WO1997002326A1

WO1997002326A1 - Reactive diluents

Info

Publication number: WO1997002326A1
Application number: PCT/GB1996/001569
Authority: WO
Inventors: Henry Deans; Benjamin Patrick Gracey; Christopher Hallett; Philip Kenneth Gordon Hodgson; Robert Paul Klaasen
Original assignee: BP Chemicals Ltd
Current assignee: BP Chemicals Ltd
Priority date: 1995-06-30
Filing date: 1996-06-28
Publication date: 1997-01-23
Anticipated expiration: 1997-12-30
Also published as: AU6310996A; WO1997002230A1; GB9513317D0; AU6311096A

Abstract

This invention relates to a formulation comprising a paint or a polymer and an ester product comprising one or more esters selected from the group consisting of: (a) 2-(2,7-octadienoxy) di-2,7-octadienyl succinate, (b) di-(2,7-octadienyl) fumarate and (c) di-(2,7-octadienyl) maleate, said ester product having a viscosity of less than 200 mPa s (cP). The esters function as reactive diluents in such formulations, especially in formulations which have alkyd resins as the polymer component.

Description

REACTIVE DILUENTS

This invention relates to use of esters of 2,7-octadienol as reactive diluents in paint and polymer formulations.

Methods of producing esters of unsaturated alcohols are well known. One such method is the direct telomerisation of butadiene with a carboxylic acid in the presence of a homogeneous catalyst comprising a noble metal compound such as eg palladium acetyacetonate and a phosphine. Such a method is described for instance in JP-A-51088907 (Mitsubishi Chemical Industries Co). In a further process, octadienol is reacted with a carboxylic acid or anhydride in the presence of other catalysts (see JP- A-47046568 (Mitsubishi) which describes the use of a zinc acetate catalyst and JP-A- 04049266 (Kuraray) which describes the use of a KCN transesterification catalyst).

Esters of octadienol, especially the maleates and fumarates are well known and methods of their preparation are described in JP-A-47 -46568 (Mitsubishi). The use of these esters as (a) co-monomers during polymerization of other monomers to form thermosetting resins, (b) plasticizers, (c) lubricating oils and (d) textile treatment agents are disclosed in this reference. Similarly JP-A-49-28175 (Mitsubishi) describes the use of some of these synthetic esters as drying oils. Again US-A-4378251 (Eastman Kodak) describes the use of some of these esters to form polymers with a viscosity of > 200 cP, preferably from 700-900 cP in coating compositions which contain in addition an organic salt of cobalt or manganese. Furthermore, US-A- 4408029 (Eastman Kodak) claims a low molecular weight polymer derived from octadienyl maleate or fumarate and having a fluid viscosity of at least 200 cP.

None of these prior published documents disclose the use of octadienyl maleates, fumarates or succinates having a viscosity of less than 200 cP as reactive diluents. Reactive diluents are usually compounds or mixtures of compounds of relatively low viscosity, a relatively high boiling point (i.e. low saturated vapour pressure) which act as solvents during the formulation and processing of the coating. A feature of reactive diluents is that such diluents can copolymerise with a resin, e.g. an alkyd. Hence reactive diluents may be used to replace part or all of the traditional solvents normally used in such formulations thereby reducing losses of the solvent to atmosphere on drying of the coating. Use of esters of di- and polyhydric alcohols which have been partially etherified with allyl alcohol as reactive diluents is described for example in EP-A-0 263 474.

Alkyd resins are well known components of decorative paints (see, for example, "The Technology of Paints, Varnishes and Lacquers" by Martens, C R, Ed., published by Robert Krieger Publishing (1974)) and can be prepared from polybasic acids or anhydrides, polyhydric alcohols and fatty acids or oils. US-A-3 819 720 describes methods of preparing such alkyd formulations. Alkyd resins are available commercially and are used in coating compositions which usually contain large amounts of solvents (eg white spirits, aromatic hydrocarbons).

It has now been found that esters of octadienyl alcohols which can be produced in commercially viable yields and purity have relatively low viscosity and therefore can be used as reactive diluents in a wide variety of polymer formulations.

Accordingly, the present invention is a formulation comprising a paint or a polymer and an ester product comprising one or more of the esters selected from the group consisting of: a. 2-(2,7-octadienoxy) di-2,7-octadienyl succinate, b. di-(2,7-octadienyl) fumarate and c. di-(2,7-octadienyl) maleate, said ester product having a viscosity of less than 200 mPa s (cP).

The esters can be derived by reacting in the presence of a catalyst a compound of the formula:

CH2=CH.CH₂.CH₂.CH2.CH=CH.CH2.OR (I) with a dicarboxylic compound of the foπnula: R'.C(O)CH=CH.C(O).R" (II) wherein R is H or an acyl group having 1-6 carbon atoms,

R' and R" may be OH or the same or different C1-C10 alkoxyl group, or, R' and R" may be such that the two taken in combination represent an -O- function thereby making (II) an acid anhydride. Thus, in the present invention (I) may be 2,7-octadienol or an alkyl ester thereof. Again, (II) may be maleic anhydride, maleic acid, fumaric acid or a dialkyi ester of maleic or fumaric acid, especially the dimethyl esters. It is to be noted that 2,7-octadienol is normally produced by telomerisation and such a process yields a mixture of isomeric octadienols ie l,7-octadien-3-ol and 2,7-octadienol. This mixture can be used to prepare the esters.

The reaction between (I) and (II) is suitably carried out in the liquid phase and in the presence of a catalyst which does not cause undue polymerisation or rearrangement of the allylic groups the molecule of (I) thereby giving rise to hydrocarbyl fragments capable of forming coloured products. Examples of a suitable catalysts include acidic or amphoteric catalysts, and may be homogeneous or heterogeneous. Specific examples of such catalysts include inter alia dibutyl tin oxide, stannous oxalate, zinc acetate, magnesium acetate, para-toluene sulphonic acid, methane sulphonic acid and phosphoric acid.

