Process for the production of (meth)acrylic acid and derivatives and polymers produced therefrom

Abstract

A method of extracting (meth)acrylic acid from an aqueous reaction medium into an organic phase in contact therewith is described. The aqueous reaction medium is formed from at least one base catalyst and at least one dicarboxylic acid selected from maleic, fumaric, malic, itaconic, citraconic, mesaconic, and citramalic acid or mixtures thereof in aqueous solution and contains the base catalyzed decarboxylation products of the base catalyzed reaction. The method includes either the addition of at least one of the said dicarboxylic acids and/or a pre-cursor thereof to the aqueous reaction medium to enhance the solvent extraction of the (meth)acrylic acid into the organic solvent or maintaining the level of base catalyst to dicarboxylic acid and/or pre-cursor at a sub-stoichiometric level during the extraction process. The method extends to a process of producing (meth)acrylic acid, its esters and polymers and copolymers thereof.

Claims

1. A process for the continuous production of (meth)acrylic acid comprising the steps of: forming an aqueous medium of at least one base catalyst and at least one dicarboxylic acid and/or a precursor thereof, wherein the at least one dicarboxylic acid is selected from fumaric, maleic, malic, itaconic, citraconic, mesaconic or citramalic acid or mixtures thereof; decarboxylating the at least one dicarboxylic acid in the presence of the at least one base catalyst under suitable conditions of temperature and pressure to produce (meth)acrylic acid and/or base salts thereof in the aqueous medium; introducing an organic solvent to the said aqueous medium for solvent extraction of the (meth)acrylic acid into an organic phase; wherein the dicarboxylic acid and/or pre-cursor thereof forms a first acid salt and a second acid salt thereof in the presence of the base catalyst; characterised in that the level of base catalyst to the said at least one dicarboxylic acid and/or pre-cursor thereof is maintained at a sub-stoichiometric level in relation to the formation of the first acid salt of the at least one dicarboxylic acid and/or pre-cursor thereof during the extraction process.

2. The process according to claim 1, wherein the concentration of (meth)acrylic acid in the aqueous phase extraction is at least 0.05 mol dm.sup.3.

3. The process according to claim 1, further comprising the additional step of adding an additional amount of at least one of the dicarboxylic acids and/or a pre-cursor thereof to the aqueous reaction medium to enhance the solvent extraction of the (meth)acrylic acid into the organic phase.

4. The process according to claim 1, wherein the dicarboxylic acid and/or a pre-cursor thereof is selected from a group consisting of citric, itaconic, citramalic, citraconic and mesaconic acid or mixtures thereof.

5. The process according to claim 4, wherein the dicarboxylic acid and/or a pre-cursor thereof is selected from a group consisting of citric, itaconic, citramalic and citraconic acid or mixtures thereof.

6. The process according to claim 1, wherein the dicarboxylic acid is selected from a group consisting of maleic, fumaric, and malic acid or mixtures thereof.

7. The process according to claim 1, wherein the dicarboxylic acid is malic acid.

8. The process according to claim 1, wherein in the case of the (meth)acrylic acid being methacrylic acid, the organic solvent is an external organic solvent with respect to the reaction medium.

9. The process according to claim 1, wherein the dicarboxylic acid is selected from citramalic or itaconic acid.

10. The process according to claim 1, wherein the organic solvent for (meth)acrylic acid extraction is selected from hydrocarbon solvents or oxygenated solvents.

11. The process according to claim 10, wherein the hydrocarbon solvents are C.sub.4-C.sub.20 hydrocarbon solvents.

12. The process according to claim 10, wherein the organic solvent include toluene, benzene, ethylbenzene, xylene, trimethylbenzene, octane, heptane, hexane, pentane, cyclopentane, cyclohexane, cycloheptane, cyclooctane, cyclohexene, methylcyclohexane, methylethylketone, methyl methacrylate, mixtures thereof; and ionic liquids which are immiscible with water.

13. The process according to claim 1, wherein the mixture of solvents for the extraction of (meth)acrylic acid is a C.sub.4-C.sub.20 hydrocarbon solvent and methyl methacrylate.

14. The process according to claim 1, further comprising the step of separating the organic phase from the aqueous phase after extraction followed by subsequent treatment of the organic phase to isolate the (meth)acrylic acid extracted in the extraction process from the organic solvent.

