.5 to 2; the solvent of step (i) and step (iii), if applicable, is selected from the group consisting of methanol, isopropanol and mixtures of methanol and isopropanol, where at least one of the solvents of step (i) and step (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF. More specifically, Q is P; R is C2-Cio-alkyl, preferably C2-Ce-alkyl, more preferably Cs-Cs-alkyl and specifically n-butyl; n is 0.5 to 2; the solvent of step (i) is methanol, isopropanol or a mixture of methanol and isopropanol, the solvent of step (iii), if applicable, is methanol, where at least one of the solvents of step (i) and step (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF. Very specifically, Q is P; R is n-butyl; n is 0.5 to 2, preferably approximately 1 ; the solvent of step (i) is isopropanol, the solvent of step (iii) is methanol, the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

The fluoride salt of the formula (M^m+)(F')_m is used in step (iv) in an amount of preferably from 1 to 2 mol, more preferably from 1.05 to 1.5 mol and even more preferably from 1.1 to 1.2 mol per mol of the onium chloride of the formula (II) provided in step (i), where the amount of the fluoride salt relates to the amount of F contained therein (meaning thus that in case of alkaline earth metal fluorides MF2 containing two mol of F per mol of salt, these are actually used in an amount of preferably from 0.5 to 1 mol, more preferably from 0.525 to 0.75 mol and even more preferably from 0.55 to 0.6 mol per mol of the onium chloride of the formula (II) provided in step (i)).

Step (iv) is generally carried out by adding the fluoride salt to the neutralized solution obtained in step (iii). The fluoride salt can be added in pure (solid) form or in solution, e.g., in (one of) the solvent(s) of step (i) or (iii), if applicable. Generally, it is added in solid form. The fluoride salt can be added all at once, portion-wise or continually, generally under thorough mixing. The reaction mixture is brought to the desired temperature and reacted.

The reaction temperature is not critical. High temperature is not required, and thus the reaction temperature preferably ranges from 0 to 50°C, more preferably from 10 to 40°C, even more preferably from 20 to 35°C and in particular from 25 to 35°C.

Reaction time in step (iv) can range from 1 to 60 hours, e.g., from 2 to 50 hours or 4 to 30 hours, and is preferably in the range of 6 to 12 or 6 to 10 hours.

If desired, the progress of the reaction can be monitored, e.g., by determining the amount of chloride in the liquid phase. Determination of the chloride content can for example be carried out titrimetrically, e.g., by titration with a silver nitrate solution, e.g., according to Mohr’s method (ISO 9297:1989-11). Generally, a chloride content of at most 0.4% by weight, preferably at most 0.2% by weight, relative to the total weight of the liquid phase is aspired.

Since alkali or alkaline earth metal chloride salts are generally less soluble than the corresponding fluorides, the chloride salts formed by anion exchange in step (iv) generally precipitate as reaction progresses. Especially if potassium fluoride is used as fluoride salt and thus potassium chloride is formed in the anion exchange, the latter precipitates virtually completely in the course of step (iv) without any further measures being taken, especially if the above-listed preferred solvents, and specifically the above-listed alkanols, are used, if the temperature does not exceed the above-given upper ranges and if the above-given preferred concentration of compounds (II) is respected. If precipitation of the chloride salt is not complete, this can be furthered, for example, by up- concentration of the reaction mixture obtained in step (iv).

In step (v), any precipitates formed in step (iv) (as explained above, these are generally the alkali or alkaline earth metal chlorides formed in the anion exchange, but also excess alkali or alkaline earth metal fluorides may be present) are removed; e.g., by filtration, sedimentation or any other method known in the art for separating solid from liquid materials. Expediently, the separated precipitates are washed or digested with a suitable solvent, generally with the solvent(s) used in step (i) and/or (iii), where applicable, to extract any onium fluoride (I) adhering to the precipitates.

In step (vi), the solution from which the precipitates have been removed in step (v), generally combined with any extracts obtained from washing/digesting the precipitates, is then depleted partially or completely of the solvent(s). This is preferably carried out under reduced pressure and at a temperature of from 25 to 50°C, preferably from 25 to 45°C, more preferably from 25 to 30°C.

Preferably, the solvent is removed only partially, the obtained mixture preferably containing 10 to 50% by weight, more preferably 15 to 40% by weight, specifically 20 to 35% by weight of residual solvent(s), based on the total weight of the mixture.