The esterification reaction is suitably carried out at a temperature from 80-200°C, preferably from 100-160°C. Witliin these temperature ranges, the viscosity of the ester formed tends to increase at the higher temperatures due to resinification of the carboxylate reactant or ester product.

It will be understood by those skilled in the art that the esterification of the octadienyl compound (I) with the dicarboxylic compounds (II) to fonn the desired ester either by direct- or trans-esterification may lead to a mixture of esters under some reaction conditions. Such a mixture of esters can be formed due to incomplete transesterification. This will be especially true where less than one molar equivalent of the octadienyl compound (I) is used per mole of the dicarboxylic compound (II) in the reaction. Thus, the reaction is suitably earned out using a molar ratio of (I) : (II) in the range of 2 : 1 to 10 : 1, preferably from 2.5 : 1 to 5 : 1. In this case, the preparative conditions employed will have a strong influence on the type of material obtained from the esterification (or transesterification) reaction, including those used for producing the esters of octadienol (I). In addition to the degree of esterification (or transesterification) achieved, there is the possibility that samples prepared at relatively higher temperatures could isomerise the maleate/fumarate and may have relatively higher densities due to some resinification. One of the esters in the ester product foπned contains an additional alkyleneoxy function in its structure arising from the Michael addition of the alcohol reactant across the unsaturated linkage of the unsaturated acid or anhydride or ester. Thus for example, when octadienol is reacted with dimethyl maleate, the product contains 2-(2,7-octadienoxy) di-(2,7-octadienyl) succinate. However, the reaction mixture may comprise primarily 2-(2,7-octadienoxy) di- (2,7 -octadienyl) succinate, di-(2,7-octadienyl) fumarate and di-(2,7-octadienyl) maleate. A mixture comprising these esters will hereafter be termed the "ester product". The esterification reaction conditions are so chosen that the viscosity of the resultant ester product does not exceed 200 mPa s (cP).

Thus, according to the present invention the ester product in the paint or polymer formulation comprising one or more of the esters selected from the group: a. 2-(2,7-octadienoxy) di-2,7-octadienyl succinate, b. di-(2,7-octadienyl) fumarate and c. di-(2,7-octadienyl) maleate, and said ester product has a viscosity of less than 200 mPa s (cP).

The ester product of the present invention can function essentially as reactive dilutents when used in paint and polymer formulations. In this context it is not necessary to isolate the individual esters in the mixed ester product formed during its synthesis. The mixed ester product may be used as such, if necessary after separation of the catalyst, colour foπning impurities, acid and excess alcohol reactant, for use as reactive diluents. Thus, curing of tlie reactive diluent after applying the formulation on a substrate surface is likely to result in the formation of a coating of hardened material. Ester products having low volatility and low viscosity ie less than 200 mPa s (cP), and usually of the order of eg 10-80 mPa s (cP), can be produced by the process of the present invention. Such properties enable them to be used as reactive diluents for cured paint and polymer formulations, especially for formulations comprising alkyd resins. Again, the low viscosity esters of octadienol are of relatively low colour. Moreover, films formed from paint formulations containing the low viscosity esters of octadienol as reactive diluent show relatively little yellowing with passage of time. The relative ratios of the low viscosity esters used as reactive diluents to the alkyd resin in a formulation can be derived from the ranges quoted in published EP-A-0 305 006. In an example of a formulation comprising high solids paints in which the reactive diluent replaces all of the traditional solvent, the ratio of reactive diluent to alkyd resin is suitably in the range from 1-50 : 99-50 parts by weight, eg 5-50 : 95-50 parts by weight, preferably from 5-25 : 95-75 and more preferably from 5-15 : 95-85 parts by weight. On the other hand, where used in a conventional paint formulation, such a diluent can replace all or part of a ttaditional solvent such as white spirit. The formulations may contain further components such as catalyst, drier, antiskinning agent, pigments, pigment stabilisers, rheology controllers (e.g. for sag control), UV and oxidation stabilisers, flow additives, microgels (e.g. to enhance hardness) and other additives. The formulations may also need to include water scavengers such as trialkyl orthoformates, molecular sieves or zeolites where the reactive diluent used is susceptible to hydrolysis such as eg some of the ester derivatives. Furthermore, where such water scavengers are used it may be necessary to use them in combination with pigment stabilizers. Where a drier (siccative) is used this may further contribute towards the solvent content of the formulation.

For formulations comprising an oxidatively curing alkyd resin and a siccative/drier such as cobalt complexes, impurities which can have a co-ordination affinity for the siccative drier such as cobalt complexes can affect adversely the drying speed and stability of the paint. Examples of such impurities include maleic acid and triethyl amine. In particular, it has been found desirable to minimise the acidity of the ester mixture used as reactive diluent in such foπnulations to a value of < 7000 ppm, preferably < 3000 ppm, more preferably < 1000 ppm w/w of KOH. It has also been found that when a mixture of esters, ie the succinates, fumarates and maleates, is used as a reactive diluent in such formulations comprising an oxidatively curing alkyd resin, the properties/performance of the diluent can be varied by changing the relative proportions of the three esters present in such a diluent For example, mixtures with a relatively lower amount of maleates exhibit better hardness and drying properties compared with those having relatively higher amounts of such maleates. Moreover, it has also been observed that formulations comprising these esters display a decreased tendency towards wrinkling especially in high solids formulations. This renders them particularly suitable when using formulations comprising high solid systems/one-coat paints have to be applied to generate a greater thickness of the relevant coating without impairing the ability of such thicker layers to harden through.

A further aspect of the present invention is that such esters when used in a relatively pure state do not cause any haze in the formulation. Where there is likely to be a risk of such haze formation, eg due to the presence of impurities such as eg resins or polymers formed during the synthesis of the esters used or during storage of such formulations, it is beneficial to use inhibitors such as eg butylated hydroxy-toluene (2,6-butoxy-4-methyl phenol) and 2,4,6-tert-bιιtyl phenol.