15. The process according to claim 1, wherein the organic solvent is introduced to the aqueous medium before or after decarboxylation.

16. The process according to claim 1, wherein the sub-stoichiometric level of base is maintained, after, if necessary, being implemented post reaction, during at least that part of the extraction process which is carried out after the decarboxylation step.

17. The process according to claim 1, wherein the sub-stoichiometric level of base is maintained throughout the reaction and extraction.

18. A method of preparing polymers or copolymers of (meth)acrylic acid or (meth)acrylic acid esters, comprising the steps of: (i) preparation of (meth)acrylic acid in accordance with claim 1; (ii) optional esterification of the (meth)acrylic acid prepared in (i) to produce the (meth)acrylic acid ester; (iii) polymerisation of the (meth)acrylic acid prepared in (i) and/or the ester prepared in (ii), optionally with one or more comonomers, to produce polymers or copolymers thereof.

19. Polyacrylic acid, polymethacrylic acid, polyalkylacrylate, polymethylmethacrylate (PMMA) and polybutylmethacrylate homopolymers or copolymers formed from the method of claim 18.

20. The process for the production of methacrylic acid comprising: providing a source of a pre-cursor acid selected from aconitic, citric and/or isocitric acid; performing a decarboxylation and, if necessary, a dehydration step on the source of pre-cursor acid by exposing the source thereof in the presence or absence of a base catalyst to a sufficiently high temperature to provide a dicarboxylic acid selected from itaconic, mesaconic, citraconic and/or citramalic acid; and using the dicarboxylic acid produced in a process according to claim 1.

Description

(1) For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example, to the following figures and examples in which:

(2) FIG. 1 shows the concentration dependence of the extraction of MAA into toluene;

(3) FIG. 2 shows a plot of partition coefficient for a range of acids against MMA fraction in toluene;

(4) FIG. 3 shows a plot of relative partition coefficient for a range of acids with MMA against MMA fraction in toluene;

(5) FIG. 4 shows the effect of adding base and dicarboxylic acid on transfer of MAA between aqueous and organic phases;

(6) FIG. 5 shows the distribution of acrylic acid between water and toluene;

(7) FIG. 6 shows a schematic view of suitable apparatus for the base catalysed decomposition of dicarboxylic acids.

SOLVENT EXTRACTION

(8) The following experimental conditions were used unless indicated otherwise: 0.1M Acids 1:1 vol:vol aq:solvent Room Temperature 1 minutes agitation time; 5 min settling time Solvent is toluene unless where stated Analysis by HPLC

Comparative Example 1

(9) A series of solvents were tested to examine the extent of transfer of methacrylic acid from an aqueous solution using the above procedure. The results are shown in table 1.

(10) TABLE-US-00001 TABLE 1 Average % relative (static) Solvent Transfer permittivity Mixed Xylenes 45.3 2.3 Toluene 48.2 2.4 Hexane 27.6 1.9 Benzene 50.1 2.3 Pentane 28.3 1.8 Cyclohexane 26.9 2.0 MMA 84.3 6.3

(11) This example shows that MAA present in the free acid form can be efficiently extracted into a range of solvents. Aromatic hydrocarbons give the highest extraction efficiencies.

Comparative Example 2

(12) Monobasic and dibasic acids likely to be present in aqueous solution following partial decomposition of mono and dicarboxylic acids expected to be found from decomposition of dibasic or tribasic acids were compared for their solubility in toluene.

(13) Each acid, initially at 0.1M solution in water was separately tested for solubility in an equal volume of toluene. The results are shown in table 2

(14) TABLE-US-00002 TABLE 2 Fraction Transferred to Acid Toluene/% monobasic MAA 54.4 CT 40.11 HIB 4.21 PY 0 dibasic IC 0 MC 0.64 MAA Methacrylic Acid CT Crotonic Acid HIB Hydroxyisobutyric Acid PY Pyruvic Acid IC Itaconic Acid MC Mesaconic Acid

(15) This example shows that the di and tricarboxylic acids useful in the process for the production of MAA are not soluble in toluene, one solvent which can be employed for the extraction of MAA. Furthermore HIB formed in equilibrium with MAA is not extracted in significant proportions and pyruvic acid formed as an unwanted by-product is also not extracted into toluene.

