Читать книгу Handbook of Enology, Volume 2 - Pascal Ribéreau-Gayon - Страница 26

The addition of tartaric acid is permitted under European Union (EU) legislation, up to a maximum of 1.5 g/l in must and 2.5 g/l in wine. In the United States, acidification is permitted using tartrates combined with gypsum (CaSO₄) (Gomez‐Benitez, 1993). This practice seems justified if the buffer capacity formula (Equation (1.3)) is considered. The addition of tartaric acid (HA) increases the buffer capacity by increasing the numerator of Equation (1.3) more than the denominator. However, the addition of CaSO₄ leads to the precipitation of calcium tartrate, as this salt is relatively insoluble. This reduces the buffer capacity and, as a result, ensures that acidification will be more effective.

FIGURE 1.6 Variations in the buffer capacity of an aqueous solution of tartaric acid (40 mM) in the presence of several amino acids (Dartiguenave et al., 2000a).

FIGURE 1.7 Variations in the buffer capacity of an aqueous solution of malic acid (40 mM in the presence of several amino acids (Dartiguenave et al., 2000a)).

Whenever tartrate addition is carried out, the effect on the pH of the medium must also be taken into account in calculating the desired increase in total acidity of the must or wine. Unfortunately, however, there is no simple relationship between total acidity and pH.

FIGURE 1.8 Hypothetical structure of interactions between tartaric acid and amino acids (Dartiguenave et al., 2000a).

A decrease in pH may occur during bitartrate stabilization, in spite of the decrease in total acidity caused by this process. This may also occur when tartrates are added to must and, in particular, wine, owing to the crystallization of potassium bitartrate, which becomes less soluble in the presence of alcohol.

The major difficulty in tartrate addition is predicting the decrease in pH of the must or wine. Indeed, it is important that this decrease in pH should not be incompatible with the wine's organoleptic qualities, or with a second alcoholic fermentation in the case of sparkling wines. To our knowledge, there is currently no reliable model capable of accurately predicting the drop in pH for a given level of tartrate addition. The problem is not simple, as it depends on a number of parameters. To achieve the required acidification of a wine, it is necessary to know the ratio of the initial concentrations of tartaric acid and potassium, i.e. crystallizable potassium bitartrate.

It is also necessary to know the wine's acid–base buffer capacity. Thus, in the case of wines from cool‐climate regions, initially containing 6 g/l of malic acid and having gone through malolactic fermentation, tartrate addition may be necessary to correct an impression of “flatness” on the palate. Great care must be taken in acidifying this type of wine; otherwise it may have a final pH lower than 2.9, which certainly cures the “flatness” but produces excessive dryness or even greenness. White wines made from red grape varieties may even take on some red color.

Table 1.10 shows the values of the physicochemical parameters of acidity in sparkling base wines, made from the cuvée or second pressings of Chardonnay grapes in the 1995 and 1996 vintages. They were acidified with 1 g/l and 1.5 g/l tartaric acid, respectively, after the must had been clarified.

Examination of the results shows that adding 100 g/l to a cuvée must or wine only results in 10–15% acidification, corresponding to an increase in total acidity of approximately 0.5 g/l (H₂SO₄). Evaluating the acidification rate from the buffer capacity gave a similar result. The operation was even less effective when there was a high potassium level, and potassium bitartrate precipitated out when the tartaric acid was added.

Adding the maximum permitted dose of tartaric acid (1.50 g/hl) to second‐pressing must or wine was apparently more effective, as total acidity increased by 35% and pH decreased significantly (−0.14), producing a positive impact on wine stability and flavor. The effect on pH of acidifying wines shows the limitations of adding tartaric acid; there may also be problems with the secondary fermentation in bottle, sometimes resulting in “hard” wines with a metallic mouthfeel.

It would be possible to avoid these negative aspects of acidification by using L(−)lactic acid. This is listed as a food additive (E270) and meets the requirements of both the Food Chemicals Codex and the European Pharmacopoeia. Lactic acid is commonly used in the food and beverage industry, particularly as a substitute for citric acid in carbonated soft drinks and is even added to some South African wines.

Its advantages compared with tartaric acid are a pK_a of 3.81 (tartaric acid: 3.01) and the fact that both its potassium and calcium salts are soluble. This enhances the acidification rate while minimizing the decrease in pH. Finally, lactic acid is microbiologically stable, unlike tartaric, malic, and citric acids. Until recently, one disadvantage of industrial lactic acid was a rather unpleasant odor, which justified its prohibition in winemaking. The lactic acid now produced by fermenting sugar industry residues with selected bacteria no longer has this odor.

TABLE 1.10 Composition of Chardonnay Wines After Tartrate Stabilization, Depending on the Time of Acidification (Addition to Must or to Wine After Malolactic Fermentation)

	Cuvée	Second pressing
1995	1996	1996
Control	Acidified must	Acidified wine	Control	Acidified must	Acidified wine	Control	Acidified must	Acidified wine
pH	3.06	2.97	2.97	3.06	2.99	2.97	3.18	3.04	3.00
Total acidity (g/l, H2SO4)	5.2	6.0	5.6	5.4	5.9	5.8	4.1	4.9	5.0
Tartaric acid (g/l)	3.6	4.0	4.3	4.4	5.2	5.0	3.4	4.6	4.8
Malic acid (g/l)	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1	0.1
Lactic acid (g/l)	4	4.3	4.4	4.2	4.1	4.1	3	3	2.7
Total nitrogen (mg/l)	274.7	221.9	271	251.6	280.3	289.8	245.9	250.4	254.4
Amino acids (mg/l)	1,051.4	703.7	1,322.6	1,254.2	1,422.7	1,471.7	1,177.5	1,350.4	1,145
Potassium (mg/l)	390	345	320	345	290	285	380	305	300
Calcium (mg/l)	71.5	90	79	60	64	61	50	55	48
Buffer capacity (NaOH, H2O)	48.1	56.6	56.2	50.3	55.5	56.9	42.4	49.1	47.7
Buffer capacity (NaOH, EtOH 11% vol.)	55.6	59.2	55.9	47.1	51.9	50.2	37.9	44.3	42

Cuvées were acidified with 1 g/l tartaric acid and second pressings with 1.5 g/l (Dartiguenave, 1998).

