Step 12: Primary Fermentation Phase 1
Fermenting grape juice into wine is about transforming sugar into alcohol with yeast. This happens in a complex web of chemical reactions which are not yet fully understood. The level of Brix measured in grape juice translates linearly into the percentage of alcohol the wine will have: 24 Brix in must yields around 13.5% alcohol in wine.
Fermentations can be divided into 2 successive phases:
Phase 1: Lag & Exponential growth phase. First, the yeast microbes need to adjust to the environment (temperature, pH etc.). As soon as the adjustment is complete, the yeast cells divide and grow in number exponentially while at the same time converting sugar. This takes a few days during which significant energy is dissipated in the form of heat, and rising temperatures and oxygen is consumed. Berries start to disintegrate, and skins start to float up to create a cap which needs to be broken up and submerged with regular punch-downs.
Phase 2: Stable & Exponential Decline Phase. This is when the yeast cells systematically convert sugar into alcohol. CO2 is generated, and temperatures tend to fall because the energy generated is dissipated by the fermentation tank into the environment. The cap of skins needs to be punched down regularly. The yeast cells start to die off when the sugar and other nutrition get scarce when the alcohol level gets too high, or the temperature falls too low.
A fermentation is called successful when all the sugar is consumed by the time the yeasts have died off. The opposite is a stuck fermentation when sugars are still present after the yeasts died and new yeasts need to be added to restart the fermentation. This is cumbersome and can be prevented by managing temperatures and yeast nutrients and selecting the appropriate yeast strains for the grape variety at hand and style of wine-making.
Choice of Yeasts
Different yeast strains produce different tasting wines even when applied to the same grapes. There are thousands of different yeast strains, many naturally available concurrently in the environment. So the challenge for the wine-maker is to decide on what yeast to use:
Fermentation with Indigenous Yeasts: We rely on the mix of yeast strains that happen to be attached to the berry skins brought in or survived in the winery from previous fermentations. This choice creates wines which genuinely reflect the local terroir, but there is a risk that the fermentation may not complete successfully.
Fermentation with Industrial Yeasts: We kill the indigenous yeasts with SO2 in Step #8 above and inoculate the must with a known, commercially available yeast or a yeast-derived from the own vineyard and propagated. This choice reduces the risks of stuck fermentations but adds uniformity to the wine produced.
Fermentations with both: We start with Indigenous Yeasts but then, in phase 2, introduce Industrial Yeasts to make sure the fermentation finished without a hitch.
For our first vintage (2009) we decided to go with a native fermentation to establish a benchmark for what happens without interventions. In the subsequent 6 years, 2010-15, we used commercially propagated yeasts to reduce the risks of stuck fermentations and have a record on which yeasts were actually doing the fermentation. As we gained more confidence in our ability to control the fermentation process we came back to native fermentations starting in 2016.
The most important yeast nutrient is Nitrogen which is metabolised by yeast to synthesise proteins. Nitrogen stimulates yeast multiplication, keeps yeast metabolism active, prevents H2S and mercaptan formation and stimulates aroma production. Nitrogen is provided as Yeast Assimilable Nitrogen (YAN). YAN is composed of ammonium ions and amino acids. Ammonium ions are the favourite ‘food’ of yeast. Easy and fast to use, ammonia impacts mostly yeast growth and population. Amino acids are harder to be assimilated. They represent the qualitative, healthy and delicate ‘food’ for yeast, which impacts their growth, health and efficiency through the fermentation as much as aroma production.
Berries contain YAN naturally. The optimal concentration for a healthy fermentation is between 150 and 350 mg per litre of must depending on its sugar content. The rules of thumb are:
For good population growth of yeast, we need at least 150 mg/L of YAN
For converting sugars to alcohol, we need 10 mg/L/Brix of YAN (e.g. for must with a sugar concentration of 25 Brix we need 250 mg/L of YAN
Too much YAN (>350 mg/L) increases stress conditions, produces off-flavours or leads to stuck fermentations.
Artisan winemakers prefer to minimalise the use of additives of any sort, including nutrients. We used no nutrients in 2009, then used them 2010 through 2015 as suggested by commercial yeast manufacturers. In 2016 and 2017 we used nutrients sparingly, only when fermentations showed signs of stalling. In 2018 we added significant amounts of nutrients because the level of YAN in the must was very low, below 100 for the Cabernet Sauvignon blocks.
Process Steps for Primary Fermentation Phase 1
When we use commercial yeasts and nutrients, we need to hydrate the yeast in a nutrient solution. Here are the steps we go through:
We make sure the must has an adequate concentration of nitrogen – food for the yeast. We measure YAN (Yeast Assimable Nitrogen) and adjust yeast nutrient in the next step as required.
We hydrate the required amount of additional yeast nutrient in 2 litres of 104°F tap water as specified by the supplier.