Comparative Example 3

(16) A series of different concentrations of MAA in aqueous solution were extracted into toluene (1:1 by volume vs aqueous solution). The percentage solubility is shown in table 3.

(17) TABLE-US-00003 TABLE 3 [MAA] in start- % extracted at 1:1 ing aq soln/M toluene to aq soln Comp Ex 3a 0.00743 12.69% Comp Ex 3b 0.0148 20.07% Comp Ex 3c 0.02878 26.76% Comp Ex 3d 0.05829 37.09% Comp Ex 3e 0.1215 52.00% Comp Ex 3f 0.2479 60.51% Comp Ex 3g 0.3 63.60% Comp Ex 3h 0.4778 68.67% Comp Ex 3j 0.7559 73.72% Comp Ex 3k 0.9576 76.71%

(18) The fraction transferred increases with the concentration of MAA. The data from table 3 were plotted according to the equation:
[MAA].sub.tol=K[MAA].sup.2aq
and the value K in the equation was evaluated as 14.6. The results are plotted in FIG. 1.

(19) This example shows that the extraction of MAA into toluene is concentration dependent. For efficient extraction, concentrations above 0.1M are preferred.

Comparative Example 4

(20) Aqueous solutions were prepared of each of the dicarboxylic acids exemplified in comparative example 2. These were extracted with an equal volume of solvent mixtures of toluene and methyl methacrylate (MMA). The resultant degrees of extraction are shown in table 4

(21) TABLE-US-00004 TABLE 4 Fraction of MMA in MMA/Toluene solvent mixture IC MC PY MAA HIB CT 0 0 0.64 0 54.4 4.21 40.11 0.1 0 1.72 0 58.85 4.8 46.72 0.2 0.29 4.5 0.3 63.01 5.14 49.88 0.3 0.81 8.26 0.7 67.25 6.38 53.62 0.4 1.69 13.02 1.17 70.31 4.82 56.56 0.5 2.89 20.56 2.07 74.28 5.76 61.15 0.6 4.34 27.82 3.01 76.77 7.32 64.67 0.7 6.56 38.06 4.17 79.42 19.71 68.07 0.8 9.57 47.19 5.57 81.42 21.47 70.86 0.9 13.1 56.33 8.05 83.02 23.32 73.21 1 17.58 63.45 10.71 84.28 23.9 75.05

(22) This example shows that MMA can be added to toluene to improve the extraction efficiency of MAA. However an optimum MMA level is observed above which dicarboxylic acids and HIB are extracted in significant amounts.

(23) In order to compare the solubilities in the organic solvents in terms of partition coefficients each sample was converted to a partition coefficient based on the equation:
[MAA].sub.solv=K[MAA].sup.2.sub.aq

(24) The data are presented in FIG. 2

(25) Only MAA, Crotonic acid and hydroxyisobutyric acid have significant solubilities in any of the solvent phases. The solubility of the components increases with the fraction of MMA in every case.

(26) The relative partition coefficients may also change with composition. FIG. 3 compares the ratio of Partition Coefficient for MAA with that for each of the other acids.

(27) Thus the comparative examples show that selectivity is higher if pure toluene is used. However use of some MMA allows a higher concentration of MAA to be extracted whilst lowering selectivity.

Comparative Example 5

(28) The extraction of a solution of 0.1M MAA in aqueous solution into an equivalent volume of toluene was determined after addition of 0.05M sodium hydroxide. The amount of MAA transferred fell from 48% to 26%. The results are shown in the first two rows of table 5

Examples 1-3

(29) Sufficient itaconic acid to give a 0.1M solution was added to the MAA+sodium hydroxide containing aqueous solution of comparative example 5 and the MAA transfer dramatically improved to 44.7% extraction into toluene. The data are shown in table 5. The experiment was repeated with citraconic or mesaconic acids instead of Itaconic acid. Very similar results were obtained.

