Tartaric acid and related compounds seem to crop up a lot in some of the posts we’ve been preparing for Picture it… Chemistry, so we thought it was about time we took a look at this substance, which has played an important part in the history of chemistry.
1) Tartaric acid is a naturally occurring organic acid that occurs in many fruits, and which has been known to winemakers for centuries. It is diprotic – that is, it can lose two protons, much like (say) sulfuric acid, but it shows much less propensity than sulfuric acid to do so. For this reason, whilst sulfuric acid is corrosive and should definitely not be ingested, tartaric acid actually finds use as a food additive (it is E334). I remember making sherbet dip as a school boy, and tartaric acid was one of the things we put in to make it fizz.
Salts of tartaric acid are very common; the tartaric acid loses one or both of the acidic protons on the ends, and the charge is balanced by a cation. Cream of tartar is potassium hydrogen tartrate (aka potassium bitartrate), where one proton is replaced by a potassium ion, and it in this form that it occurs in winemaking – under certain conditions, it can crystallise on the corks of wine bottles, to make what are known as ‘wine diamonds’. In fact, this is the origin of most commercial tartaric acid, which is a by-product of fermentation or wine making (https://www.tartaric.com/tartaric-products/).
A second salt of tartaric acid is potassium sodium tartrate, in which the other hydrogen atom has been lost and replaced with a sodium ion. This is commonly known as Rochelle salt, after the French town where it was first isolated, but sometimes it is called Seignette salt, after the man who did it. Crystals of it display the phenomenon of piezoelectricity, which means that when they are squeezed they generate an electrical field. This is most commonly seen in everyday life in lighters for gas stoves, the kind where squeezing a handle compresses a piezoelectric crystal (though not Rochelle salt) and generates a spark.
A further salt of tartaric acid is potassium antimony tartrate, which was historically known as tartar emetic. Antimony is a powerful emetic, and so a method of inducing vomiting in the 16th and 17th centuries was to drink wine which had been left in an antimony cup (Figure 2). The tartaric acid in the wine dissolves some of the metal from the cup, and when the wine is drunk it causes vomiting. An accidental incidence of the same principle occurred in Newcastle in 1929, when a company there provided their workers with lemonade that had been prepared in cheap buckets. The lemonade was made from crystals that contained tartaric acid (much like my sherbet!), and the cheap buckets contained antimony. 56 people ended up in hospital with symptoms that encompassed burning, nausea, colic, vomiting and collapse (Proc. R. Soc. Med., 1977, 70, 756–763. ).
2) Rochelle salt is used in many situations that require the controlled reduction of the group 11 metals (copper, silver and gold, also called coinage metals). ‘A’ level chemists may well be familiar with the Fehling’s test for reducing sugars, in which a positive test causes a deep blue solution to deposit a brick red precipitate. The reagent is prepared by mixing an aqueous solution of copper sulfate with a strongly alkaline solution of Rochelle salt, which creates a deep blue copper tartrate complex. In the test the aldehyde groups of the sugars are oxidised to carboxylic acids, and the copper(II) ions are reduced to copper(I), which precipitates as the red oxide. Despite the antiquity of this test, it was not until 2013 that it was established that the active species is similar to that shown in Figure 3 – because of the strong alkalinity of the solution, the central -OH groups of the tartrate are deprotonated, and they are the groups that coordinate to the copper (Eur. J. Inorg. Chem., 2016, 1798-1807).
Similar chemistry is involved when Rochelle salt is used for stabilising solutions that contain silver; historically, many methods of silvering mirrors use it in a process known as the ‘Rochelle salt method’ (Publications of the Astronomical Society of the Pacific, 1911, 23, 13-32 ). In these systems, the tartrate ion serves as the oxidising agent as well as helping to solubilise the silver ions, but the reaction is slow – this is why it leaves behind a nice layer of silver (J. Chem. Soc., 1926, 129, 2178-2182). A more modern version of this uses the reduction of silver by potassium bitartrate (above) to produce silver nanoparticles, and it works for gold as well. (ChemPhysChem 2016, 17, 2551-2557). In these systems additives are used to keep the metal in solution as small particles and to stop them forming continuous layers.
As shown in Figure 1, tartaric acid has a chain of four carbon atoms; the two end ones are part of carboxylic acid groups (-COOH), and the two middle one are bonded to alcohol (-OH) groups. These two central carbon atoms are chiral centres, which means that the molecule comes in forms which are related by mirror symmetry. This is a bit like left- and right-handed gloves; they look the same, but are in fact mirror images of each other, and the technical term for this is to say that they are enantiomers. Figure 4 shows the molecule redrawn to emphasise this, and shows both enantiomers.