If any further precipitates are formed during the partial removal of the solvent, these are expediently removed (e.g., by filtration or sedimentation; if desired followed of washing or digesting the precipitates with a suitable solvent, generally with the solvent(s) used in step (i) and/or (iii), where applicable, to extract any onium fluoride (I) adhering thereto) before step (vii) is carried out.

If in step (vi) the solvent(s) has/have been completely removed, the residue is preferably dissolved in a suitable solvent before step (vii) is carried out. Suitable and preferred solvents are those listed above in context with step (i) and/or (iii).

The anhydrous hydrogen fluoride is used in step (vii) in an amount of from 0.1 to 5 mol, preferably from 0.5 to 2 mol, more preferably in an amount of approximately 1 mol per mol of the onium chloride of the formula (II) provided in step (i). “Approximately” in this context means to include minor deviations from this figure due, for example to metering errors or impurities present in the hydrogen fluoride. The deviation is generally at most 10%, preferably at most 5%.

“Anhydrous hydrogen fluoride” relates to anhydrous hydrogen fluoride as obtained in technical processes for the preparation thereof. Generally, it contains <100 mg of water per kg of hydrogen fluoride.

Step (vii) is carried out at a temperature of preferably from 0 to 40°C, more preferably from 5 to 30°C and specifically from 10 to <30°C.

Step (vii) is generally carried out by adding the amount of anhydrous HF required to obtain the desired “n” in n*HF in formula (I) (i.e. by adding n mol of HF per mol of com- pound (II) used in step (i)) to the reaction mixture obtained in step (vi), expediently under thorough mixing. Anhydrous HF is preferably added in liquid form. To this purpose, given its boiling point of 19.5°C at 1013 mbar and to avoid the need of applying pressure, the reaction mixture obtained in step (vi) is expediently cooled to below this temperature, preferably to 0 to 15°C or more preferably to 5 to 12°C, before anhydrous HF is added. Since the reaction with the onium fluoride is exothermic, it is expedient to cool the reaction mixture during addition of HF and the progress of the reaction of step (vi), preferably to 0 to 40°C, more preferably to 5 to 30°C and specifically to 10 to <30°C.

The obtained reaction mixture contains the onium fluoride of the formula (I) in a solvent (mixture). The reaction mixture is generally used as such, but if desired can of course also be partially or completely depleted of the solvent(s) contained therein.

The method according to the invention allows to significantly reduce the amount of fluoride source and simplify the process without impairing yield and quality of the desired onium fluoride. Moreover, the reaction can be carried out in steel reactors without risking (premature) corrosion.

As explained above, onium fluorides are suitable as catalysts.

The present invention relates thus also to the use of an onium fluoride of the formula (I) as defined above (or of the product obtained with the process according to the invention) as a catalyst in the preparation of isocyanate trimers containing iminooxadiazinedione groups, in particular of isocyanate trimers containing iminooxadiazinedione groups and derived from (cyclo)aliphatic diisocyanates.

The present invention relates also to a process for preparing isocyanate trimers containing iminooxadiazinedione groups, comprising reacting at least one (cyclo)aliphatic diisocyanate in the presence of onium fluoride catalyst of the formula (I) as defined above (or with the product obtained with the process according to the invention), and when the reaction has reached a predetermined degree of conversion of the (cyclo)aliphatic diisocyanates, stopping the reaction by addition of at least one catalyst poison for the catalyst, and if necessary separating off unreacted (cyclo)aliphatic diisocyanate.

In the preparation of polyurethanes, the isocyanate starting materials are often provided in form of their cyclic trimers. Symmetric cyclic trimers are isocyanurates, while non-symmetric cyclic trimers are iminooxadiazinedione. Suitable isocyanate starting materials are all those known in the art. “(Cyclo)aliphatic” means cycloaliphatic, aliphatic (aliphatic meaning non-cyclic) and mixed cycloaliphatic- aliphatic. Examples for suitable (cyclo)aliphatic diisocyanates are tetramethylene diisocyanate, pentamethylene diisocyanate (PDI), hexamethylene diisocyanate (HMDI or HDI), dodecyl diisocyanate, 1 ,4-diisocyanato-4-methylpentane, 2-methylpentane-