(30) TABLE-US-00005 TABLE 5 Concentration % of MAA in Transfer aqueous Added into solution/M NaOH/M Added Acid/M Toluene Comp Ex 1 0.1 48.0 Comp Ex 5 0.1 0.05 26.0 Ex 1 0.1 0.05 Itaconic Acid, 0.1 44.7 Ex 2 0.1 0.05 Citraconic Acid, 0.1 48.1 Ex 3 0.1 0.05 Mesaconic Acid, 0.1 46.3

Examples 4-30 and Comparative Examples 6-9

(31) 0.1M concentrations of various di and tricarboxylic acids added to an aqueous solution of 0.1M MAA containing different levels of NaOH were extracted with an equal volume of toluene.

(32) The quantity of MAA extracted fell much more slowly as sodium hydroxide concentration increased, in the presence of one of the added carboxylic acids than in the absence of added di/tri carboxylic acid. The effect was most marked with citric and mesaconic acids. Table 6 shows the experimental data, which are presented graphically in FIG. 4.

(33) TABLE-US-00006 TABLE 6 [MAA]/M [NaOH]/M Acid [Acid] % transfer Comp Ex 1 0.1 0 None 48 Comp Ex 5 0.1 0.05 None 26.04 Comp Ex 6 0.1 0 Itaconic 0.1 47.99 Ex 4 0.1 0.025 Itaconic 0.1 44.59 Ex 5 0.1 0.05 Itaconic 0.1 41.53 Ex 6 0.1 0.075 Itaconic 0.1 30.7 Ex 7 0.1 0.1 Itaconic 0.1 20.88 Ex 8 0.1 0.125 Itaconic 0.1 17.68 Ex 9 0.1 0.15 Itaconic 0.1 3.84 Comp ex 7 0.1 0 Citraconic 0.1 47.58 Ex 10 0.1 0.025 Citraconic 0.1 47.71 Ex 11 0.1 0.05 Citraconic 0.1 48.06 Ex 12 0.1 0.075 Citraconic 0.1 47.29 Ex 13 0.1 0.1 Citraconic 0.1 45.52 Ex 14 0.1 0.125 Citraconic 0.1 35.05 Ex 15 0.1 0.15 Citraconic 0.1 24.21 Ex 16 0.1 0.2 Citraconic 0.1 8.12 Comp Ex 8 0.1 0 Mesaconic 0.1 47.36 Ex 17 0.1 0.025 Mesaconic 0.1 46.98 Ex 18 0.1 0.05 Mesaconic 0.1 46.32 Ex 19 0.1 0.075 Mesaconic 0.1 45.66 Ex 20 0.1 0.1 Mesaconic 0.1 44.05 Ex 21 0.1 0.125 Mesaconic 0.1 39.16 Ex 22 0.1 0.15 Mesaconic 0.1 35.15 Ex 23 0.1 0.2 Mesaconic 0.1 23 Comp Ex 9 0.1 0 Citric 0.1 47.82 Ex 24 0.1 0.025 Citric 0.1 48.27 Ex 25 0.1 0.05 Citric 0.1 48.12 Ex 26 0.1 0.075 Citric 0.1 47.44 Ex 27 0.1 0.1 Citric 0.1 46.18 Ex 28 0.1 0.125 Citric 0.1 41.83 Ex 29 0.1 0.15 Citric 0.1 39.19 Ex 30 0.1 0.2 Citric 0.1 28.35

Examples 31-34

(34) Table 7 illustrates the use of higher organic phase to aqueous phase ratios leading to higher degrees of extraction of a solution of 0.3M MAA.

(35) TABLE-US-00007 TABLE 7 aq:toluene v/v % transfer Ex 31 1:1 64 Ex 32 1:2 72 Ex 33 1:3 76 Ex 34 1:4 85

Examples 35-39

(36) Table 8 further shows that the use of serial extractions can increase the MAA transfer still further. The starting solution was 0.3M MAA in water.

