Microwave-assisted synthesis of 4-oxo-2-butenoic acids by aldol-condensation of glyoxylic acid

4-Oxobutenoic acids are useful as biologically active species and as versatile intermediates for further derivatisation. Currently, routes to their synthesis can be problematic and lack generality. Reaction conditions for the synthesis of 4-oxo-2-butenoic acid by microwave-assisted aldol-condensation between methyl ketone derivatives and glyoxylic acid have been developed. They provide the desired products in moderate to excellent yields for a wide range of substrates, by applying a simple procedure to accessible starting materials. The investigation revealed different conditions are required depending on the nature of the methylketone substituent, with aryl derivatives proceeding best using tosic acid and aliphatic substrates reacting best with pyrrolidine and acetic acid. This substituent effect is rationalised by frontier orbital calculations. Overall, this work provides methods for synthesis of 4-oxo-butenoic acids across a broad range of substrates.

Introduction 4-Oxo-2-butenoic acids are interesting building blocks for drug discovery. Several derivatives have shown biological activity in their own right, for example to treat cancer, 1,2 neurodegenerative, 3 metabolic, 4 and gastric 5 conditions as well as antimicrobial 6,7 or antifungal 6 properties (Fig. 1). In addition, their high reactivity makes them versatile intermediates for further derivatisation.
Preparation of 4-oxo-2-butenoic acids has oen proven to be scope-limited, with Friedel-Cras acylations 1 used for aromatic substrates (Fig. 2) and oxidative furan-opening 8 for aliphatic ones (Fig. 2). Although we managed to obtain the desired 4-oxo-2-butenoic acid product when applying the oxidative furan oxidative conditions to an electron-rich aromatic example, we were unable to identify conditions compatible with electron-decient aromatic substrates.
Aldol-condensation with glyoxylic acid is compatible with a larger range of starting materials with a few examples available in the literature, mainly using acetophenone derivatives as substrates (Fig. 2). Literature reported conditions are typically acid-promoted, most commonly by acetic acid, 7,9-12 sulphuric acid, [13][14][15][16][17] phosphoric acid, 13,18 toluene-4-sulfonic acid 19 and formic acid, 20 with two acids frequently used together. 12,19,[21][22][23][24][25][26][27][28] The acid promotors are usually used neat or in large excess under reux for relatively long periods of time. A few base-promoted procedures are also described in the literature, using potassium carbonate 29,30 or sodium hydroxide at reux or under reduced pressure. 31 All these conditions are quite harsh on the reactants and resulting products, thus limiting the scope of the transformation.
Therefore, we decided to investigate the aldol-condensation with glyoxylic acid to identify efficient conditions for the preparation of 4-oxo-2-butenoic acid derivatives. We aimed to condensation product. Attempts to force the water elimination by addition of tosyl chloride led to an improved conversion of the starting material. It was proposed that the tosyl chloride was hydrolysed in situ and the reaction was catalysed by tosic acid. Accordingly, carrying out the direct tosic acid-promoted aldolcondensation provided the desired product in good yield (70%), conrming the previous hypothesis (Table 1, entry 1). These conditions were applied to the more electron decient 4cyanoacetophenone, in which case, even aer prolonged heating, some starting ketone remained (Table 1, entry 2). To improve the conversion towards the formation of the product 2 as well as reducing the heating time, microwave-assisted heating was performed using similar conditions, leading to a moderate isolated yield of 2 (32%, Table 1, entry 3). Optimisation of the irradiation time and temperature allowed an increase of the yield to 45% with a reduced reaction time of 1 h ( Table 1, entry 4). Shorter reaction time led to incomplete conversion of the starting material whereas increase of the temperature from 160 C to 180 C led to partial or total degradation of the desired product (Table 1, entries 5-7), suggesting that heating at 160 C for 1 hour were giving the best results. These ndings suggest that increased temperature and pressure allowed by the microwave reactor are key to drive the   reaction while allowing a simple reaction set-up. Therefore, these optimised conditions were also applied to 4-methoxyacetophenone and showed an improved yield of 94% (Table 2). When applying these conditions to cyclohexylmethyl ketone starting material, no desired product 3 was formed ( Table 1, entry 8). The major product was the aldol adduct intermediate. However, treatment of cyclohexylmethyl ketone with glyoxylic acid, in the presence of pyrrolidine and acetic acid ( Table 1, entry 9), using microwave-assisted heating, enabled the isolation of the desired product 3 in 25% yield. Increasing the temperature from 80 C to 100 C led to product degradation, even with decreased reaction time ( Table 1, entry 10), but decreasing the temperature to 60 C improved the yield to 52% (Table 1, entry 11).
Electron donating and withdrawing substituents and ortho, meta and para substitution patterns were well tolerated. No desired product formation was observed when these conditions were applied to aliphatic substrates 3, 12-14 and 16-18. Pentan-2-one, however, yielded the desired product 15 under the TsOHpromoted conditions (by NMR) as a minor component relative to the expected internal aldol-condensation product 19 (ratio 1 : 1.5, Fig. S1 †). Aer heating at 100 C for 16 hours, 19 was the only product formed (99% yield). Hence, this suggests, as expected, that the tosic acid mediated reaction may not be compatible with methyl ketones bearing an additional enolisable centre.
The pyrrolidine-acetic acid conditions were also applied to the selected substrates (Table 3). With aliphatic substrates, these conditions yielded the desired products 3 and 12-17 with yields between 43% and 92%, in which no product was obtained with the TsOH-promoted conditions. In this case, substrates with additional enolisable centres were tolerated (3,(12)(13)(14)(15)(16)(17) and the internal aldol adduct was not observed. No desired product was obtained for the t-butyl ketone 18, presumably due to increased steric hindrance. Aromatic products 1, 4, 5 and 7-11 were obtained in poor 4 to 12% yields, much lower than with the TsOH-promoted conditions. However, no product was observed for the electronpoor examples 2 and 6 nor for the aliphatic chain 15 (Table 3).
For all the examples above, the obtained 4-oxo-2-butenoic acids all had the E conformation for the alkene bond conrmed by NMR. No trace of the Z alkene was observed, demonstrating the stereoselectivity of this transformation.
Finally, scale-up of the synthesis of 2 from 0.69 mmol to 6.9 mmol demonstrated the scalability of the TsOH-promoted reaction with no change in yield. Scale up of the synthesis of 3 using the pyrrolidine-acetic acid conditions from 4.0 mmol to 7.2 mmol gave increased yield (from 38% to 52%).
Based on the assumption that the reaction proceeds by attack of the protonated glyoxylic acid by the enol form of the methylketone under the tosic acid promoted conditions, and by either the enol or the enamine in the presence of pyrrolidine/ acetic acid (Scheme 1) as the rate determining step, the observed differences in reactivity were rationalised from the calculated energy gaps between the protonated glyoxylic acid LUMO and the enol or enamine HOMOs (Fig. 3). Calculations employed the RHF/6-31+G** level of theory in the Gaussian09 suite of programs. 32 Geometries were optimised and frequencies computed to verify that they are minima. . This provides an explanation as to why the pyrrolidine-acetic acid conditions work better for aliphatic substrates, and that these reactions likely proceed via the enamine as the predominant pathway.

