A number of systems are known already that are capable of oxidising aldehydes to carboxylic acids or other reaction products under relatively mild oxidising conditions. Thus, for example, oxidants such as permanganate or dichro ate may be employed, but they inevitably introduce toxic materials, thereby creating waste disposal problems. Alternatively, rather more exotic, and hence costly oxidising agents have been proposed such as complexes of chromium with oxygen, chloride and either 1 ,10-phenanthroline or 2,2'-bipyridyl in a paper by Chakraborty, T K et al in Synthesis Communications, JjD (12), 951 (1980) and benzyltriethylammonium permanganate in a paper by Scholtz, D, Monatsh. Chem. 110(6), 1471 (1979). The use of such latter compounds is often of interest in producing laboratory scale quantities of material, but become decreasing viable at increased scale. It will be recognised that for one or more reasons, such oxidants do not meet the criteria for commercial scale oxidation of aldehydes to a satisfactory extent. It is an object of the present invention to provide a

process for the oxidation of aromatic aldehydes that employs widely available and relatively cheap reagents and which can be carried out under relatively mild conditions.

^■ According to the present invention, there is provided a process for the oxidation of aromatic aldehydes in which an aldehyde obeying the general formula:-

in which n represents an interger from 1 to 3, and the or each X represents hydrogen or an electron-withdrawing substituent or a mildly electron-donating substituent that is ortho, meta or para to the aldehyde substituent or a strongly electron-donating substituent that is meta to the aldehyde substituent, provided that at least one ortho position is occupied by hydrogen, is brought into contact with an alkali metal perborate and acetic acid at a mild temperature and permitted to react until at least a proportion of the aldehyde has been oxidised to the corresponding aromatic carboxylic acid.

It is recognised that a number of qualitative terms have been employed herein. In the context of the extent of electron-donation, "mildly" indicates donation that is similar to or less than that of a methyl group, whereas "strongly" indicates a greater extent of electron donation, exemplified by a methoxy group.

The substituent X, it will be recognised, conveniently can be chosen from a very wide range of substituents, thereby demonstrating the inherent usefulness and widespread applicability of the instant invention process. In particular, it can be chosen from halo, aldehyde, nitrilo, carboxylic acid, nitro, thioalkyl, alkyl or alkoxy substituents. The halo group can be fluoro, chloro, bromo or iodo group, and the alkyl or thioalkyl preferably contains only a small number of carbon atoms. Further oxidation can be observed when a hydroxyl substituent is ortho or para to the aldehyde substituent, and the sulphur

atom in the thioalkyl group demonstrates a tendency to be oxidised to a sulphoalkyl group. Extremely good yields have been observed when a deactivating ie electron-withdrawing substituent is present, in the ortho or para positions as well as the meta position. In a further variation, the substituent can also comprise a nitrogen hetero-atom within the aromatic nucleus or in a yet further variation the benzene nucleus can be replaced by a di or poly-cyclic aromatic nucleus in which the aldehyde substituent is located at the 2 carbon atom.

Whilst the invention has been demonstrated most extensively in respect of the oxidation of aldehydes that are substituted by only one non-aldehyde substituent or none, it will be recognised that present invention encompasses the oxidation of those aromatic aldehydes that are substituted by a plurality of non-aldehyde substituents, such as two or three. Naturally, when a plurality of such substituents are present, each can each be the same as or different from the other or others. It is most advantageous to so array them as to minimise or eliminate steric hindrance effects. Suitable examples include di-halo benzaldehydes and 3,5-dimethoxy benzaldehyde. It is particular important to avoid 2,6-disubstitution, probably for steric hindrance reasons. From the foregoing it will be recognised that the invention process preferentially oxidises the aldehyde substituent, and accordingly provides a convenient route for the synthesis a wide range of substituted benzoic acids.

For the avoidance of doubt, it will be understood that the present invention process relates specifically to compounds in which the aldehyde group is a direct and immediate substituent of the aromatic nucleus. It does not extend to the oxidation of saturated aliphatic aldehydes, demonstrated by n-butanal and n-heptanal, or to aromatic compounds in which one or more aliphatic carbons are interposed between the nucleus and the aldehyde, such as phenylacetaldehyde or 2,2-diphenyl cetaldehyde.

