Polyphenols are known to have a role in must and wine oxidation and wine aging. They are major agents of wine evolution, either positive development or spoilage. (Wildenradt et al. 1974, Singleton et al. 1985, Singleton 1987, Cheynier et al. 1988).

However, polyphenols are difficult to monitor in wineries due to available technologies to analyze them, mainly UV visible spectrophotometry and liquid chromatography (HPLC). These methods have several disadvantages:

• They are time consuming due to sample preparation (filtration, dilution, acidification…)
• They can be difficult to achieve on-site due to equipment cost and skills required to run the analysis
• Results with HPLC are very detailed but difficult to interpret and make real-time decisions.

Therefore, information is missing for winemakers to make the right decisions at the right moment on polyphenol consideration.

Besides, electrochemistry was widely used by academics to study polyphenols oxidation, (Lunte et al. 1988, Hapiot et al. 1996, Makhotkina et al. 2013), phenolic composition (Kilmartin et al. 2002, De Beer et al. 2004, Makhotkina et al. 2012) and polyphenol extraction during winemaking. (Zou et al. 2002, Makhotkina et al. 2012) However, this method is also difficult to achieve in wineries to have real-time results because of electrode fouling during the measurement: oxidation of the polyphenols makes them adsorb on the electrode surface, which requires cleaning the electrodes between each measurement as stated in the description of the methods in all these references.

Solution to avoid electrode cleaning is to use printed disposable electrodes (Avramescu et al. 2002). Specific carbon-printed disposable electrodes for must and wine application were developed and these electrodes were shown to allow following wine phenolics and oxidation (Ugliano et al. 2013, 2015a, 2015b, 2015c, 2016). Measurements presented in this article were achieved with these carbon printed disposable electrodes and a Nomasense PolyScan P200 (Wine Quality Solutions). Measurement principle is based on linear sweep voltammetry, which consists in applying increasing voltages to the sample. At each voltage, different compounds are oxidized and consequently, electrons are liberated giving birth to a current. The resulting intensity vs voltage curve (Figure 1) is a “fingerprint” of the oxidizable compounds of the sample. This fingerprint evolves along winemaking process.

To facilitate real-time interpretation for winemakers and support decision making with PolyScan P200, indexes are calculated from this curve:

• PhenOx: total polyphenols content (in mg/L gallic acid equivalent) correlated to Folin Ciocalteu index
• EasyOx: easily oxidizable compounds.

Several examples of the use of these indexes as a decision aid during winemaking follow.

Monitoring extraction of polyphenols in white grapes pressing

The main consideration on white must is to stabilize the wine against oxidation in the early steps of the process. To achieve that aim, excessive polyphenols must be eliminated either by must fining or oxygenation. The difficulty for a winemaker is to evaluate early and fast the concentration of polyphenols in juice:

• during pressing to collect poor and rich ones in different tanks
• or before they are settled and before beginning of alcoholic fermentation to fine the juices or oxygenate them.

Monitoring polyphenol concentration allows optimizing pressing programs to separate polyphenol-poor juices from polyphenol-rich juices. By sending them to different settling tanks, the process can then be adapted to the concentration of polyphenols. When no significant increase in this concentration during pressing is observed (FIG. 2), changes in the program must be made, for example by reducing the number of cage rotations or by reducing the maceration time of the grapes. On the other hand, when the polyphenol level is stable at the beginning of pressing and then increases as indicated by the evolution of the PhenOx index in FIG. 3, it is easy to separate the juices at the beginning of the plateau, at 800 mbar when an increase of 100 units is observed.