Such inhibitors not only have the advantage of preventing haze formation but also render the foπnulations safer to handle by inhibition of other unwanted reactions in the formulation such as eg peroxidation. The present invention is further illustrated with reference to the following Examples: EXAMPLES:

1. Examples of the preparation of the reactive diluents Example SI:

The following apparatus was assembled: A five-litre flanged flask with an insert pipe for a nitrogen sparge, a thermowell for thermocouple, and a Dean and Stark apparatus with double-walled condenser. The flask was heated with an electric heating mantle which was controlled with a eurotherm controller connected to the thermocouple. The nitrogen sparge pipe was inserted so that the nitrogen flow agitated the flask contents and provided mixing during the course of the reaction. The nitrogen flow also served to entrain out the liberated methanol and force the reaction to completion.

To the flask was added dimethyl maleate (867.2g), 2,7-octadienol (2301g) and zinc acetate (31.44g). The mixture was sparged with niϋOgen for 10 minutes to remove air and the nitrogen flow was then reduced to a level which ensured efficient mixing. The mixture was then heated in stages to 130°C (ie 80°C for 10 minutes, then 100°C for 10 minutes and then 120°C for 10 minutes). The progress of the reaction was monitored by the methanol collected in the Dean and Stark apparatus. When 90% of the predicted methanol had been collected, the reaction mixture was sampled hourly and analysed by GC. The reaction was adjudged complete when the level of the "half ester" (methyl octadienyl maleate/fumarate) fell to below 0.3% w/w, this took approximately 8 hours. At this point the heating was switched off and the reaction mixture allowed to cool to room temperature. The product from the reaction was then decanted from any solids in the reaction flask. This product was then charged to a heated decanter (40°C) with an equal volume of 5% w/w aqueous sodium hydroxide solution. The mixture was stirred for 20 minutes and then allowed to separate and the lower aqueous phase decanted. The base wash was then repeated and the remaining organic phase was washed with saturated brine until the aqueous phase reached a steady pH. The organic phase was then heated (100°C) under reduced pressure (< 500 Pa (< 5 mBar)) on a rotatory evaporator to remove residual water and the majority of the excess octadienol. After cooling, the product was filtered and transferred to a 5-litre three necked round bottomed quickfit flask . This flask was equipped with a still head condenser and receiver flask ( Perkin triangle), a thermocouple, a steam inlet pipe, and a eurotheπn conuOlled heating mantle. The apparatus was evacuated to 4000 Pa (40 mBar) and the product heated to 120°C. The supply of steam was then connected and the residual traces of octadienol were removed. The purification was judged complete when the volume of the heads product aqueous phase increased to more than 5 times that of the organic phase. After cooling down, the product was then treated with activated carbon (l%w/w, 100°C 2Hrs, <500 Pa (<5 mBar) on a rotatory evaporator. The cooled mixture was filtered through dried celite to obtain tlie final product which had the following analyses: OH number - 7 mg KOH/g (titration) total acid - 226 ppm KOH/g (titration) maleic acid / anhydride - < lOppm (HPLC)

Fumaric acid - <10ppm (HPLC)

Zinc - <5ppm (atomic absoφtion, detection limit) sodium - 30ppm ( atomic absorption) chlorine - <10ppm (atomic absorption detection limit) GC "CPSil5" column - 2,7-octadienyl methyl fumarate/ maleate

(0.1% w/w) di-(2,7-octadienyl) maleate (11% w/w) di-(2,7-octadienyl) fumarate (43% w/w) 2-(2,7-octadienoxy) di-(2,7-octadienyl) succinate (41% w/w) higher boilers (0.01% w/w) The GC assignment was supported by GC/MS and a Hnmr and C nmr studies. The GC/MS used a VG Trio- 1000, operated according to the manufacturers instructions under the following conditions: - GC column 25mx0.32mm DB5 (0.25 micron film) temperature programme 40°C (3 mins) @ 10°C/min to 320°C(10mins) injection 1 microlitre (1 % solution in acetone) on column 40°C ammonia chemical ionisation (CI) - scan range 50-800 scan rate 1/s It was found that the deduction of molecular weights from the CI spectra is rather less straightforward than is usual on account of (a) extensive rearrangements of

+ + fumarates in paiticular giving [M+3] and [M+20] ions in addition to the usual [M+l] and [M+l 8] ions and (b) extensive fragmentation exhibited by some species. As a result, the GC peaks have been assigned by interpretation. These assignments

1 13 were confirmed by H and C nmr. Table 1 gives assignments of the observed ¹³C nmr peaks. It should be noted that the two isomeric octadienols ( 2,7-octadienol and l,7-octadien-3-ol) though not separable by the GC method used can nevertheless be identified by nmr and are recorded in tlie nmr assignment Table 1. The correspondence to the GC was again confirmed by integration of the nmr spectrum of several samples in which the composition varied. The product of this Example will hereafter called sample AK1. Example S2: The following apparatus was assembled:

A five-litre flanged flask with an insert pipe for a nitrogen sparge, a thermowell for thermocouple, and a Dean and Stark apparatus with double-walled condenser. The flask was heated with an electric heating mantle which was controlled with a eurotherm controller connected to the theπnocouple. The nitrogen sparge pipe was inserted so that the nitrogen flow agitated the flask contents and provided mixing during the course of the reaction. The nitrogen flow also served to entrain out the liberated methanol and force the reaction to completion.

