The chemistry of Sorbic acid

ethylnitrolic acid

Aqueous Chlorination of Sorbic Acid

Enough of the intro stuff, now for some real chemistry.

Attempts at chlorinating Sorbic Acid with Sodium Chlorite

These experiments were developed from a paper by Spacu and Dumitrescu (1) on the determination of sorbic acid in foods by chlorinating the acid with excess sodium chlorite and then measuring the excess chlorine in the reaction mixture. Sodium chlorite is a fairly accessible bleaching /sterilizing agent and can sometimes be purchased on line from various suppliers. It comes as a white crystalline powder that consists of about 80% NaClO2 with the remainder being mainly NaCl (salt).

Sodium chlorite reacts with excess hydrochloric acid to give sodium chloride and chlorine, approximately according to the equation:

NaClO₂ + 4HCl -> NaCl + 4Cl + 2H₂O

The reaction with sorbic acid is, according to Spacu and Dumitrescu:

NaClO₂ + 4HCl + 2C₆H₈O₂ -> NaCl + 2C₆H₈O₂Cl₂ + 2H₂O



In the original work they were looking at low concentrations and higher concentrations do not always favour the same reaction route, however, in order to make recovery of the resulting compound easier I used fairly concentrated solutions. One problem was that I had no data of the properties, such as melting point and solubility, of the target compound so the first few experiments were carried out very much in the dark.

Experimental
The first 4 experiments were carried out on the same basic scheme:

Approximately 4g of sodium chlorite (80%) and 10g of potassium sorbate were dissolved in 40ml of water. In a separate beaker 23ml of 30% hydrochloric acid were diluted with 50ml of cold water. The mixed salt solution was then added dropwise into the rapidly stirred acid over about 10 to 20 minutes. The addition was accompanied but a slightly exothermic reaction and the mixture turns slightly yellow though this colour fades after 10 to 20 minutes. Faster addition causes a slightly more pronounced exotherm and some chlorine is lost from the solution.

In Experiment 1 the beaker was stood in a water bath at room temperature, about 16 C, as soon as the exothermic reaction was noted. When the addition was complete and the colour discharged the precipitated solid was filtered off, washed with a little water and dried to give 2.295g. The white compound can be recrystallized from dilute ethanol, azeotropic isopropanol or water. The latter solvent proved the best but about 30ml per g are required. When alcohols are used for the recrystallization the product was sticky.

After the filtrate had been discarded the product was identified as un-reacted sorbic acid.

Experiment 2 was conducted in exactly the same way except than the water bath was omitted. The ppt was filtered off as before (1.405g of sticky unreacted sorbic acid) and the filtrate extracted with 40ml of ether and then two further portions of 30ml each. The extracts were combined and the ether recovered by distillation to give a small volume of colourless liquid that soon crystallised into small rosettes of blades, these were filtered from the small volume of slightly thick aqueous liquid; yield was 2.741g after drying. This material is quite distinct form sorbic acid being highly soluble in water.

Experiment 3
In order to reduce the amount of unreacted sorbic acid this experiment was conducted as in Exp. 2 but with 4.616g of sodium chlorite. Unfortunately this generated a stronger exotherm (circa 40° C) and the ppt became very sticky, it stuck to the stir bar, the beaker walls and the anything else it touched. The sticky white material was ether soluble so the entire mixture was extracted with 50ml of ether in the beaker before pouring it into the separating funnel. It was then extracted with two further 30ml portions of ether. The ether was gently distilled from the combined extracts and the very pale yellow oily residue poured into a bowl and allowed to cool and the last of the ether to evaporate. About 2.5-3g of sticky white crystals remained after a few hours. Recrystallization from azeotropic isopropanol (87%) gave 0.822g of slightly sticky unreacted sorbic acid. Foolishly I rejected the filtrate; this would have been better evaporated and leached with water.

Initial Observation:
1) Higher reaction temperatures favour increased consumption of sorbic acid.
2) Higher reaction temperatures increase the amount of sticky polymer like by-product.

From these observations the obvious next step is to try a lower temperature reaction.