(37) TABLE-US-00008 TABLE 8 aq:toluene v/v % transfer 1:1 vol Ex 31 1:1 63.6 1:2 vol Ex 32 1:2 72.0 Ex 35 2 1:1 80.2 1:3 vol Ex 33 1:3 75.9 Ex 36 1:2 + 1:1 84.9 Ex 37 3 1:1 88.1 1:4 vol Ex 34 1:4 84.9 Ex 38 2 1:2 88.0 Ex 39 4 1:1 92.4

Example 40

(38) In a further experiment 0.01M citramalic acid decomposition was conducted with reaction flow in order to test the use of toluene extraction during the reaction; in this experiment, the flow of aqueous solution of dicarboxylic acid was mixed with an equal rate of flow of toluene before entering the reactor. Conditions were as follows: 0.01M Citramalic acid (CM) in water with 50 mM NaOH, 2000 psi at variable temperature, with a fixed residence time of 480 seconds. Initial flow consisted of CM and NaOH dissolved in water and toluene in a 50:50 ratio by volume. The yields of products in the two phases detected by HPLC analysis are displayed in table 9. Analysis of the organic phase indicated an absolute MAA yield of 3.42%, with no other products detected. The yield of MAA detected in the aqueous phase was 34.61%, therefore the partition coefficient for MAA between the toluene and aqueous phases=28.5 after cooling to ambient temperature. Thus the solvent may be added to the aqueous phase before the decomposition period as well as after cooling.

(39) TABLE-US-00009 TABLE 9 Detected in Detected in Aqueous Phase Toluene Phase Mass Balance 54.83 0.00 Conversion 93.25 0.00 PY 3.62 0.00 CC 4.53 0.00 IC 0.76 0.00 HIB 3.85 0.00 CM 0.00 0.00 MC 0.71 0.00 MAA 34.61 3.42 Key: - IC Itaconic Acid MC Mesaconic Acid CC Citraconic Acid HIB Hydroxyisobutyric Acid PY Pyruvic Acid

Examples 41-46 and Comp Ex 10

(40) Solutions of a mixture of dibasic acids and methacrylic acid were prepared in water containing 0.1M of each acid. Sodium hydroxide was added to each solution at a different concentration as shown in table 10. The aqueous solution was extracted with an equal volume of toluene at room temperature. The quantity in the organic and aqueous layers are shown in the table.

(41) TABLE-US-00010 TABLE 10 water toluene [NaOH] [MAA] [CC] [IC] [MC] [MAA] [MAA] Comp 0 0.1 0.1 0.1 0.1 0.052 0.048 Ex 10 Ex 41 0.025 0.1 0.1 0.1 0.1 0.048 0.052 Ex 42 0.05 0.1 0.1 0.1 0.1 0.050 0.050 Ex 43 0.075 0.1 0.1 0.1 0.1 0.052 0.048 Ex 44 0.1 0.1 0.1 0.1 0.1 0.051 0.049 Ex 45 0.125 0.1 0.1 0.1 0.1 0.050 0.050 Ex 46 0.15 0.1 0.1 0.1 0.1 0.051 0.049

(42) In the presence of 0.3M of combined dicarboxylic acids, the addition of base has no effect on the concentration of MAA extracted. In fact, by comparison with data in example 5, and table 5, it is obvious that the amount extracted was the same as for a solution free of dicarboxylic acid and base. This shows the effectiveness of the presence of the dicarboxylic acid in preventing the loss of organic solvent solubility in the presence of base.

Comparative Example 11

(43) Solutions of acrylic acid in water were extracted with toluene under the same conditions as in comparative example 3 except that the acid was changed from MAA to AA.

(44) The starting concentrations and the quantity extracted into toluene are shown in table 11.

(45) TABLE-US-00011 TABLE 11 Conc/M [organic]/M [aq]/M) Comp Ex 11a 1 0.20 0.80 Comp Ex 11b 0.75 0.12 0.63 Comp Ex 11c 0.5 0.064 0.44 Comp Ex 11d 0.25 0.026 0.22 Comp Ex 11e 0.125 0.0070 0.12 Comp Ex 11f 0.0625 0.0025 0.060 Comp Ex 11g 0.0312 0.00098 0.030 Comp Ex 11h 0.0156 0.00052 0.015 Comp Ex 11j 0.0078 0.00021 0.0076

(46) The relative concentration between the aqueous and organic phases is plotted according to the equation
[AA.sub.tol]=K[AA.sub.aq].sup.2
and shown in FIG. 5.

(47) The excellent straight line fit has a much lower slope than for example 3, indicating that AA much prefers the aqueous layer.

Comparative Example 12

(48) In order to increase the solubility of the AA in the organic layer a higher polarity is likely to be required. The extraction of a 0.1M aq AA solution was studied with an equal volume of a mixture between toluene and butanone.