However, in everyday winemaking, optimizing press separation for each lot can be tedious, above all in large wineries. In that cases, once press cycles are optimized, measurement can be made in settling tanks only. This allows optimizing blending of juices before settling and adapting process to stabilize juice against oxidation (protection against oxygen for polyphenol-poor juices, fining or oxygenating for polyphenol-rich juices). This is what the following example shows, in a French cooperative winery where several press methods are used in parallel. Every day, many settling tanks are filled. Blending of similar quality juices and adaption of the winemaking process to the juice quality is difficult to achieve without a simple indicator of this quality. The use of linear sweep voltammetry to measure the polyphenol content of the juices in each settling tank is relevant to help decision in this situation. In previous example, over the course of a day, three qualities of juice were distinguished (Table 1):

• Low polyphenol concentration juices with PhenOx around 500. S0₂ was added before settling and inert gas was added on top of the tanks to avoid any enzymatic oxidation.
• Medium concentration juices with PhenOx around 750. These juices were fined with small doses of enological additives such as PVPP or vegetal protein.
• High polyphenol concentration juices corresponding to hard press juice with PhenOx around 900. These juices were then fined with high doses of enological products such as PVPP and charcoal.

In this example, using this technology has an economic impact thanks to :

• the adaptation of the doses of enological products to polyphenol concentrations, also allowing to avoid spoiling the wine organoleptic profile due to “over-fining”
• the recovery of additional volumes of juice of medium polyphenol concentrations which were not usually separated from the hard presses
• the early stabilization against oxidation that allows blending back wines together, increasing the volume of wines with good price and bringing back some organoleptic properties that are mainly present in press juices, such as fatness.

The measurements presented here were made on Sauvignon Blanc but many other varieties (Chardonnay, Grenache Blanc, Colombard, Gros Manseng, Muscat …) were also monitored and comparable results were obtained.

Monitoring red polyphenol extraction during traditional maceration

Monitoring extraction of polyphenols during red traditional maceration is helpful to:

• Determine the end of extraction for racking off the skins. This is of interest when winemakers need to rack off as soon as possible to liberate tanks for grapes entering the winery.
• Compare the polyphenolic concentration and composition of each tank to adapt the aging process and compare vintages (Bolkan, 2017).

Data shown here was collected on Tempranillo in Spain.

Figures 4 and 5 show the extraction of polyphenols during traditional Tempranillo maceration. The extraction has an immediate effect in first tank (Figure 4) and finishes on day 6. On the contrary, in the second tank (Figure 5), extraction starts after six days due to pre-fermentative cold soak which does not allow as fast extraction as running alcoholic fermentation.

On another Tempranillo tank, anthocyanin extraction was followed by Puissant Léon method in parallel to PolyScan measurements. A similar profile was observed (Figure 6), allowing to make decisions on a similar basis. However, EasyOx index does not allow to determine an absolute anthocyanin concentration. The measurement by linear sweep voltammetry therefore allows the winemaker to make decisions on a similar basis, while simplifying the measurement since the technique is practiced on disposable printed electrodes and does not require sample preparation.

To finish, considering the level of EasyOx and PhenOx reached at the end of alcoholic fermentation allows comparing the tanks together and to index average value that is usually observed.

In the example of Figure 7, tanks of the same grape variety range from 500 to 700 of PhenOx and from 190 to 270 of EasyOx, giving evidence of concentration differences. Moreover, tank composition shows different features: tank number 13 has a level of EasyOx that is above average and a PhenOx level at the average. On the contrary, tank number 7 has a PhenOx above average and a Easyox value close to the average. This reveals a difference in oxidizable compound types and studies are ongoing to determine if this has an impact on wine development, as tannin/anthocyanin ratio was shown to have. (Durner et al., 2015, Gambutti et al. 2017).

Besides the monitoring of Tempranillo presented in these examples, similar results have been obtained on different grape varieties such as Merlot, Cabernet Sauvignon, Syrah, Grenache, or Mourvèdre.

Conclusion

Evaluating the progress of oxidation mechanisms in must and wine is still almost unsurmountable for winemakers with traditional analytic methods. However, real-time monitoring of polyphenolics oxidation could facilitate decision making as it comes to the choice of the most adapted winemaking process for a defined quality of juice or wine to fulfill winemaker intentions. Linear sweep voltammetry on disposable printed electrodes was shown to follow extraction of polyphenols during early stages of winemaking. Application development is still ongoing to widen the use of this method to more winemaking steps.