Experiment 4 was run with the 3.830g of sodium chlorite and 10.005g of potassium sorbate dissolved in 40ml of water and then chilled in the fridge to about 4°. 23ml of 30% hydrochloric acid were diluted with 50ml of water and placed in the freezer until it had reached -16° C.
The acid was placed on the stirrer plate and stirred rapidly as the mixed salt solution was added. A dense white ppt formed that eventually stopped the stir bar from moving; this is almost certainly simply precipitated sorbic acid. The temperature on completion of the addition was -5° so little or no exothermic reaction had occurred. In fact no reaction apart from the liberation of free sorbic appeared to have taken place so the mixture was left to stand and warm up slowly to room temperature overnight. The following morning the slurry was less thick and was easily vacuum filtered to give 3.800g of washed and dried sorbic acid. The yellow filtrate was extracted with three x 50ml portions of dichloromethane (DCM) and then the combine extracts dried with magnesium sulphate. The DCM was distilled off until only about 10-15ml remained in the flask; this was poured out into a bowl and allowed to evaporate slowly. No crystals formed but eventually a small volume of bright yellow, pleasantly ethereal smelling liquid remained. The smell of the liquid is reminiscent of perchloroethylene and I suspect that decarboxylation has occurred to give a polychloro pentane or pentene.

Conclusions
The desired reaction product appears to be exceptionally soluble in water and this may account for the low recovery of the material in Exp.2, basically a poor partition coefficient between water and ether. If this is the case, five or six extractions with ether may be required and possibly the addition of KCl to assist extraction.

The optimum temperature is probably about 15-20° C since higher temperatures favour the formation of a sticky by-product. Lower temperatures involve longer reaction times and this seems to result in a completely different reaction path.

A slightly larger excess of sodium chlorite coupled with a cooler temperature may reduce losses through unreacted sorbic acid and a greater number of ether extractions cycles may improve the yield of product.

These findings point to my next set of experiments but I also intend to investigate the direction of mixing with an experiment where the acid is added to the mixed salt solution.

1) Spacu & Dumitrescu; 1967; Determination of sorbic acid with sodium chlorite, Talanta, v14, p981.

3-Hexenoic acid by the reduction of Sorbic acid with Dithionite
6th May 2018

This experiment was inspired by a series of U2Us with the SM member Clearly-not-atara in which we discussed the possibility of sodium dithionite as a reductant for conjugate double bonds. Sorbic acid was chosen as the substrate for this experiment as it is, for me, the most readily available compound with conjugate double bonds. At the time the experiment was done I didn't have access to the original work only a schematic overview of dithionite chemistry from the internet (it looks like part of a slae brochure for a Na dithionite producer!).

The reaction is simple enough:


I am not sure about the nature of the counter oxidation product of the sodium dithionite, possibly sodium bisulphite. This would increase the acidity of the reaction mixture and this may be the reason that the schematic indicated that sodium bicarbonate should be added.

15g of potassium sorbate (0.1 moles) were dissolved in 200ml of water in a 600ml beaker equipped with a magnetic stirrer and heated to 80-90°C. 27.01g sodium dithionite (0.15 moles) and 16g of sodium bicarbonate (0.2 moles) were ground together in a pestle and mortar and then added to the hot sorbate solution in small portion over about 10 to 15 minutes. After each addition of the solid there was some brief frothing and the solution turned yellow. Heating and stirring were maintained for about 30 minutes after the completion of the dithionite addition.

The slightly cloudy solution was allowed to cool to room temperature and then acidified with 70ml of 30% hydrochloric acid to about pH 1. The solution became a cloudy suspension so the mixture was heated almost to boiling but this caused evaporation of much of the intensely “cheesy” smelling acid so the suspension was cooled to room temperature again. A skin of polymerisation product formed on the surface which was removed and saved for further investigation (Note 1). The still cloudy suspension was extracted with 50ml of dichloromethane (DCM) but much insoluble material now prevented a clean separation. The entire volume of liquid was filtered through a 7cm Buchner funnel and placed in a separating funnel (Note 2). Over a couple of hours a clean separation occurred and the DCM layer was removed, the residue being extracted again with a further 25ml of DCM. The combined DCM extracts were dried with magnesium sulphate and then the solvent was allowed to evaporate.

When the DCM had evaporated the residue consisted of colourless crystals in an oily acid. The two phases were separated by filtration in a small Hirsch funnel. The white crystals were very sparingly soluble in water and appear to be unreacted sorbic acid. The oil appears to be the 3-hexenoic acid, the yield was 5.81g, However, this is probably not a true reflection of the yield as it is volatile and the boiling of the reaction mixture in contact with air may have caused significant amounts to evaporate or polymerise. The weight of the crystals was not recorded but was <<1g.

Discussion
3-Hexenoic acid exists in two configurations; cis and trans. Both cis and trans isomers are liquids at room temperature (Mps of both are about 12°C) when pure. However, a mixture of the two is likely to be liquid to lower temperatures. The acid recovered in this experiment remains liquid to about 0-3°. If stored at 4° C for several weeks small colourless crystals form. The small amount a material precluded vacuum distillation with the equipment that I have. The acid was not investigated further but the strong cheese-like smell is characteristic of this type of acid, both saturated and mono-unsaturated.