Current production quality, combined with low prices, should make it possible to allow experimentation in the near future, and perhaps, even a lifting of the current ban on the use of lactic acid in winemaking.

The fact that a wine has an acid–base buffer capacity also makes deacidification possible. The additives authorized for deacidifying wines are potassium bicarbonate (KHCO₃) and calcium carbonate (CaCO₃). They both form insoluble salts with tartaric acid, and the corresponding acidity is eliminated in the form of carbonic acid (H₂CO₃) that breaks down into CO₂ and H₂O. A comparison of the molecular weights of these two salts and the stoichiometry of the neutralization reactions lead to the conclusion that, in general, one gram of KHCO₃ (MW = 100) added to one liter of wine produces a drop in acidity of 0.49 g/l, expressed in grams of H₂SO₄ (MW = 98). Adding one gram of CaCO₃ (MW = 100) to a liter of wine produces a decrease in acidity equal to its own weight (exactly 0.98 g/l), expressed in grams of sulfuric acid.

In fact, this is a rather simplistic explanation, as it disregards the side effects of the precipitation of insoluble potassium bitartrate salts and especially calcium tartrate, on total acidity as well as pH. These side effects of deacidification are only fully expressed in wines with a pH of 3.6 or lower after cold stabilization to remove tartrates. It is obvious from the pH expression (Equation (1.2)) that, paradoxically, after removal of the precipitated tartrates, deacidification using CaCO₃ and, more particularly, KHCO₃ is found to reduce the [salt]/[acid] ratio, i.e. increased true acidity. Fortunately, the increase in pH observed during neutralization is not totally reversed.

According to the results described by Usseglio‐Tomasset (1989), a comparison of the deacidifying capacities of potassium bicarbonate and calcium carbonate shows that, in wine, the maximum deacidifying capacity of the calcium salt is only 85% of that of the potassium salt. Consequently, to bring a wine to the desired pH, a larger quantity of CaCO₃ than KHCO₃ must be used as compared with the theoretical value. On the other hand, CaCO₃ has a more immediate effect on pH, as the crystallization of CaT is more complete than that of KHT, a more soluble salt.

In practice, CaCO₃ causes an instantaneous reduction in total acidity that is absolutely foreseeable (1 g/l expressed as H₂SO₄ for 1 g/l CaCO₃ added). Unfortunately, it creates an increase in calcium, which may induce difficult‐to‐control precipitations later on. Nevertheless, its use must be favored in winemaking. In contrast, KHCO₃ leads to a low, progressive acidity drop, which continues throughout the precipitation of KHT. It is not easy to determine the necessary dose for a given deacidification. In general, a lower dose than the theoretical one is used. However, the absence of calcium enables easier stabilization after treatment. Its use is recommended for minor deacidification of finished wine (for example, addition of 40–45 g/hl for an acidity drop, expressed as H₂SO₄, of 0.3 g/l).

In any case, such an operation requires a certain degree of prudence. It is actually controlled by the legislation of various countries. Too significant a correction for acidity should not be sought, considering the fact that deacidification with these methods only affects tartaric acid. This accentuates the tartrate/malate imbalance in the total acidity in wines that have not completed malolactic fermentation, as the potassium and calcium salts of malic acid are soluble.

There is a way of deacidifying these wines while maintaining the ratio of tartaric acid to malic acid. The idea is to take advantage of the insolubility of calcium tartromalate, discovered by Ordonneau (1891). Wurdig and Muller (1980) used malic acid's property of displacing tartaric acid from its calcium salt, at pH above 4.5 (higher than the pK_a2 of tartaric acid), in a reaction (Figure 1.9) producing calcium tartromalate.

FIGURE 1.9 Formation of insoluble calcium tartromalate when calcium tartrate reacts with malic acid in the presence of calcium carbonate.

The technology used to implement this deacidification, known as the DICALCIC process (Vialatte and Thomas, 1982), consists in taking volume V, calculated from Equation (1.5) below, of wine to be treated and adding to it the quantity of CaCO₃ necessary to obtain the desired deacidification of the total volume (V_T):

(1.5)

In Equation (1.5), A_i and A_f represent initial and final acidity, respectively, expressed in grams of H₂SO₄ per liter of the total volume V_T. The volume V of wine to be deacidified by crystallization and elimination of the calcium tartromalate must be poured over an alkaline mixture consisting, for example, of calcium carbonate (1 part) and calcium tartrate (2 parts). Its residual acidity will then be very close to 1 g/l as H₂SO₄.

It is important that the wine should neutralize the CaCO₃/CaT mixture and not the reverse, as the formation of the stable, crystallizable, double tartromalate salt is only possible above pH 4.5. Below this pH, precipitation of endogenous calcium tartrate occurs, promoted by homogeneous induced nucleation with the added calcium tartrate, as well as precipitation of potassium bitartrate by heterogeneous induced nucleation (Robillard et al., 1994).

The addition of calcium tartrate is necessary not only to ensure that the tartaric acid content in the wine does not restrict the desired elimination of malic acid by crystallization of the double tartromalate salt but also to maintain a balance between the remaining malic and tartaric acid.