1.5-diisocyanate, 2-butyl-2-ethylpentamethylene diisocyanate, 2,2,4- or 2,4,4-trimethyl-

1.6-hexamethylene diisocyanate, lysine alkyl ester diisocyanate, where alkyl stands for Ci-C -alkyl, as well as cycloaliphatic isocyanates such as isophoronediisocyanate (5- isocyanato-1-(isocyanatomethyl)-1 ,3,3-trimethylcyclohexane; IPDI), 1,4- bisisocyanatocyclohexane, bis-(4-isocyanatocyclohexyl)methane, 2- isocyanatopropylcyclohexyl isocyanate, 3(4)-isocyanatomethyl-1 -methylcyclohexyl isocyanate, 2,4'-methylenebis(cyclohexyl) diisocyanate, and 4-methylcyclohexane 1,3- diisocyanate (H-TDI). Among these, preference is given to hexamethylene diisocyanate (HDI), pentamethylene diisocyanate (PDI), 5-isocyanato-1-(isocyanatomethyl)-1 ,3,3- trimethylcyclohexane (IPDI), 2-methyl pentane- 1,5-diisocyanate, 2,4,4-trimethyl-1 ,6- hexane diisocyanate, 2,2,4-trimethyl-1,6-hexane diisocyanate and 4-isocyanatomethyl- 1,8-octane, and more preference to hexamethylene diisocyanate (HDI) and pentamethylene diisocyanate (PDI). Specifically, the (cyclo)aliphatic diisocyanate is hexamethylene diisocyanate.

The catalyst (I) is used in amount of preferably 20 ppm to 500 ppm, more preferably 50 ppm to 300 ppm, in particular 50 ppm to 150 ppm, based on the weight of the (cyclo)aliphatic diisocyanate. In this context, 1 ppm corresponds to 0.0001% by weight (10^-4 % by weight), relative to the total weight of the reference substance.

The process can be carried out according to generally known methods as described, for example, in EP 0962455, and preferably comprises dissolving the catalyst in a solvent (if not yet present in solution, as obtained for example in step (vii) described above) and adding the (obtained) solution to the (cyclo)aliphatic diisocyanate. Suitable and preferred solvents are those listed above in context with step (i) and/or (iii).

The reaction temperature is generally not critical can be in the range of 20 to 200°C or 30 to 120°C or 40 to 100°C.

When the reaction has reached a predetermined degree of conversion of the (cyclo)aliphatic diisocyanates, the reaction is stopped by addition of at least one catalyst poison for the catalyst.

The reaction is preferably stopped when the degree of reaction RNCO, which is calculated as the quotient of the difference between the NCO content of the starting isocyanate before trimerization and the NCO content of the reaction mixture after termination of the reaction divided by the NCO content of the starting isocyanate before trimerization, is 3% to 60%, preferably 3% to 50%, more preferably 5% to 20%.

The catalyst poison is preferably selected from organic and inorganic acids and acid derivatives (but not HF). Suitable acid derivatives are the halides (especially the acid chlorides) and anhydrides. Examples for suitable organic and inorganic acids and acid derivatives are sulfonic acids, e.g., p-toluene sulfonic acid or dodecyl benzene sulfonic acid; benzoic acid, benzoyl chloride, phosphoric acid, acidic esters thereof (the esters being suitably Ci-Cs-alkyl esters, e.g., C4- or Cs-dialkylesters, such as dibutyl- or di-(2- ethylhexyl)-phosphate), phosphorous acid and acidic esters thereof (here, too, the esters being suitably Ci-Cs-alkyl esters). Among these, preference is given to p-toluene sulfonic acid.

After deactivation, the deactivated catalyst may be removed from the reaction mixture by known means, such as adsorptive binding of the catalyst and subsequent removal by filtration and thermal deactivation.

Any unreacted monomer may, after deactivation (and removal, if desired) of the catalyst system, be separated off by known methods, for example by (thin-layer) distillation or extraction, and then recycled.

The thusly prepared isocyanate cyclic trimers are suitable isocyanate components for the production of polyurethanes. Before being subjected to the reaction to form polyurethanes, they may optionally be modified by reacting the isocyanate groups to incorporate urethane, urea, biuret and/or allophanate groups or by reacting some or all of the NCO groups with reversible blocking agents. Suitable blocking agents include phenols, lactams such as epsilon-caprolactam, oximes, di- and triazoles, amines such as diisopropylamine and CH-acid compounds such as malonic acid dialkyl esters and acetoacetic ester.