Naturally occurring tartaric acid is only ever found as one of these enantiomers, which is officially called (2R,3R)-tartaric acid, or L-(+)-tartaric acid. This is true of most naturally occurring molecules that can come as enantiomers – normally, only one of those mirror images is found in nature, and why this is the case is one of the great unsolved scientific problems. However, heating naturally occurring (+)-tartaric acid can convert it to the other, (-) form, and this is how a mixture of equal amounts of both L-(+)-tartaric acid and D-(-)-tartaric acid was accidentally isolated in France in 1819. This mixture was at first called racemic acid (from the latin rasemus, meaning ‘grape’), since it was thought to be a different substance from the tartaric acid that was usually produced. Louis Pasteur noticed that crystals of the sodium ammonium salt of racemic acid came in left- and right-handed forms, which happens because it crystallises in such a way that any individual crystal contains only the (+)- tartate or the (-)-tartrate, but never both. He separated the crystals by hand, and showed that one form was identical to sodium ammonium tartrate derived from tartaric acid, and thus that racemic acid was a mixture of tartaric acid and something that appeared to be the same but with the other handedness. From this, he had the intellectual insight to deduce that this all reflected some underlying facet of the structure of tartaric acid, and thus identified chirality (Acta Cryst., 2009, A65, 371-389). Chemists still call the process of turning one enantiomer into a mixture of equal quantities of both enantiomers ‘racemisation’, and the resulting mixture is ‘racemic’.
The reason that the early chemists regarded racemic acid as a different compound to tartaric acid was that it had different physical properties – it crystallised differently, and had different solubility – and this is because when two chiral objects come together they can interact in different ways according to their chirality. This can be illustrated using our glove analogy from above. A left hand (hands are also chiral!) and a left glove fit together differently to a left hand and a right glove; the interaction of two chiral objects is different for different chiralities. In racemic acid, there are two different chiral chemicals interacting (L-(+)-tartaric acid and D-(-)-tartaric acid), whereas in naturally occurring tartaric acid there is just one (L-(+)-tartaric acid); so, racemic acid has different properties to tartaric acid.
3) Most uses of tartaric acid and its salts in the modern chemistry laboratory are based around its chirality – specifically, using it to cause something else to be chiral – to the extent that whole books have been written on the subject. Generally these methods fall into two distinct categories, which are called resolution and stereospecific synthesis. Resolution is the process of taking a mixture of enantiomers and separating them, whereas stereospecific synthesis is the (much more recently invented) process of only making one of them in the first place.
A classic example of tartrate’s use in resolution is seen in the resolution of metal tris-phenanthroline complexes, of the kind shown in Figure 5. (J. Chem. Educ., 1962, 39, 481-483). These cationic complexes occur as two enantiomers which are mirror images of each other, but when a solution of tartar emetic is added (containing the antimony D-tartrate anion), one enantiomer interacts much more strongly with the anion than the other, causing crystallisation of complexes containing only the strongly interacting cation. This can easily be separated by filtration from the other cations, which have remained in solution – the cation has been resolved into its two enantiomers.
The stereosynthetic uses of tartrate are numerous, and normally use esters of tartaric acid where the protons have been replaced by organic groups. A famous example of this is in a procedure called the Sharpless epoxidation, for which the American chemist Barry Sharpless was awarded a half share of the 2001 Nobel Prize in chemistry.
An epoxidation is the addition of an oxygen atom to a double bond, and this can be achieved using t-butylhydroperoxide with a titanium catalyst; however, there are two possible enantiomers of the product (Figure 6). By adding diethyltartrate to the reaction, it is possible to synthesise only one of these isomers.
This turns out to be exceedingly useful in organic synthesis, especially for making natural products (recall that many naturally occurring chemicals only come in one enantiomeric form, as mentioned above). Figure 7 shows a recent example plucked at random from the 2019 chemistry literature, which is a step in the synthesis of the anthracycline (+)-nivetetracyclate A (Org. Lett., 2019, 21, 785-788). The substrate is basically a flat molecule composed of four benzene rings joined side-by-side, so in theory the incoming oxygen could be attached to either face with equal probability. However, in practice the diethyltartrate favours reaction on the underside (of the molecule as drawn) 96.5% of the time, and only 3.5% of the time does it react on the top face. The difference of 93% between those two figures is called the enantiomeric excess.
The procedure works by creating a chiral pocket around the catalyst, into which the substrate can only fit in a certain way – much like the way an enzyme works. The exact nature of this pocket is still a matter for debate (J. Phys. Chem. A, 2019, 123, 1022-1029), but it’s generally accepted to be a dimeric species along the lines of figure 8.
Contributors: Chris Adams
Main image: by Glen Jackson, from https://www.pexels.com/photo/chemistry-crystal-tarttaric-acid-233902/
Figure 2: From Wikimedia Commons at https://commons.wikimedia.org/wiki/File:Hoerner%26KluferFehling.png
Figure 3: Science Museum, London, via Wellcome Images, firstname.lastname@example.org, http://wellcomeimages.org
Figure 5: By our very own Ben Mills, and from wikimedia commons at https://commons.wikimedia.org/wiki/File:Ferroin-cation-3D-balls.png
Figure 8: From Wikimedia Commons at https://commons.wikimedia.org/wiki/File:Transition_state.png
All other figures drawn ourselves.