The alkali perborate is particularly conveniently a sodium perborate on account of the bulk availability and excellent storage and handling properties of the two industrially available products sodium perborate monohydrate and sodium perborate tetrahydrate, which have respectively the empirical formulae NaBθ3.H ₂ 0 and aBθ3.4H ₂ 0, though these do not properly represent the structure of the compounds. Whilst there are various ways in which the compound can be introduced into the reaction medium, a particularly safe way comprises introducing it progressively, such as in small portions or continuously during an introductory period, either at or below the desired reaction temperature.

Advatageously, it has been found that the invention reaction conditions permit the oxidation of the aldehyde to be effected using only a small excess of perborate beyond the stoichiometric amount of 1 mole per mole of aldehyde. In practice, it is preferable to employ a mole ratio for perborateraldehyde selected in the range of from 1:1 to 2:1 and particularly from 1.1:1 to 1.5:1. Naturally, there is some variation in the efficiency of the reaction depending upon which substituents are present, but by appropriate selection of conditions, it is possible to obtain very high conversion to the carboxylic acid at a mole ratio chosen within the aforementioned ranges.

The reaction medium particularly conveniently comprises glacial acetic acid. The concentration of substrate in the reaction medium can be selected within a very wide range, for example from 0.1M to a saturated solution. By the expression "mild temperature" in the context of the present reaction is meant that there is no need to maintain a high temperature during the reaction or even approach closely reflux temperature for the medium. In many instances, it is convenient to employ a temperature that is above ambient, and preferably above 40 ^& C, up to about 70°C. Very effective oxidations have been achieved in the region of or around 45 to 50°C throughout the

reaction period.

The precise mechanism for the present invention is open to discussion at present. It is speculated that there may be some mode or modes of interaction between the perborate oxidant and the reaction medium which can result in the generation in situ of one or more active species that is or are resposible for the effective oxidation reaction. It will be recognised, though, that the present invention stands by virtue of its demonstrated results and not by the truth or otherwise of any particular point of speculation. The total reaction period will usually be determined in practice by taking into account the reaction temperature and the substrate and will often include a period during which perborate is introduced and a subsequent period in which the reaction is allowed to progress. The perborate introduction period is often chosen within the range of 10 to 60 minutes. The subsequent period is often chosen in the range of from 15 minutes to 10 hours and for many of the substrates in the range of from 15 minutes to 120 minutes. Some reaction can occur whilst the perborate is being introduced so that the total reaction period is often selected in the range of from 30 minutes to 10 hours, and for many substrates from 30 to to 150 minutes. The presence of an electron-withdrawing substituent tends to enable a shorter reaction period to be selected and an electron-donating substituent tends to demand a longer reaction period. The reaction can be monitored, for example by thin layer chromatography and recovery of the product commenced when the monitoring indicates that either a desired proportion of the substrate has been converted to the product, or the reaction rate has slowed thereby indicating that little further product could be obtained. In practice, reaction periods can be gauged in small scale trials and refined in bulk-scale operation. The invention process is particularly suitable for a batch style reaction procedure, but it will be recognised that by a suitable choice of reactor design such a tubular once through reactor, it is a practical proposition to carry

out the reaction continuously, especially for those substrates that employ a relatively short reaction period. The product can be recovered from the reaction mixture by-removal of at least part of the reaction medium and preferably all of it, such as by evaporation, preferably under reduced pressure, and subsequent addition of water to the residue, thereby causing formation of a solid product. A suitable amount of water is often chosen in the range of 10 to 100 parts w/w per part of substrate. The solid can then be separated using conventional solid/liquid separating techniques such as centrifugation, filtration or settling.

The aqueous residue comprises a saturated solution of the product. Accordingly, a further amount of product can be recovered by subsequently contacting the aqueous residue with a low molecular weight aliphatic ester such as ethyl acetate or a similar solvent having low boiling point and low miscibility with water, separating the two phases and removing the solvent, such as by evaporation, preferably under reduced pressure. A convenient ratio of stripping solvent to aqueous residue is often chosen in a total v/v ratio of 1:1 to 3:1. The conventional techniques of solvent stripping, viz continuous co- or counter-current contact or multiple batch contact are applicable. It will also be recognised that product losses can be reduced additionally to the solvent stripping technique or alternatively instead of that technique by recycling the aqueous residue, either before or after its contact with the above-mentioned solvent, in place of at least a fraction of the water that is added to the reaction mixture residue in an early phase of the product recovery.