(49) TABLE-US-00012 % extracted % extracted % Butanone Maleic acid Acrylic acid 0 0 5.01 10 0.32 14.57 20 1.46 25.26 30 3.41 35.45 40 5.19 44.14 50 10.62 53.47 60 10.77 57.31 70 15.01 63.39 80 19.88 67.47 90 27.09 70.04 100 34.32 65.56

(50) There is a very large increase in the extent of extraction as the butanone concentration increases, although the selectivity of extraction falls. It is likely that a mixture containing sodium salts will show a much improved separation between acrylic acid solubility and maleic acid solubility and that an appropriate choice of solvent of intermediate polarity will allow sufficiently effective a separation that the acrylic acid can be further purified by e.g. distillation.

Preparative ExamplesExperiments Conducted Using the Flow Reaction Use the Procedure as Outlined Below

(51) Flow Reaction Procedure

(52) A reactant feed solution was prepared comprising itaconic, citraconic, mesaconic acid or citramalic acid at a concentration of 0.5 M and sodium hydroxide also at a concentration of 0.5 M. The itaconic acid used (>=99%) was obtained from Sigma Aldrich (Catalogue number: L2,920-4); citraconic acid (98+%) was obtained from Alfa Aesar (L044178); mesaconic acid (99%) was obtained from Sigma Aldrich (Catalogue number: 13,104-0). The citramalic acid solution is prepared by dissolving solid (R)-()-citramalic acid (commercially available from VWR International) with sodium hydroxide catalyst in nano-pure water to the required concentration.

(53) The deionised water used for solvation of the acids/NaOH was first degassed via sonication in an Ultrasound Bath (30 KHz) for a period of 5 minutes.

(54) This reactant feed solution was fed into the reactor system via a Gilson 305 HPLC pump module fitted with a Gilson 10 SC pump head. The rate at which the reactant feed solution was pumped into the reactor system depended on the residence time required and the volume of the reactor. The feed rate was also dependent on the density of the reaction media which in turn depended on the reaction temperature.

(55) The reactant feed solution was pumped to the reactor via 1/16 internal diameter stainless steel (SS 316) pipe (Sandvik). The reactor consisted of a straight section of SS 316 pipe, encased in an aluminium block fitted with two 800W Watlow heater cartridges. The transition of the SS 316 piping from 1/16 to was achieved with Swagelok SS 316 reducing unions and required an intermediate step of pipe (i.e. 1/16 pipe to pipe to pipe).

(56) The volume of the reactor was calculated theoretically, and confirmed from the difference in weight when the reactor was filled with water and when it was dry; for the experiments described, the volume of the reactor was 19.4 cm.sup.3. After the pipe reactor, the piping was reduced back down to 1/16, before meeting a Swagelok SS 316 1/16 cross-piece. At this cross-piece, a thermocouple (type K) was used to monitor the temperature of the exit feed.

(57) Reactor volume (used for residence time) is defined as the volume of the section of pipe between the two to reducers located immediately before and after the aluminium block.

(58) The product mixture is finally passed through a heat exchanger (a length of pipe within a pipe through which cold water was passed in contra flow) and a manual Tescom Back-Pressure Regulator through which back-pressure (pressure throughout the whole system between this point and the pump head) was generated: the pressure employed was 3000 psi for all experiments described. Samples were collected in vials before being prepared for analysis.

(59) The required temperature for reaction was achieved using a thermostat fitted with a Gefran controller (800 P), which mediated power applied to the two Watlow cartridge heaters. Each set of experiments involved working at a single temperature while varying residence time between runs. The required flow rate for the first run was set at the Gilson pump module. The pump was then left for a period of around 20 minutes, pumping only deionised water, in order for the heat-transfer between the aluminium block to have become consistent. The heat-transfer was deemed to have achieved equilibrium when the temperature indicated by the thermocouple located at the reactor exit feed position did not change (accurate to 1 C.) for a period of more than 5 minutes. At this stage the inlet of the pump was transferred from the container of deionised water to the container of the prepared reactant mixture. The total volume of the apparatus (including reactor) was approximately double that of the reactor itself; this was previously determined experimentally. For a particular flow rate, the reactant mixture was left pumping for approximately three times the required period for it to have begun emerging from the final outlet, in order to ensure that a steady-state of reaction had been achieved. After this time a 20 ml sample of the apparatus exit solution was collected for analysis. Both the rate of collection of the exit solution and the rate at which the reaction solution was consumed were recorded against time in order to monitor the consistency of the pump efficiency. Following sample collection from a particular run, the pump inlet was switched back to the container of deionised water, and the flow rate was increased to its maximum for a period of approximately 10 minutes to ensure that all remaining material from the previous run had been purged from the system. This procedure was then repeated for the subsequent residence time to be investigated.