From the experiment above and the observations made subsequently it would appear that slightly more dithionite is required to get complete reduction and a slightly longer reaction time may also help to improve conversion. The solution should not be boiled after acidification but probably treated with a filter aid and filtered before extraction with ethyl acetate or ether rather than DCM. The acid is fairly volatile and it is reported that it can be vacuum distilled without difficult.

Note 1; the skin was dissolved in dilute NaOH solution to give a yellowish solution, charcoal treated and filtered, acidified and extracted first with DCM and then with ethyl acetate. The later solvent appears to be a better extractant and in spite of the fact it was used after the DCM on evaporation it left the greater amount of residue. The residue was an acid oil smelling like the main product but a little darker in colour. A little over 1.5ml resulted. It deposited a few colourless crystals on standing in the fridge for several weeks but neither this nor the main product showed further tendencies to polymerise.
Note 2; the residue on the filter was found to be mainly sulphur.

Well I have had a go at a Diels-Alder type condensation between recrystallized p-benzoquinone and sorbic acid. I dissolved 0.05 Moles of the former and slightly more than a molar equivalence of sorbic acid in 65ml of benzene and refluxed for 4 hours. The light amber solution is now cooling but masses of pale green crystals have formed already. I have based my procedure on the classic Woodward et al paper on the total synthesis of reserpine (Tetrahedron 1958 p1-65 Yes that's one paper all 65 pages of it!!). He condensed vinylacrylic acid with benzoquinone to obtain two product (isomers), one a tetrahyronaphthaquinone carboxylic acid and the other an aromatic rearrangement product a dihydro-5,8-dihydroxynaphthalene-1-carboxylic acid. The former seem to rearrange into the later irreversibly under prolonged heating or under the influence of acids (therefore the former cannot be Fischer esterified).

Since I am using sorbic acid which has an extra methyl group on the end and I deliberately used prolonged refluxing to force the formation of the aromatic isomer I hope to have ended up with an extra methyl group in the 4- position relative to the carboxylic acid ie 1,4-dihydro-4-methyl-5,8-dihydroxynaphthalene-1-carboxylic acid. I am not sure how to confirm that this is indeed what I have prepared but I am going to try to make it fully aromatic by adding bromine across the 2,3 double bond, then de-hydrohalogenating this compound and finally oxidizing it to a naphthaquinone derivative. More details to follow

An interesting possibility is the lower temperature condensation of benzoquinone with two moles of sorbic acid to give a polyhydro-anthraquinone derivative. My next experiment! This need to be done before the initial condensation product aromatises.

I just tried to scale up the direct combining of sorbic acid and benzoquinone by the solventless fusion process that I tried before but at 10x the scale. With less than one gram of reactants in the original experiment the reaction was quiet with no obvious reaction apart from the melt becoming dark. I ruined this experiment by trying to leach the reaction mixture with sodium bicarbonate and obtained a dark brown creosote like liquid.

I tried this again with 5g of each reactant, roughly 1:1 molar ratio, ground together first and then gently warmed over a free ethanol flame in a small round-bottom flask. The mixture melted quite suddenly to a rich orange liquid. Then while I was holding the flask but not in then flame any more, the mixture began to darken and then suddenly erupted emitting copious amounts of acrid purple smoke which condensed to a purple dust on everything around about. No actual damage was done but I wasn't 'alf surprised and I have a bit of a clean up!.

So beware, if you decide to try this reaction without solvent!

The toluene solvated reaction is currently refluxing away quietly.

Update:- well I refluxed the toluene solution for 9 hours and then let it cool. This morning I found some yellow-brown-orange crystals in a light brown liquid with a rim of soft gloopy tar at the top of the liquid level. I warmed the flask and poured the contents into a beaker, this left behind most of the gloopy tar, another gift from chemistry hell, in the flask. The flask is now full of chromate sulphuric acid mixture and the contents of the beaker crystallising ontop of a layer of yet more tar.

I think the conclusion of yesterdays experiments is that the secret is patience and reaction temperatures at or below 80 C.

I have also recrystallised all of the compound C and obtained a small quantity of very pale yellow aggregates of diamond shaped crystals, currently drying.

[Edited on 17-2-2021 by Boffis]

I have a jar or 1,2,3,6-tetrahydrophthalic acid:



Does anyone know if this compound is a dienophile like maleic anhydride? Would it be better as an ester rather than a free acid.

My plan is to try and condense it with sorbic acid to form a hydro-8-methyl-naphthalene-2,3,5-tricarboxylic acid. Possible?