Having described the invention in general terms, specific embodiments will now be described more precisely by way of non-limiting example only. Examples 1 to 31

Each of these Examples was carried out using the following general procedure. A substrate identified in the Table as

ArCHO, 10 ^-2 moles, was dissolved in glacial acetic acid, 30 is. The solution was stirred and maintained at about 45- 50 ^ϋ C during the addition of sodium perborate tetrahydrate, 1.2-χ 10~ ² moles, in small portions over a period of 20 minutes and throughout the subsequent reaction period. The progress of the reaction was monitored by withdrawing a number of small samples from the reaction mixture at intervals for immediate analysis by thin layer chromatography. The reaction was allowed to continue until the analyses indicated that all the substrate had been consumed. This varied from about 30 minutes for 4-chlorobenzaldehyde up to as long as 8 hours for 3,5-dimethoxybenzaldehyde. The acetic acid was removed from the reaction mixture by evaporation under reduced pressure and water, 50 mis, added to the residue. The solid which separated out, crude product, was recovered by filtration, and dried in air. The - filtrate was contacted with ethyl acetate, 3 x 25 mis portions, and the combined organic phase was dried with anhydrous magnesium sulphate, and evaporated under reduced pressure, thereby precipitating a further amount of crude product. The two crude products were combined, dried and recrystallised to provide the yield given in the Table, which is the molar percentage of purified carboxylic acid product, based on the substrate present initially. The identity of the product was subsequently confirmed by melting point comparison with the reading given in the literature and by infra red spectral analysis.

In Examples 17 to 19 and 29, the mole ratio of sodium perborate:substrate was multiplied accordingly to allow for the presence of respectively a second aldehyde substituent and an oxidisable sulphur compound.

The Table

Example No Substrate % Yield

1 C ₆ H ₅ CHO 93

2 ^• 2-CH ₃ C ₆ H ₄ CHO 77

3 3-CH ₃ C ₆ H ₄ CHO 84

4 4-CH ₃ C ₆ H ₄ CHO 92

5 2-ClC ₆ H ₄ CHO 82

6 3-ClC ₆ H ₄ CHO 93

7 4-ClC ₆ H ₄ CHO 94

8 2-FC ₆ H ₄ CHO 79

9 3-FC ₆ H ₄ CHO 86

10 4-FC ₆ H ₄ CHO 86

11 2-BrCgH ₄ CHO 71

12 3-BrC ₆ H ₄ CHO 91

13 4-BrC ₆ H ₄ CHO 90

14 4-IC _g H ₄ CHO 87

15 2,4-Cl ₂ C ₆ H ₃ CHO 87

16 3,4-Cl ₂ C ₆ H ₃ CHO 77

17 2-CHOC ₆ H ₄ CHO 70

18 3-CHOC ₆ H ₄ CHO 90

19 4-CHOC ₆ H ₄ CHO 93

20 4-COOHC ₆ H ₄ CHO 81

21 4-NCC ₆ H ₄ CHO 71

22 2-N0 ₂ C ₆ H ₄ CHO 86

23 3-N0 ₂ C ₆ H ₄ CHO 83

24 4-N0 ₂ CgH ₄ CHO 90

25 4-iPrC ₆ H ₄ CHO 79

26 3-OHCgH ₄ CHO 70

27 3-CH ₃ OCgH ₄ CHO 83

28 3,5-(CH ₃ 0) ₂ CgH ₃ CHO 90

29 4-CH ₃ SC ₆ H ₃ CHO 90

30 2-CHOC _1Q H ₇ 90

31 4-CHOC _ς H _Δ N 86

From the Table above, it can be seen that a very wide range of substituted benzaldehydes can be oxidised to the corresponding substituted benzoic acids under the conditions of the present invention. Furthermore, by comparison

between sets of Examples, such as 2 to 4, 5 to 7 and 8 to 10, it can be seen that the process .is effected more efficiently when the substituent is present in the meta or para position rather than in the ortho position around the nucleus. Examples 17 to 19 demonstrate the formation of dicarboxylic acids from the corresponding di-aldehyde. Examples 27 and 28 demonstrate the formation of methoxy and dimethoxy benzoic acid respectively when the methoxy group is meta to the aldehyde. However, if the methoxy group were ortho or para, the resultant product would be the corresponding methoxyphenol as a result of a Dakin reaction and not the corresponding substituted benzoic acid,