Conclusions
We successfully identied two sets of conditions that provide E-4oxo-2-butenoic acids using microwave-assisted aldol-condensation with glyoxylic acid. These provide the desired products in good to excellent yields for substrate with various electronic and steric properties (electron-rich, -neutral and -poor aromatic, ortho-, meta-, and para-substituted aromatic, linear and cyclic aliphatic). The reaction can be scaled-up to mmol scale with no effect on the yield. It was observed that tosic acid promoted conditions work better for aromatic substrates whereas acid acetic and pyrrolidine are preferred for aliphatic substrates. These ndings can be rationalised by the relative differences in HOMO-LUMO energy gaps for the enol and enamine intermediates. Overall, we describe a convenient and efficient synthesis for 4-oxo-2-butenoic acids which should allow an easier access to these biologically-relevant molecules and their derivatives.

Experimental
General procedure A for the synthesis of 1, 2 and 4-11 In a Biotage microwave vial, acetyl derivative (1 eq.), glyoxylic acid monohydrate (3 eq.) and TsOH monohydrate (1 eq.) were dissolved in dioxane (2.5 mL mmol À1 ). The vial was closed with a 20 mm aluminium crimp cap with a PTFE/silicone septum and heated in the microwave for 1 h at 160 C using the low absorption mode while stirred at 600 rpm with a PTFE stirring bar. 2 M HCl aqueous solution was added to the mixture. This was extracted 3 times with CH 2 Cl 2 . Combined organic phases were dried over MgSO 4 . The solvent was removed under vacuum. A typical scale was 2.4 mmol but the reaction was successfully scaled up and down.