The thusly prepared isocyanate cyclic trimers, optionally in blocked form, are especially suitable for the manufacture of optionally water-dispersible one- and two-component polyurethane coating compositions because their viscosities are reduced when compared to isocyanurate-polyisocyanates, while their properties profile is equally high or is improved. They are more stable towards the occurrence of flocculation or turbidity, even when highly diluted in lacquer solvents, when compared to corresponding products containing mainly isocyanurate groups. Their resistance towards the effects of moisture (e.g., the formation of a skin in open packaging or the matt appearance of surfaces lacquered at high humidity and a high ambient temperature, so-called "downglossing") is also improved when compared with products containing isocyanurate groups. The present invention also relates to a process for producing polyurethane coatings, comprising reacting the cyclic trimeric isocyanates obtained according to the process described above with at least one binder selected from the group consisting of polyacrylate polyols, polyester polyols, polyether polyols, polyurethane polyols, polyurea polyols, polyetherols, polycarbonates, polyester polyacrylate polyols, polyester polyurethane polyols, polyurethane polyacrylate polyols, polyurethane- modified alkyd resins, fatty acid-modified polyester polyurethane polyols, copolymers with allyl ethers and copolymers or graft polymers thereof.

The present invention relates moreover to the use of the cyclic trimeric isocyanates obtained according to the process described above as a curing agent, preferably in a material selected from the group consisting of coating materials in primers, primer surfacers, pigmented topcoats, basecoats and clearcoats in the sectors of refinishing, automotive refinishing, large vehicle finishing and wood, plastic, and OEM finishing, in utility vehicles in the agricultural and construction sector and as curing agent in adhesives and sealants.

The invention is now illustrated by the following examples.

EXAMPLES

Example 1

In a stainless steel reactor, 760 kg of Cyphos® 443P (70% solution of tetra-n-bu- tylphosphonium chloride in isopropanol; from Cytec; 1.804 kmol) was diluted with 760 kg of methanol. Prior to this, the hydrogen chloride content in the Cyphos® starting material was determined by acid-base titration. The hydrogen chloride was neutralized by addition of an equimolar amount of potassium methylate, used as a 30% solution in methanol (49 kg, 0.21 kmol), to the diluted solution, whereupon the solution immediately became turbid and potassium chloride precipitated. To this neutralized solution were added 120 kg of potassium fluoride (2.065 kmol) and the reaction mixture stirred for 8 hours at 30°C. The chloride content of the liquid phase was steadily < 0.2% (determined according to Mohr’s method (ISO 9297:1989-11) by titration with silver nitrate). Precipitated potassium chloride together with excess potassium fluoride was filtered off via a suction filter, and the filter cake was washed with 160 kg of methanol. Isopropanol and methanol were distilled off from the filtrate in vacuo at 30°C, resulting in a concentrate containing 19% by weight of methanol and 8% by weight of isopropanol, the chloride content being < 0.4% by weight, relative to the total weight of the composition. The concentrate was filtered, cooled to 10°C and 35 kg of anhydrous liquid hydrogen fluoride (1.75 kmol) were added at < 30°C. The thusly obtained tetrabutylphosphonium hydrogen difluoride solution was filled into a container for further use and storage.

The low chloride content in the concentrate (< 0.4% by weight, relative to the total weight of the concentrate) shows that conversion and yield are essentially quantitative.

No surface corrosion in the steel reactor was observed.

For comparative reasons, the above process was repeated. However, after filtering off potassium chloride and excess potassium fluoride and washing the filter cake with methanol, a second portion of 120 kg potassium fluoride was added. Analysis of the solution showed that the second addition did not further reduce the chloride content in solution. It is thus clear that this second addition can be skipped without impairing the quality and quantity of the obtained product.

Comparative example 1

The following method corresponds essentially to that of EP 0962455.