To the flask was added dimethyl maleate (1004.1g), 2,7-octadienol (2620.2g) and stannous oxalate (36.5g). The mixture was sparged with nitrogen for 10 minutes to remove air and the nitrogen flow was then reduced to a level which ensured efficient mixing. The mixture was then heated in stages to 130°C ( e.g. 80°C for 10 minutes, then 100°C for 10 minutes and then 120°C for 10 minutes). The progress of the reaction was monitored by the methanol collected in the Dean and Stark apparatus. In order to drive the reaction to completion the temperature was raised to 140°C after 40hrs at 130°C. When 90% of the predicted methanol had been collected the reaction mixture was sampled hourly and analysed by GC. The reaction was adjudged complete, when the level of the "half ester" (methyl octadienyl maleate/fumarate) fell to below 0.3% w/w and this took approximately 72 hours. At this point the heating was switched off and the reaction mixture allowed to cool to room temperature. The product from the reaction was then decanted from any solids in the reaction flask. This product was then charged to a heated decanter (40°C) with an equal volume of 5% w/w aqueous sodium hydroxide solution. The mixture was stirred for 20 minutes and then allowed to separate and the lower aqueous phase decanted. This base wash was then repeated and the remaining organic phase was washed with saturated brine until the aqueous phase reached a steady pH. The organic phase was then heated (100°C) under reduced pressure (< 500 Pa (< 5 mBar)) on a rotatory evaporator to remove residual water and the majority of the excess octadienol. After cooling, the product was filtered and transferred to a 5-litre three-necked round-bottomed quickfit flask . This flask was equipped with a still head condenser and receiver flask (Perkin triangle), a thermocouple, a steam inlet pipe, and a eurotherm controlled heating mande. The apparatus was evacuated to 4000 Pa (40 mBar) and the product heated to 120°C. The supply of steam was then connected and the residual traces of octadienol were removed. The purification was judged complete when tlie volume of the heads product aqueous phase increased more than 5 times that of the organic phase. After cooling down, the product was then treated with activated carbon (1% w/w, 100°C 2Hrs, <500 Pa (< 5 mBar)) on a rotatory evaporator. The cooled mixture was filtered through dried celite to obtain the final product which had the following analyses: OH number - 1 mg KOH/g (titration) total acid - 113 ppm KOH/g (titration) maleic acid / anhydride - < lOppm (HPLC)

Fumaric acid - <10ppm (HPLC) tin - <5ppm (atomic absorption, detection limit) sodium - <20ppm ( atomic absoφtion, detection limit) chlorine - <10ppm (atomic absoφtion detection limit) GC "CPSil5" column - 2,7-octadienyl methyl fumarate/ maleate

(1.1% w/w) di-(2,7-octadienyl) maleate (69% w/w) di-(2,7octadienyl) fumarate (22% w/w) 2-(2,7-octadienoxy) di-(2,7-octadienyl) succinate (5% w/w)

The GC assignment was supported by GC/MS and a H nmr and ^" C nmr studies. The GC/MS used a VG Trio- 1000, operated according to the manufacturers instructions under the following conditions:

GC column 25mx0.32mm DB5 (0.25 micron film) - temperature programme 40°C (3 mins) @ 10°C/min to

320°C (10mins) injection 1 microlitre (1 % solution in acetone) on column 40°C ammonia chemical ionisation (CI) scan range 50-800 - scan rate 1/s It was found that the deduction of molecular weights from the CI spectra is rather less straightforward than is usual on account of (a) extensive rearrangements of fumarates

+ + + in particular giving [M+3] and [M+20] ions in addition to the usual [M+l] and

[M+l 8] ions and (b) extensive fragmentation exhibited by some species. As a result the GC peaks were assigned by inteφretation. In addition to the assigned peaks an additional species was identified which was assigned to a lactone.

1 13

This assignment was confirmed by H and C nmr. Table 1 gives assignments of the observed C nmr peaks. It should be noted that the two isomeric octadienols ( 2,7-octadienol and l,7-octadien-3-ol) though not separable by the GC method used can be identified by nmr and are recorded in the nmr assignment Table 1. The correspondence to the GC was confirmed by integration and analysis of several samples in which the composition varied. The lactone found by GC/MS was also observed in the nmr and quantified at approximately 7.1% (tenative structure given in Table 1). The product of this Example will hereafter called sample AK2. Example S3:

The following apparatus was assembled: A five-litre flanged flask with an insert pipe for a nitrogen sparge, a thermowell for thermocouple, and a Dean and Stark apparatus with double-walled condenser. The flask was heated with an electric heating mantle which was controlled with a eurotherm controller connected to the theπnocouple. The nitrogen sparge pipe was inserted so that the nitrogen flow agitated the flask contents and provided mixing during the course of the reaction. The nitrogen flow also served to entrain out the liberated methanol and force the reaction to completion.

To the flask was added dimethyl maleate (914.3g), 2,7-octadienol (2375g) and stannous oxalate (31.2g). The mixture was sparged with nitrogen for 10 minutes to remove air and the niUOgen flow was then reduced to a level which ensured efficient mixing. The mixture was then heated in stages to 130°C ( e.g. 80°C for 10 minutes, then 100°C for 10 minutes and then 120°C for 10 minutes). The progress of the reaction was monitored by collection of the methanol collected in the Dean and Stark apparatus. In order to drive the reaction to completion the temperature was raised to 140°C after 7hrs at 130°C. When 90% of the methanol had been collected the reaction mixture was sampled hourly and analysed by GC. The reaction was adjudged complete when the level of the "half ester" (methyl octadienyl maleate/fumarate) fell to below 0.3% w/w and this took approximately 31 hours. At this point the heating was switched off and the reaction mixture allowed to cool to room temperature. The product from the reaction was then decanted from any solids in the reaction flask. This product was then charged to a heated decanter (40°C) with an equal volume of 5% w/w aqueous sodium hydroxide solution. The mixture was stirred for 20 minutes and then allowed to separate and the lower aqueous phase decanted. This base wash was repeated and the remaining organic phase was washed with saturated brine until the aqueous phase reached a steady pH. The organic phase was then heated (100°C) under reduced pressure (< 500 Pa (< 5 mBar)) on a rotatory evaporator to remove residual water and the majority of the excess octadienol. After cooling, the product was filtered and transferred to a 5-litre three-necked round-bottomed quickfit flask . This flask was equipped with a still-head condenser and reciever flask (Perkin triangle), a thermocouple, a steam inlet pipe, and a eurotherm controlled heating mantle. The apparatus was evacuated to 4000 Pa (40 mBar) and the product heated to 120°C. The supply of steam was then connected and the residual traces of octadienol were removed. The purification was judged complete when tlie volume of the heads product aqueous phase increased more than 5 times that of the organic phase. After cooling down the product was then treated with activated carbon (l%w/w, 100°C 2Hrs, < 500 Pa (< 5 mBar)) on a rotatory evaporator. The cooled mixture was filtered through dried celite to obtain tlie final product which had the following analyses: OH number - 4 mg KOH/g (titration) total acid - 46 ppm KOH/g (titration) maleic acid / anhydride - < lOppm (HPLC)

