Non-metallic crystalline instabilities

General introduction

There are two well-recognised causes of non-metallic crystalline deposits in wines, both resulting from the precipitation of phenolic compounds. The two phenolics involved in these instabilities, however, have quite different origins. The first is the flavonol compound quercetin, which is a part of the natural composition of grapes, although practices in grape harvesting can also contribute to its occurrence in wine (Ziemelis 1982). The second phenolic constituent that can precipitate as a crystalline deposit is ellagic acid, and incidences of this instability result from practices at the end of the winemaking process (Pocock et al. 1984). Both of these deposits are entirely organic and are the products of hydrolysis reactions proceeding during the life of the wine. Importantly, both ellagic acid and quercetin come from soluble precursors that are broken down in these hydrolysis reactions.

Deposits involving these two phenolics can affect both red and white wines but as an instability problem they are more commonly experienced in white wines. When found in red wines, ellagic acid or quercetin is often part of a mixed deposit which may include potassium bitartrate, calcium L-tartrate, and pigmented polyphenolic material. There is, however, no evidence that the occurrence of either of these crystalline phenolics in an unstable red wine is responsible for, or in any way initiates, the formation of any accompanying deposits.

Flavonol Hazes

Background to flavonol deposits and haze problems

Quercetin glycosides and possibly other flavonol glycosides, for example those of kaempferol, are the soluble precursors involved in flavonol hazes. It seems, however, that quercetin, which is the most abundant flavonol of grapes, is almost exclusively found as the offending crystalline deposit. Flavonol glycosides are much richer in the skins than in the flesh and pulp of the grape. Leaves too can also contribute to flavonol glycosides in must. Recent evidence indicates that sun exposure of fruit will enhance the flavonol glycoside content of skins (Price et al. 1995). Therefore, trellising practices maximising sun exposure of fruit will increase these compounds in must, as will excessive leaf matter in a mechanically harvested crop. It is to be stressed that the flavonol glycosides are not themselves directly responsible for any instability problems. Furthermore, other than increasing the potential for instability, grape composition is not impacted negatively by having enhanced levels of these natural constituents.

Properties of flavonol glycoside precursors

The nature of the carbohydrate part of the flavonol glycosides is quite variable, with a simple glucoside being only one of several glycosides that are usually present. Glucuronides (ie glycosides involving glucuronic acid) together with several other glycosides may dominate. Under the acid conditions of wine, each of the glycoside species may hydrolyse at different rates. Additionally, this diverse array of glycosides will respond more or less rapidly to the activity of various enzymes that may be present to give the insoluble quercetin aglycone. There is very little of the free aglycone in must at crushing and its progressive generation from the different glycosides is the cause of the instability problem.

Appearance and properties of flavonol deposits

Crystalline deposits of quercetin can take a number of forms. They have been found as small beam-shaped crystals and as yellow, pale green or tan coloured needles. The latter have been observed even as long fibres which initially could be mistaken for asbestos particles. The occurrence of quercetin in this fibrous form is the reason that the instability is sometimes described as a haze. The crystalline deposit is found on analysis to be quercetin dihydrate, and is readily identified by IR spectroscopy. UV analysis also shows λmax 370 nm with a bathochromic shift to λmax 430 nm on addition of aluminium trichloride solution.

Strategies to counter flavonol haze problems

Modern viticultural practices emphasising sun exposure of fruit mean that grape composition with higher levels of flavonol glycosides is a feature of contemporary winemaking. Other sources of flavonol glycosides, notably leaves from mechanically harvested fruit, can be minimised by ensuring efficient elimination of leaf material during the harvesting operation. Prolonged skin contact encourages the transfer of flavonol glycosides and so light bodied white wines may experience problems less often than their full bodied counterparts. Pressings will also be rich in flavonol precursors (Ziemelis 1982). Maceration in red winemaking ensures maximum transfer from skins in these styles (Ziemelis and Pickering 1969). Finally, the use of enzymes with glycosidase activities for pectin removal will hasten hydrolysis of the soluble flavonol glycoside precursors, to give the insoluble quercetin. Enzymes with other functions that may be employed during the vinification and which have glycosidases as side-activities, can also catalyse flavonol glycoside hydrolysis. Prolonged storage of red wines in wood etc prior to finishing and bottling provides an opportunity for some self-stabilisation, thus limiting (but from experience, not necessarily eliminating) the likelihood of instability.

A predictive test with a glycosidase enzyme applied to a small volume of a suspect wine prior to finishing and bottling has been employed with some success (Ziemelis 1982). Observation of quercetin crystals in the enzyme-treated test sample indicates a susceptibility to deposit formation.

Ellagic acid

Properties of ellagic acid and characteristics of ellagic acid instability

Ellagic acid occurs as a microcrystalline deposit, with crystal shapes of rods, prisms or beams. The crystals are usually light brown when formed in white wines. From red wines the instability usually shows itself as part of dark red sediment comprising small crystals of ellagic acid together with amorphous pigmented polyphenolic material.

Ellagic acid has a very high melting point (in excess of 300ºC), and this property may lead to a mis-diagnosis of the precipitate as being of inorganic origin. It is also insoluble in water, which is the cause of its precipitation. Ellagic acid is soluble in NaOH and is readily identified by IR spectroscopy.

Ellagic acid is a product of the hydrolysis of oak wood tannins, specifically ellagitannins. Such hydrolysis is a natural process which occurs during the contact of wine with oak wood. On storage of wine in oak barrels, all of the ellagic acid formed usually precipitates out before the wine is bottled. Most commonly, instability problems arise from the contact of wine with wood (and especially chips or shavings etc) shortly before bottling. In these circumstances the extracted ellagitannins are incompletely hydrolysed before bottling and the ellagic acid is produced and precipitated in the bottle.

Procedures to minimise the likelihood of ellagic acid instability

Experience has shown that to avoid instability problems wine should be stored for a minimum period of four weeks after final wood treatment and before bottling (Pocock et al. 1984). There have, however, been cases in which wines stored in bulk at low temperature for several months have still exhibited ellagic acid instability. Reasons for this apparent anomaly may have been due to the low storage temperature slowing the hydrolysis of the ellagitannin precursor. High levels of glucan and/or other polysaccharides in wines may also exert a ‘protective-colloid’ effect and delay the precipitation of ellagic acid.

Non-grape wines and ellagic acid instability

Fruit wines from strawberries and loganberries etc, can show ellagic acid instability even in the absence of any wood treatment. These non-grape berry fruits have a naturally high content of ellagitannins, which will hydrolyse in the finished fruit wine giving an ellagic acid deposit.


Pocock, K.F., Strauss, C.R., Somers, T.C. 1984. Ellagic acid deposition in white wines after bottling: a wood-derived instability. Aust. Grapegrower Winemaker. (244): 87.

Price, S.F., Breen, P.J., Valladao, M., Watson, B.T. 1995. Cluster Sun Exposure and Quercetin in Pinot noir Grapes and Wine. Am. J. Enol. Vitic. 46(2): 187-194.

Ziemelis, G. 1982. Flavonol haze – a new form of wine instability arising from technological change. Clarke, J. (ed.) The Institute of Brewing (Australia and New Zealand section): proceedings of the seventeenth convention; 7-12 March 1982; Perth, WA. Sydney, NSW: The Institute of Brewing (Australia and New Zealand section): 75-76.

Ziemelis, G., Pickering, J. 1969. Precipitation of flavonols in a dry red table wine. Chemistry and Industry. 49: 1781-1782.