In a stainless steel reactor, 380 kg of Cyphos® 443P (70% solution of tetra-n-bu- tylphosphonium chloride in isopropanol; from Cytec; 0.902 kmol) was diluted with 400 kg of methanol. 60 kg of potassium fluoride (1.035 kmol) were added, and the mixture was stirred at 30°C for 8 hours. The chloride content of the liquid phase was 0.6 to 0.7%. Precipitated potassium chloride together with excess potassium fluoride was filtered off via a suction filter, and the filter cake was washed with 80 kg of methanol. To the filtrate, another 60 kg of potassium fluoride (1 .035 kmol) were added, and the reaction mixture was stirred for another 4 hours at 30°C until a chloride content of < 0.2% (determined according to Mohr’s method (ISO 9297:1989-11) by titration with silver nitrate) was obtained. The precipitated salts were again filtered off and washed with methanol. Isopropanol and methanol were distilled off from the filtrate in vacuo at 30°C, resulting in a concentrate containing 20% by weight of methanol and 8% by weight of isopropanol, the chloride content being approx. 0.4%. The concentrate was filtered, cooled to 10°C and 17.5 kg of anhydrous liquid hydrogen fluoride (0.875 kmol) were added at < 30°C. The thusly obtained tetrabutylphosphonium hydrogen difluoride solution was filled into a container for further use and storage.

The low chloride content in the concentrate (ca. 0.4% by weight, relative to the total weight of the concentrate) shows that conversion and yield are essentially quantitative.

Surface corrosion in the steel reactor was observed yet after just one pass. The comparison of example 1 and comparative example 1 shows that the method of the invention allows to simplify the conversion of onium chlorides to the corresponding onium fluorides by requiring just one addition step of the fluoride salt, and to distinctly reduce the amount of required fluoride salt for obtaining quantitative conversion. Moreover, corrosion can be reduced.

Claims

1. A method for producing an onium fluoride of the formula (I) R₄Q⁺ F- ■ nxHF (I) where

Q is N or P; each R is independently selected from the group consisting of Ci-C2o-alkyl, C3- Cs-cycloalkyl, Cs-Cs-cycloalkyl-Ci-C^alkyl, Ce-C -aryl, C6-C -aryl-Ci-C4- alkyl, and a 3- to 8-membered saturated, partially unsaturated or maximally unsaturated heterocyclic ring containing 1 , 2 or 3 heteroatoms or heteroatom groups selected from the group consisting of O, N, S, S(O) and S(O)2 as ring members; or two R form together a C2-C2o-alkylene bridging group which may be interrupted by 1 , 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1 , 2 or 3 aromatic or heteroaromatic rings; and the other two R are either as defined above or, independently, also form together a C2-C2o-alkylene bridging group which may be interrupted by 1 , 2 or 3 heteroatoms selected from the group consisting of O, N and S, where two O may not be adjacent, and/or by 1 , 2 or 3 5- or 6-membered aromatic or heteroaromatic rings; and n is from 0.1 to 5; which method comprises

(i) providing a solution of an onium chloride of the formula (II)

R₄Q⁺ CI- (II) where Q and R are as defined above, in a solvent;

(iv) reacting the reaction mixture obtained in step (iii) with a fluoride salt of the formula (M^m+)(F )_m, where M is an alkali metal or alkaline earth metal and m is the charge thereof, m being thus 1 if M is an alkali metal and 2 if M is an alkaline earth metal; (v) removing any precipitate present in the reaction mixture obtained in step (iv);

(vi) partially or completely removing the solvent; and

2. The method according to claim 1 , where Q is P.

3. The method according to any of claims 1 or 2, where each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and Cs-Ce-cyclo- alkyl-Ci-C2-alkyl; or two R form together a C4-C6-alkylene bridging group and the other two R are either independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cy- cloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl, or, independently, also form together a C4-Ce-alkylene bridging group; where preferably each R is independently selected from the group consisting of Ci-C2o-alkyl, Cs-Ce-cycloalkyl and C3-C6-cycloalkyl-Ci-C2-alkyl.

4. The method according to claim 3, where each R is independently Ci-C2o-alkyl, preferably C2-Cio-alkyl.

5. The method according to claim 4, where each R is independently C2-Ce-alkyl, preferably Cs-Cs-alkyl, and where more preferably each R is n-butyl.

6. The method according to any of the preceding claims, where n is 0.5 to 2, and is preferably approximately 1.

7. The method according to any of the preceding claims, where the solvents of steps (i) and (iii) are independently selected from the group consisting of Ci-Cs- alkanols, C2-Cs-alkanediols and mixtures thereof.