Fumaric acid - <10ppm (HPLC) tin - <10ppm (atomic absoφtion) sodium - <20ppm ( atomic absoφtion, detection limit) chlorine - <10ppm (atomic absoφtion detection limit)

GC "CPSil5" column - octadienyl methyl fumarate/maleate

(0.1 1 % w/w) di-(2,7-octadienyl) maleate (73% w/w) di-(2,7-octadienyl) fumarate (22% w/w) - 2-(2,7-octadienoxy) di-(2,7-octadienyl) succinate (3% w/w) The GC assignment was supported by GC/MS and a H nmr and '^" C nmr studies. The GC/MS used a VG Trio- 1000, operated according to the manufacturers instructions under the following conditions: - GC column 25mx0.32mm DB5 (0.25 micron film) temperature programme 40°C (3 mins) @ lOC/min to 320°C(10mins) injection 1 microlitre (1% solution in acetone) on column 40°C - ammonia chemical ionisation (CI) scan range 50-800 scan rate 1/s It was found that the deduction of molecular weights from the CI spectra is rather less straightforward than is usual on account of (a) extensive rearrangements of fumarates

^■4- 4- + in particular giving [M+3] and [M+20] ions in addition to the usual [M+l] and [M+l 8] ions and (b) extensive fragmentation exhibited by some species. As a result the GC peaks were assigned by inteφretation. In addition to the assigned peaks an additional species was identified which was assigned to a lactone. This assignment was

1 1 confirmed by H and C nmr. Table 1 gives assignments of the observed ¹³C nmr peaks. It should be noted that the two isomeric octadienols ( 2,7-octadienol and l,7-octadien-3-ol) though not separable by tlie GC method used can be identified by nmr and are recorded in the nmr assignment Table 1. The correspondence to the

GC was confirmed again by integration of the nmr spectrum of several samples in which the composition varied. The lactone found by GC/MS was also observed in the nmr and quantified at approximately 6.2% (tenative structure given in Table 1). the product of this Example will hereafter be called AK2R. Example S4: The process of Example S 1 was repeated to give a product with the following analysis : OH number - <1 mg KOH/g (titration) total acid - 91 ppm KOH/g (titration) maleic acid / anhydride - < lOppm (HPLC)

Fumaric acid - <10ppm (HPLC)

Zinc - oppm (atomic absoφtion, detection limit) sodium - < 20ppm ( atomic absoφtion) chlorine - <1 Oppm (atomic absoφtion detection limit)

GC "CPSil5" column - 2, 7-octadienyl methyl fumarate/ maleate

(0.02% w/w) di-(2,7-octadienyl) maleate (1.4% w/w) - di-(2,7-octadienyl) fumarate (55% w/w) 2-(2,7-octadienoxy) di-(2,7-octadienyl) succinate (42% w/w) higher boilers (<0.01% w/w) 1 1 ^

The GC assignment was supported by GC/MS and a Hnmr and C nmr studies. The product of this Example will hereafter be called sample AKIR.

TABLE 1

8 7 6 5 4 3 2 1

2.7-Octadienol: CH₂=CH.CH₂.CH₂.CH₂.CH=CH.CH₂.OH

1 2 3 4 5 6 7 8

Major 63 132.8 129.5 31.3 28.6 33 138.5 114.5

Minor 58 132 129 26.6 28 33 135.5 114.5

8 7 6 5 4 3 2 1 9 10

Fumarate/Maleate: CH₂=CH.CH₂.CH₂.CH₂.CH=CH.CH₂.O.(=O)C.C=C.C(=O).OR (R = 2,7-Octadienyl)

Fumarate/Maleate 1 2 3 4 5 6 7 8 9 10

Maleate (major isomer) 65.24 123.34 135.81 31.12 27.53 32.65 137.8 114.3 164.27 129.33

Maleate (minor 60.4 - - 26.39 28.04 32.65 - - 164.42 129.39 isomer)*

Fumarate (major 65.35 123.31 136.16 31.2 27.58 32.73 137.81 114.42 163.96 133.12 isomer)

Fumarate (minor 68.72 122.83 135.02 26.49 28.1 32.79 137.81 114.51 164.07 133.21 isomer)*

* Some of the assignments are tentative

TABLE 1 (continued)

2-(2.7-octadienoxy) di-2.7-octadienyl succinate:

8 7 6 5 4 3 2 1 9 10 1 1 12 1' <- 2' - 8' -» 1" 2" 3" 4" 6 5" 7" 8"

Lactones: Species believed to be lactones have been identified in the octadienyl ester samples. Whilst there appear to be two possible isomers of this lactone, the C NMR data fits the following structure:

O

/ \| 2

,CH₂ C-O \ / CH₂=CH.CH₂.CH₂.CH=CH.CH.CH.CH₂.C(=O).O.CH₂.CH=CH.CH₂.CH₂.CH₂.CH=CH₂ 8 7 6 5 4 3 2 11 10 9 1'

1 1' 2 3 9 10 1 1 12 1'

Isomer 1 71.28 41.47 125.33 170.49 30.45 39.47 176.43 64.94

Isomer 2 69.81 45.00 126.58 170.01 31.69 41.59 176.30 64.96

Example S5

Preparation of 2-7-octadienyl Fumarate

All operations were carryed out under a nitrogen atmosphere. The following apparatus was assembled : a 2 litre three-necked round bottomed flask equipped with condenser, water pump (with a backflow trap), stirrer, eurotherm-controUed heating mande, thermowell and a pressure-equalising dropping funnel. The dry apparatus was then charged with: To the flask :

500g (3.922 moles) of octadienol ( water level <0.05%w/w by Karl Fischer analysis) and

500mls cyclohexane (water level <0.05%w/w); To the pressure equalising dropping funnel

200g (1.3075 moles) fumaryl chloride. The system was then evacuated to 0.5 bar. The mixture heated to about 55C at which point a gentle reflux of cyclohexane commenced. The fumaryl chloride was then added dropwise to the stirred flask. Once the addition was completed the heating was continued for a further hour before allowing the mixture to cool room temperature.