(60) Analysis

(61) Quantitative analysis of products was achieved using an Agilent 1200 series HPLC system equipped with a multi wave-length UV detector. Products were separated using a Phenomenex Rezex RHM monosaccharide H+(8%) column held at 75 C., protected by a guard column. The method used was isocratic, implementing a 0.4 mlmin.sup.1 flow rate of aqueous 0.005 M H.sub.2SO.sub.4 mobile phase. The compounds contained in product samples were found to have optimum UV absorbance at the shortest wavelength capable of the MWD detector of 210 nm (bandwidth 15 nm). All product compounds were calibrated for their UV detection, by correlating their UV absorbance against a range of concentrations. Linear response ranges were determined for each compound, and the most compatible range of concentrations found for all compounds of interest was between 510.sup.3 M and 110.sup.3 M. Thus, adequate quantitative detection of most products was achieved with a 1 to 100 dilution of samples obtained from the apparatus before HPLC analysis (a dilution of 1 to 100 would mean that when starting with a 0.5 M reaction solution, any product generated in a yield of between 20%-100% would fall within the linear response range of concentrations). Where compounds fell outside this linear response range (e.g. a yield of less than 20%), a second HPLC analysis was conducted using a dilution of 1 to 10. Any samples which were not accurately quantified using the 1 to 10 dilution method were considered to be trace in concentration and therefore negligible.

(62) Procedure

(63) The following procedure was carried out. The reagent mixture comprising acid and sodium hydroxide was first prepared. The required flow rate to achieve the residence time was calculated using the reactor volume and the density of water (calculated from temperature).

(64) FIG. 6 shows a schematic representation of the apparatus for the present invention. Reaction solution 18 was located in receptacle 20 which was connected to inlet 16. The inlet was connected via conduit 22 to the reactant pump 2 which was operable to pump the solution 18 to the reactor tube 24 tube which was housed in a heater cartridge 26 which extended circumferentially along the reactor 24 length. The conduit 22 between the pump 2 and the reactor 24 proceeded from the pump via a valve 28 for operation control, pressure monitor 30 and pressure relief valve 32. In addition, a trip switch 34 was connected to the pressure monitor 30, reactant pump 2 and a temperature monitor 14. The temperature monitor 14 was located in conduit 22 immediately after reactor 24 and before outlet 6. In addition, after the monitor 14, the conduit proceeded to the outlet via a filter 36, heat exchanger 8 and back pressure regulator 4. At the outlet 6, the product was collected in collection receptacle 38.

(65) The reactor 24 also included a temperature control unit 10, 12 to control the temperature of the reactor 24. The apparatus also included a quenching system which includes a separate inlet 40 for quench water 44 in quench water receptacle 42. The inlet 40 was connected to the outlet 6 via conduit 46 which included a separate quench pump 48 followed by a valve 50 for control of the quench water. The quench water conduit 46 met the reaction conduit 22 immediately after the temperature monitor 14 of the reactor 24 and before filter 36 to quench any reaction after the reactor. The quench pump 48 and temperature controller unit 10, 12 were also connected to trip switch 34 for necessary shut down when the trip criteria are met.

(66) The reactor pump 2 was turned on and deionised water was pumped into the system. The back pressure regulator 4 was gradually adjusted to the required pressure (3000 psi).

(67) The pump operation efficiency was checked at 5 ml min.sup.1 by recording time taken to collect a volume of 20 ml of water from system outlet 6. >90% efficiency was acceptable.

(68) The pump flow rate is then set to that required for the run.

(69) The water supply (not shown) to the heat exchanger 8 was set to a low-moderate flow, depending on the reaction temperature and pump flow rate for the experiment.

(70) The heater thermostat 10 fitted with a temperature controller 12 was set to the required temperature for the run.