8. The method according to claim 7, where the solvents of steps (i) and (iii) are independently selected from Ci-Cs-alkanols, preferably from methanol, isopropanol, 2-ethylhexanol and mixtures thereof.

9. The method according to claim 8, where the solvents of steps (i) and (iii) are independently selected from Ci-Cs-alkanols, preferably from methanol, isopropanol and mixtures thereof, and specifically methanol, optionally in admixture with isopropanol.

10. The method according to claim 9, where the solvent of step (i) is selected from methanol, isopropanol and mixtures thereof, and is specifically isopropanol; and the solvent of step (iii) is methanol.

11. The method according to any of the preceding claims, where the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal hydroxides, alkali metal or alkaline earth metal carbonates, alkali metal or alkaline earth metal hydrogen carbonates and alkali metal or alkaline earth metal Ci-Cs- alkanolates.

12. The method according to claim 11 , where the base of step (iii) is selected from the group consisting of alkali metal Ci-Cs-alkanolates; preferably from alkali metal Ci-Cs-alkanolates and more preferably from potassium Ci-Cs-alkanolates.

13. The method according to any of the preceding claims, where the fluoride salt of step (iv) is of the formula MF, where M is an alkali metal, preferably sodium or potassium, more preferably potassium.

14. The method according to any of the preceding claims, where the solvents of steps (i) and (iii) are independently a Ci-Cs-alkanol or a mixture of different Ci- Cs-alkanols and the base of step (iii) is selected from the group consisting of alkali metal or alkaline earth metal Ci-Cs-alkanolates, where the Ci-Cs-alkanolate is derived from the Ci-Cs-alkanol or one of the Ci-Cs-alkanols of the mixture of different Ci-Cs-alkanols used as solvent in step (i) and/or (iii), and the alkali metal or alkaline earth metal of the base corresponds to M of the fluoride salt used in step (iv).

15. The method according to claim 14, where the solvents of steps (i) and (iii) are independently methanol, isopropanol or a mixture of methanol and isopropanol, where preferably at least one of the solvents of steps (i) and (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate; and the fluoride salt used in step (iv) is KF.

16. The method according to claim 15, where Q is P; R is C2-Cio-alkyl, preferably C2- Ce-alkyl, more preferably Cs-Cs-alkyl and specifically n-butyl; n is 0.5 to 2; the solvents of steps (i) and (iii) are independently methanol, isopropanol or a mixture of methanol and isopropanol, where preferably at least one of the solvents of steps (i) and (iii) is or comprises methanol; the base used in step (iii) is potassium methanolate and the fluoride salt used in step (iv) is KF.

17. The method according to any of the preceding claims, where one, two or all three of the following conditions a), b) and/or c) apply: a) the reaction mixture to be reacted in step (iv) contains the onium chloride of the formula (II) in an amount of from 0.5 to 3 mol, preferably from 1 to 2.5 mol and more preferably from 1.5 to 2.2 mol per kg of solvent; and/or b) the fluoride salt of the formula (M^m+)(F')_m is used in an amount of from 1 to 2 mol, preferably from 1.05 to 1.5 mol and more preferably from 1.1 to 1.2 mol per mol of the onium chloride of the formula (II) provided in step (i), where the amount of the fluoride salt relates to the amount of F contained therein; and/or c) anhydrous hydrogen fluoride is used in step (vii) in an amount of from 0.1 to 5 mol, preferably from 0.5 to 2 mol, more preferably in an amount of approximately 1 mol per mol of the onium chloride of the formula (II) provided in step (i).

18. The method according to any of the preceding claims, where one, two or three of the following condition d), e) and/or f) apply: d) step (iv) is carried out at a temperature of from 0 to 50°C, preferably from 10 to 40°C and more preferably from 20 to 35°C; and or e) in step (vi) partially or completely removing the solvent is carried out under reduced pressure and at a temperature of from 25 to 50°C, preferably from 25 to 45°C; and/or f) step (vii) is carried out at a temperature of from 0 to 40°C, preferably from 5 to 30°C and more preferably from 10 to <30°C.

19. The method according to any of the preceding claims, where in step (vi) the solvent is removed only partially.