The reaction mixture was then transfered to a heated separating funnel (40C) and washed twice with a aqueous 5% sodium hydroxide solution. The organic phase was then washed with saturated brine until the aqueous layer solution reached a steady pH.

The organic top layer was then concentrated on a rotatory evaporator, operated at (i) 200mBar and 80C, to remove cyclohexane and then (ii) 100C and O.lmBar to remove excess octadienol. The final product was a colourless oil. GC analysis confirmedthe product to be di-(2,7 octadienyl) fumarate. 2. Viscosity measurements of the reactive diluents

2.1 Method

The viscosity of each diluent was measured at 25°C using a suspended level viscometer. Densities of the diluents were taken as an average of three readings made at 25°C using density bottles with a nominal 10 cm^ capacity, calibrated with water. 2.2 Results

TABLE 2 : Viscosity of reactive diluents

Diluent Viscosity (m Pa.s)

AK1 (Example SI) 59

AK2 (Example S2) 19

AK2R (Example S3) 18

AKIR (Example S4) 51

Table 2 demonstrates that the esters / reactive diluents of this invention have relatively low viscosity (<200 mPa.s) and are suitable for use as reactive diluents. 3. Examples of the use of the reactive diluents in paint formulations

Reactive diluents, such as those of the present invention, must meet a range of criteria including low odour and low toxicity, low viscosity and the ability to "cut" the viscosity of the paint to facilitate application on the surface to be coated therewith. Furthermore, the diluent should not have a markedly adverse effect on the properties of the paint film such as drying speed, hardness and degree of wrinkling. The reactive diluents described above have therefore been tested in paint applications using both clear and pigmented paints. The diluents have been compared with paints formulated using white spirit, a conventional thinner. The results demonstrate the excellent performance of the diluents of this invention.

3.1 Tests of pigmented paint formulations NB. It is well known by those skilled in the ait that day-to-day fluctuations in conditions can introduce some variability into experimental data. To minimise these errors, the tests presented below were conducted as follows: ca. five to ten paint formulations were prepared simultaneously and comprised one reference (white spirit) and about four to nine reactive diluent-based paints. These samples were tested at the same time under identical conditions. Comparison of performance data from within these groups of foπnulations allowed errors due to random sources to be minimised. Hence in the following examples the reader will realise that the appai'ent variation in performance data from some diluents results from the use of different paint formulations made on different days from the same diluent. a) Methods used to prepare paint formulations NB. In the examples below, % reactive diluent refers to the ratio of reactive diluent to alkyd (e.g. 30% reactive diluent implies 30g diluent to every 70g alkyd). i Pigmented paint formulations based on a high solids alkvd resin

High solids alkvd reference with white spirit diluent In a mini motor mill 534 grams of a high solids alkyd (Setal® 293, ex. Akzo Nobel Resins) and 423 grams of titanium dioxide (Kronos® 2310, ex. Kronos) were milled. Thereafter 40 grams of a combi siccative (Nuodex Combi® APB, ex. Servo) and 2.9 grams of methyl ethyl ketoxime were added and mixed thoroughly. This mixture was diluted with white spirit to an application viscosity of 0.5 Pa.s. High solids alkyd with 30 % of reactive diluent

In a mini motor mill 374 grams of a high solids alkyd (Setal® 293, ex. Akzo Nobel Resins) and 423 grams of titanium dioxide (Kronos® 2310, ex. Kronos) were milled. Thereafter 160 grams of 'AK1 ' , 40 grams of a "combi siccative" (Nuodex Combi® APB, ex. Servo) and 2.9 grams of methyl ethyl ketoxime were added and mixed thoroughly. This mixture was diluted with white spirit to an application viscosity of 0.5 Pa.s. A similar method (with appropriate adjustments to the proportions of the components of the paint) was used to prepare paints with 10 and 20% reactive diluent. ii) Pigmented paint formulations based on a conventional alkvd resin Conventional alkyd refererence with white spirit diluent In a mini motor mill 563 grams of a conventional alkyd (Setal® 270WS70, ex. Akzo Nobel Resins) and 306 grams of titanium dioxide (Tioxide® TR92, ex. Tioxide) were milled. Thereafter 2.4 grams of Siccatol® Co (10%) (ex. Durham Chemicals), 8.5 grams of Siccatol® Sr (10%) (ex. Durham Chemicals), 11.5 grams of Siccatol®Ca (5%) (ex. Durham Chemicals) and 3.0 grams of methyl ethyl ketoxime were added and mixed thoroughly. This mixture was diluted with white spirit to an application viscosity of 0.5 Pa.s.

Conventional alkyd with 20 % of reactive diluent

In a mini motor mill 448 grams of a conventional alkyd (Setal® 270WS70, ex. Akzo Nobel Resins) and 306 grams of titanium dioxide (Tioxide® TR92, ex. Tioxide) were milled. Thereafter 80 grams of 'AK-1 ' , 2.4 grams of Siccatol® Co (10%) (ex. Durham Chemicals), 8.5 grams of Siccatol® Sr (10%) (ex. Durham Chemicals), 11.5 grams of Siccatol® Ca (5%) (ex. Durham Chemicals) and 3.0 grams of methyl ethyl ketoxime were added and mixed thoroughly. This mixture was diluted with white spirit to an application viscosity of 0.5 Pa.s.