(71) Once the required temperature had been reached (as indicated by thermostat 10), reactor outlet temperature was monitored by the reactor temperature monitor 14 until the value (accurate to 1 C.) was observed to remain static for a period of at least 5 minutes (this usually took approximately 20 minutes).

(72) The pump inlet 16 was switched from the deionised water container (not shown) to the prepared reagent mixture container 18 (this requires stopping the pump flow for a few seconds). The initial volume of reagent mixture in container 18 was recorded.

(73) Calculations can indicate the period before product solution will begin to emerge from the system outlet 6. However, in practice, this was confirmed by the visual and audible presence of gas bubbles exiting the apparatus (generated from the decomposition of reagents). This was allowed to continue for a period that is 3 the period taken for the product solution to emerge. This ensured that the product mixture is homogenous.

(74) At the outlet 6, 20 ml of product solution was collected and the time taken for this collection was recorded. A final time and volume reading was also taken for the reagent mixture.

(75) After product collection, the pump inlet was transferred back to the deionised water container, and the pump was set to prime mode (maximum flow rate) and left for a period of approximately 10 minutes.

(76) The flow rate of the pump was then set to the required value for the subsequent run.

(77) Again the reactor outlet temperature was monitored and was considered steady when the value did not change for a period of at least 5 minutes (this usually took approximately 10 minutes).

(78) This experimental method was repeated until all required runs for the experiment had been performed.

(79) After all runs had been completed, the deionised water was pumped into the system with the pump on prime mode and the heater (thermostat) was switched off.

(80) When the reactor outlet temperature had dropped below 80 C., the pump was switched off and the water supply to the heat exchanger was also ceased.

(81) Methacrylic Acid Extraction

(82) Solutions prepared according to the preparative procedure above were extracted with an equal volume of toluene. In the first set of experiments no extra acid was added. In the second set the acid used for the original high temperature decomposition was added such that the total concentration of dicarboxylic acids (Itaconic, citraconic, mesaconic, Citramalic) plus 2-hydroxyisobutyric acid equalled 0.5M, which was the starting concentration for the original decomposition. The results in table 10 show that addition of acid has a very large impact on the amount extracted at the high concentrations of base present.

(83) TABLE-US-00013 TABLE 10 Example Example Example Example Example 47 48 Example 49 Example 50 51 52 53 Feed IC IC IC IC IC MC CC Original 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Feed conc/M MAA 19.25% 64.73% 58.36% 56.74% 54.42% 44.89% 44.93% ICA 16.35% 0.99% 0.84% 0.00% 0.16% 7.72% 5.88% Citramalic 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% CCA 36.76% 1.69% 0.50% 0.00% 0.26% 16.27% 12.40% MCA 15.18% 2.08% 0.64% 0.08% 0.30% 13.28% 9.93% HIB 11.26% 23.04% 22.12% 19.33% 13.07% 13.72% 14.25% PY 0.36% 3.06% 2.69% 2.63% 2.67% 1.31% 1.77% CT 0.07% 0.91% 0.74% 0.53% 0.63% 0.63% 0.65% Acids Mass 99.23% 96.50% 85.89% 79.31% 71.51% 97.82% 89.81% Balance No added Acid % Extracted 11.55% 0.05% 1.00% 0.00% 0.00% 7.02% 2.01% pH 4.87 6.65 >7 >8 >8 5.34 5.70 Acid Added % Extracted 20.21% 29.43% 28.31% 28.04% 27.90% 30.56% 29.74% pH 4.39 4.45 4.47 4.47 4.46 4.05 4.16

Comparative Example 12

(84) The efficiency of MAA extraction into a mixture of 2-butanone and o-xylene in the ratio 75:25 was studied. The presence of xylene in this organic mixture partly restricts the solubility of butanone in the aqueous phase, which is a significant issue where butanone is used alone as the organic phase; at this particular ratio, the distribution coefficient for MAA is reported to be a maximum of approximately K=7.00..sup.23 In this case it was found that roughly 80% of MAA was extracted into the organic phase, which appeared extremely desirable; however, other dicarboxylic acids concerned in the decomposition experiments (i.e. IC, CC etc.) also showed a slight affinity to the organic phase of up to 11%.

(85) Attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

(86) All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.

(87) Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

(88) The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.