We carefully hydrate the yeast in the solution; this process is essential to ensure that the yeast cells assimilate to the environment: We add the yeast, stir gently and let the suspension stand for 20 minutes. Then we mix in 2 litres of grape juice and let the solution stand until it cools down to the temperature of the must in the fermentation tank + 15°F
We pour the acclimated solution into the fermentation tank and start the punchdowns.
When we go for native fermentation, we may set a bucket or two of crushed grapes aside a week earlier, punch it down daily and observe whether fermentation starts indigenously. The bucket is ready to be used to inoculate the main fermentation tanks when the fermentation is active (i.e. producing enough CO2 to form a 1-2 inch cap in the bucket). The alternative is to simply wait until the fermentation starts on its own; this creates a “Warm Soak” waiting period of 3-4 days.
Here is the process chart:
At inoculation, we may want to add nutrients, mainly if the YAN level measured in Step #6 above is below 120 mg/L Punchdowns in the primary fermentation tanks should start 12 hours later.
The punchdown process is:
Three times a day, take the tank cover off and blow off the Argon or CO2 blanket with a fan
Punch down the cap making sure not to crush seeds at the bottom of the tank (picture on the right). Decide whether to further increase the oxygen supply in the must. If yes, macro-oxygenate once a day during the first 3 days: Inject 10-40 ppm of pure oxygen through a diffusion stone into the must (equipment - the picture on the right). The advantage of O2 diffusion is that we can measure the amount of oxygen injected, but not the amount that bubbles up and is not absorbed.
An alternative to punchdown is delestage, but here there is no measurement at all of the oxygen supplied. Delestage requires draining the fermentation tank into a holding vessel, leaving the remaining skins exposed to air for 20 to 100 minutes and then pouring the contents of the holding vessel back over the skins into the tank. Delestage should not be repeated more than 3 times and should be followed by a punchdown at least 16 hours later. (see this article for a good description https://winemakermag.com/237-delestage-fermentation-techniques ).
Take another tasting sample and comparison taste.
Take two 2mL samples for chemical analysis once a day. Enter results into new records of the FermentationActions table.
Squeegee and wipe down inside walls of the tank with disinfectant (KMBS solution on a paper towel), cover the must with a new blanket of Argon or CO2, if the fermentation is not yet producing enough CO2 itself, and put the tank cover back on
Adjust heating or cooling to keep the temperature in 70-80 °F range
Around 3-4 days following inoculation we expect to see a peak in the level of Free Anthocyanins (hopefully above 1,000) and sugar levels having dropped 1/3rd in Brix. At this point, we move on to Fermentation Phase 2.
Dealing with Fermentation Problems
A long lag phase or abrupt stop in the conversion of sugars to alcohol indicate a problem. An abrupt stop in fermentation activity can happen as a consequence of a severe temperature drop – no longer an issue for us since we can control the temperature in our fermentation tank. A problematic delay in the onset of fermentation activity is indicated when the lag phase is longer than 5 days. This can happen when:
A native fermentation is attempted with indigenous yeasts. It may help to raise the temperature, but it is safer to switch to inoculation with industrial yeast instead.
The yeast used for inoculation did not develop properly. This can be confirmed by counting the density of viable yeast cells under a microscope. It should have reached 10 to 100 million cells / mL - an analysis better left to a commercial lab (e.g. Enartis). The remedy is to re-inoculate
There is a nutrient deficiency as indicated by low YAN levels relative to the Brix level of the must. The remedy is to add more yeast nutrition.
There are toxins or spoilage microbes in the must. This can be confirmed by lab analysis of the must revealing excess SO2, pesticides, copper or iron residuals or spoilage microbes. If the analysis indicates Lactic Acid Bacteria as spoilage microbes, then the must should be treated with Lysozyme and SO2. If the analysis indicates non-microbial toxins then fining is recommended with Bentonite, yeast hulls or an industrial product like Enartis Cellferm.
When restarting a fermentation, it is advisable to use a special yeast which ferments vigorously and can adapt to high alcohol, high volatile acidity and has low nutrition needs.
Data management is identical to what we described in Steps 3-11, with one exception. We measure the chemical properties twice, using both the “Must” and the “MUF” settings on the OenoFoss equipment. This is because the measurements for MUF (Must Under Fermentation) are not calibrated and need to be interpolated to calibrated Must-measurements.The following screenshot shows the “Juice Analysis” tab for October 19, 2017, at 10 am.
Tracking Results for 2018
We made the following choices in 2018.
We had 4 fermentation batches, two for the Merlot-CabFranc mixes, one large one for the Cabernet Sauvignon must and 1 small one for the Petit Verdot must. We waited 2-4 days until the fermentations started on their own, but had to add significant amounts of nutrition 500 -1500 ppm) to compensate for the low YAN levels (85 – 175 ppm). We used Nutriferm from Enartis and Microessentials from Gusmer. We reached one third sugar depletion in 2-3 days before the Anthocyanins peaked. We occasionally heated the must, so temperatures stayed in the 75 – 82 dF range. We infused between 15 – 27 ppm of pure oxygen.
We will review the results at the end of Step 17..