Each of the pigmented paints prepared as described above was stored in two tins. One tin was kept at 23°C for 7 days before testing. The second tin was stored at 35°C for 2 weeks before drying speed tests were performed. Tlie results below derive from the tins stored at 23°C unless stated. b) Test methods used with pigmented paint formulations

Coating viscosity was measured in accordance with the ICI Cone & Plate method (ISO2884) at 10,000 s^"1, 23°C, 50% RH (Pa.s).

Konig hardness (ISO 1522) was measured at 23°C, 50% RH (s) of a coating film applied with a 100 μm applicator on a glass substrate.

The Fischer spherical indention test is based on ISO6441 (um) and was performed on a coating film applied with a 100 μm applicator on a glass substrate. Drying performance was determined at 10°C, 85% RH with a Beck-Koller drying recorder (hours) under daylight lamps. Films were applied to glass substrates using a 90 μm applicator. The measurements quoted (in hours) are : a) For high solids alkyd resin-based pigmented paints : "phase 1 ", the dust drying time; b) For conventional alkyd resin-based pigmented paints : "phase 2" (touch dry time) and "phase 3" (through-dry time). c) Results of the tests of pigmented paints based on reactive diluents i) Pigmented paints based on the high solids alkvd

Example Tl Paints containing 30% reactive diluent were prepared from the diluents AK1,

AK2, AK2R.

The drying speed results in Table 3 show that paints based on the reactive diluents of this invention dry over a period which is acceptable to the industry.

The Fischer indentation test and Konig hardness test results in Table 4 show that paint films containing the reactive diluents of this invention are relatively hard. This is a considerable advantage since many so-called reactive diluents cause plasticisation of tlie paint film. In contrast, the results in Table 4 are evidence that the diluents of this invention are bound into the paint film. TABLE 3 (Drying performance)

TABLE 4 (Hardness Data)

Reactive Diluent Used Indentation (microns) Konig hardness **

Ester Mix AK1 1.61 * 0.96# 23#

Ester Mix AK2 2.07* 1.30# 12@ 18#

Ester Mix AK2R 2.08* 1.26# 12@ 18#

* Film cured for 1 week at 23°C

# After 100 hours at 50°C @ After 3 weeks at 23°C

** In number of swings. Multiply by 1.4 to convert to seconds.

The results in Table 5 show the maximum thickness of paint which can be applied to a glass substrate without the appearance of unsightly wrinkles in the dried film. The reactive diluents of this invention allow relatively thick films to be applied - this is a considerable advantage for high solids and "one-coat" paints. The data in the table also show that very high solids contents can be achieved with the reactive diluents of this invention.

TABLE 5

Diluent Maximum film thickness* Solids content **

AK1 300 94

AK2 300 97

AK2R 300 97

* Expressed as bar coater gap width (microns) ** Solids content of paint in weight% Example T2

Further paint formulations were prepared using the high solids resin with different concentrations of diluent (expressed as % cf. the resin, i.e. 10% means 10% diluent to 90% resin). Table 6 shows the hardness data from these samples, again demonstrating that films with very good hardness can be obtained by using the diluents of this invention.

TABLE 6 (Hardness Data)

Diluent Used Indentation (microns) Konig hardness (**)

AK1 (10%) 1.37* 1.07# 14@ 18#

AK1 (20%) 1.24* 0.82# 16@ 24#

AK1 (30%) 1.19* 0.66# 20@ 32#

White spirit 1.29* 1.1# 13@ 16# * After 1 week curing at 23°C

# After 100 hours at 50°C @ After 3 weeks at 23°C

** In number of swings. Multiply by 1.4 to convert to seconds. ii) Pigmented paints based on the conventional alkvd It is possible to use the reactive diluents of the present invention in paints based on a wide range of resins. Further paint foπnulations were prepared using the conventional alkyd resin with different concentrations of diluent (expressed as % cf. the resin, i.e. 10% means 10% diluent to 90% resin). Example T3 The data in Table 7 show that a VOC content of below 300g/l can be achieved using the diluents of this invention at the 20% loading. Furthermore, the presence of the diluent in the paint had no effect (cf. the conventional white spirit-based paint) on the maximum film thickness which could be applied without the appearance of insightly wrinkles in the paint. TABLE 7

Diluent Maximum film thickness* VOC content**

AKIR 10% 200 323

AKIR 20% 200 288

White spirit 200 366 * Expressed as bar coater gap width (microns)

** Volatile organic compound content, grammes per litre.

Tables 8 and 9 summarise the drying and hardness test data from the pigmented paint formulations based on conventional alkyd resins. The drying data in Table 8 show that the paints based on the reactive diluents of this invention dry in a period acceptable to the industry, even after storage at 35°C for two weeks. Furthermore, the hardness of the paint films is excellent.

TABLE 8

* After paint storage for 1 week at 23°C ** After paint storage for 2 weeks at 35°C

TABLE 9 (Hardness Data)

Diluent Used Indentation (microns) Konig hardness **

AKIR (10%) 1.46* 0.83# 41@ 46#

AKIR (20%) 1.53* 0.69# 46@ 51#

White spirit 1.47* 0.91# 33@ 35#

* Film cured for 1 week at 23 °C

# After 100 hours at 50°C @ After 3 weeks at 23°C

** In number of swings. Multiply by 1.4 to convert to seconds.

3.2 Tests of clearcoats (un pigmented paints) a) Methods used to prepare unpigmented "clearcoat" formulations i) Materials Used:

Unpigmented ("clearcoat"; paint formulations were prepared using the high solids alkyd resin SETAL® 293 described above in Section 3.1(a)(i). In addition to the diluent, Siccatol® 938 drier (ex AKZO NOBEL) and methyl ethyl ketone-oxime (hereafter "MEK-oxime") anti-skinning agent were used. Where used, the white spirit was Exxon type 100. The nominal proportions of the above materials in the paint formulations were:

TABLE 10

Materials Parts by weight

Resin + Diluent 100.0

Siccatol 938 6.7

MEK-oxime 0.5

Note that, for white spirit foπnulations only, the proportions of drier and antiskinning agent were calculated on the basis of the resin only. Thus, the concentration of these components in the paint was lower than for other diluents. ii) Method of Preparation of Clearcoat Foπnulations Alkyd resin and diluent were mixed in glass jars for 2 hours (eg using a Luckham multi-mix roller bed) in the proportions required to achieve a viscosity (measured via the ICI cone and plate method using a viscometer supplied by Research Equipment (London) Limited) of 0.68 ± 0.03 Pa s (6.8 ± 0.3 poise). Typically, this resulted in a mixture which was ca. 80% w/w resin. If further additions of diluent or resin were required to adjust the viscosity to 0.68 ± 0.03 Pa s (6.8 ± 0.3 poise), a further hour of mixing was allowed. The required quantity of drier was added and, after mixing (1 hour), the required amount of anti-skinning agent was added. After final mixing for at least 30 minutes, the viscosity of the mixture was measured to ensure that the viscosity was between 0.61 and 0.69 Pa s (6.1 and 6.9 poise).

The mixture ("formulation") was then divided into two tins and sealed so as to leave ca. 10-15% v/v headspace of air in the sealed jars. One of the tins was stored at 23°C in darkness for 7 days before paint applications tests were perfonned. The second tin was stored ("aged") at 35°C in daylight for 14 days before applications tests were performed. b) Test methods used for Clearcoat Formulations: i) Application of paint film:

Thin films were applied to cleaned glass test plates using Sheen cube or draw¬ bar applicators with a nominal 75μm gap width. ii) Viscosity: The viscosity of each foπnulation was measured according to BS 3900 Part A7 with an ICI cone and plate viscometer (supplied by Research Equipment (London) Limited) at 23°C and at a shear rate of 10,000 reciprocal seconds. iii) Drying Performance: Drying perfonnance was measured using films applied to 30 cm x 2.5 cm glass strips and BK drying recorders. The BK recorders were enclosed in a Fisons controlled temperature and humidity cabinet so that the drying experiment could be perfonned at 10°C and at 70% relative humidity. Sample performance was assessed on the basis of the dust drying time, T2. iv) Incorportion of the diluent into the paint film

An indication that the reactive diluent is incorporated into the paint film during the "cure", rather than undergoing evaporation, was obtained by monitoring the weight of tlie paint film following application. Weight measurements were done on a coating film applied with a 150 μm applicator on a glass substrate at 23°C and 50%RH. The data are expressed as a % of the coated film weight, after the white spirit component of the siccative drier had evaporated (which took about 10 minutes).

NB. It is well known by those skilled in the art that day-to-day fluctuations in conditions can introduce some variability into experimental data. To minimise these errors, the tests presented below were conducted as follows: ca. five to ten paint formulations were prepared simultaneously and comprised one reference (white spirit) and four to nine reactive diluent-based paints. These samples were tested at the same time under identical conditions. Comparison of performance data from within these groups of formulations allowed errors due to random sources to be minimised. c) Drying Speed Data from clearcoats

The results in Tables 1 1 and 12 show that clearcoat paints based on the reactive diluents of the present invention reach the dust dry stage of drying within a period which is within about 4 hours of a traditional white spirit-based paint. These properties are regarded as satisfactory by the industry. TABLE 11

Diluent Used Drying times (hours)

Paint stored for 1 week Paint stored for 2 weeks at 23°C at 35°C

AK1 5.69 5.25

AK2 6.25 7.16

AK2R 6.16 7.00

White spirit 3.9 3.81

TABLE 12

Diluent Used Drying time (hours)

Paint stored for 1 week Paint stored for 2 weeks at 23°C at 35°C

AKIR 4.9 5.2

White spirit 3.3 3.3

Weight measurements of two coated films are summarised in Table 13, and show that the diluent remains in the paint film, rather than evaporating.

TABLE 13

Time (hours) Reference (High solids resin AK1 + high + white spirit) solids alkyd

0 100 100

0.5 97.99 97.28

1.0 98.31 97.34

5 102.22 100

24 100.64 99.41

72 98.91 99.31

195 97.28 100.03

Example T5 Some furtiier clearcoat foπnulations were prepared in using the di-octadienyl fumarate prepared in Example S5 (which had a viscosity of 9.6 mPa.s). The drying speed results are given in Table 14, and are again satisfactory by industry standards.

TABLE 14

Diluent Used Drying time (hours)

Paint stored for 1 week at 23°C

Formulation from di-octadienyl 5.4 fumarate (Example S5)

White spirit 3.5

Claims

Claims:

1. A formulation comprising a paint or a polymer and an ester product comrprising one or more esters selected from the group consisting of a. 2-(2,7-octadienoxy) di-2,7-octadienyl succinate, b. di-(2,7-octadienyl) fumarate and c. di-(2,7-octadienyl) maleate, said ester product having a viscosity of less than 200 mPa s (cP).

2. A formulation according to Claim 1 wherein the polymer in said formulation comprises alkyd resins.

3. A formulation according to Claim 1 or 2 wherein the ratio of the ester product to the polymer is in tlie range from 5-50 : 95-50 parts by weight.

4. A formulation according to any one of the preceding Claims wherein said ester product has a viscosity of the order of eg 10-80 mPa s (cP).

5. A formulation according to any one of the preceding Claims wherein said formulation contains one or more further components selected from the group consisting of catalyst, drier, antiskinning agent, pigments, pigment stabilizers and water scavengers.

6. A fonnulation according to any one of the preceding Claims wherein the ester mixture used as reactive diluent in such foπnulations has an acidity value of

< 7000 ppm w/w of KOH.

7. A foπnulation according to any one of the preceding Claims wherein the ester mixture used as reactive diluent in such foπnulations has an acidity value of

< 1000 ppm w/w of KOH.

8. A foπnulation according to any one of the preceding Claims wherein said foπnulation comprises in addition a polymerisation inhibitor.

9. A foπnulation according to Claim 8 wherein the inhibitor is 2,6-butoxy-4- methyl-phenol or 2,4,6-te/-t-butyl phenol. ^'