On December 3, 2019, I started a study to compare the performance of a Flextank vessel to that of a second-year barrel, each equipped with a demijohn-sized WineStix oak stave made from Allier French oak with a medium-plus toast level. Performance was to be assessed both quantitatively by measuring and analyzing pertinent enological parameters and qualitatively by tasting and evaluating the wine each month throughout the duration of the study. This study is expected to last a minimum of 12 months and up to 24 months.

Please refer to my December 9, 2019 blog for a full description of the study objectives, methods, and parameters to be analyzed.

SUMMARY OF CONCLUSIONS

Following are some conclusions based on test results and observations at the 6-month point of this study.

• The barrel-aged and Flextank wines have very similar chemistries except for some small analytical but imperceptible differences that do not all translate to sensory differences yet.
• Both wines have very similar sensory profiles with the Flextank being just a little bit smoother owning to the lower acidity (TA) resulting from the greater drop in tartrates. The Flextank sample also measured lower tannins which, coupled to lower acidity, will indeed feel smoother.
• The lower acidity in the Flextank sample is likely due to the significantly greater formation of tartrates although there is no obvious explanation at this time. An analysis of tartaric acid and potassium concentrations may shed some light as to why this happened. It is possible that the barrel sample still has a greater amount of protective colloids preventing further tartrates from forming.
• Clarity, measured as turbidity (NTU), is analytically – not visually – better in the barrel sample owing to its larger body and lower center of gravity compared to the Flextank; this physical aspect of barrels hastens faster clarification but, again, not visually apparent as the differences are very small.
• The Flextank sample exhibited more SO₂ binding based on measured total SO₂ although dissolved oxygen (DO) and free SO₂ analyses reveal similar behaviors in both wines. Over time with future sulfite additions, it might be expected that the Flextank wine end up with slightly higher total SO₂.
• The barrel sample has slightly higher tannins, as measured by a Total Phenolic Index (TPI), which suggests that the oak is still contributing some tannins, and higher color intensity but only as measured by spectrophotometric analysis, i.e. the higher intensity is not visually apparent as both wines have the same brilliance of red.
• Volatile acidity (VA) was greater in the barrel and increasing at a faster rate than in the Flextank sample likely due acetic acid bacteria manifesting their oxidative effects in the ullage.
• When needing to retrieve samples from the Flextank, it is much easier to work via the access port in the top lid, i.e., at the threaded port for the green ported or solid cap.

WINE ANALYSIS AND RESULTS

The table below shows measured parameters for the base wine on December 3, 2019 – what is referred to as time T0 – and for the oak-barrel and Flextank samples at the 3- and 6-month intervals, i.e., T0+3 and T0+6. The base wine is a Touriga Nacional from grapes from Lodi, CA purchased in the Fall of 2019 from Musto Wine Grape Company.

After 6 months, both wine samples are visually the same with the same sensory profile and very similar chemistries except for several small differences described below.

SENSORIAL AND VISUAL ASSESSMENTS

Both wines exhibit subtle notes of blackcurrant with just a hint of oak now detectable on the nose and palate but with still a touch of greenness likely due to underripe pyrazines. Both wines are fairly full-bodied with good body structure and light to medium color although the Flextank sample is just a tad smoother. Sample analysis and racking operations confirmed that this added smoothness is due to the  lower total acidity (TA) in the Flextank sample due to the greater extent of tartrates dropping; interestingly, there was however no change in pH. Tartrates were expected based on my assessment of tartrate stability using the Concentration Product method, which predicts stability based on the concentrations of tartaric acid, potassium and ethanol, and pH. The difference in TA between the barrel and Flextank samples is only 0.29 g/L (0.029%), but the difference in tartrates was significant. I will measure tartaric and potassium concentrations in both samples to understand better what happened; I would expect the Flextank sample to now have less tartaric acid and significantly less potassium.

The visual appearance of both samples is the same, at least to the naked eye. However, an analysis of turbidity (NTU) reveals that the barrel sample is clearing faster. This was to be expected given that the barrel has a wider body and lower center of gravity than the Flextank.

SULFUR DIOXIDE (SO₂) AND DISSOLVED OXYGEN

From an oxygen transfer perspective, it appears that the barrel and Flextank are behaving the same, at least based on measured free SO₂ (FSO2) and dissolved oxygen (DO) paramters. But upon closer investigation, if we look at total SO2 (TSO2), we can conclude that there is less SO₂ binding in the barrel sample, suggesting that the HDPE Flextank material reduces SO2 losses compared to the barrel. It would therefore be expected that wine aged in a Flextank will have higher total SO₂ over time from future sulfite additions.

COLOR AND TANNINS

From the spectrophotometric analysis, we see that the barrel sample is exhibiting a slightly redder color based on absorbance measurements at 520 nm, i.e., A₅₂₀, which gives the wine a slighter higher color intensity (IC) after 6 months, but nothing visually apparent. Hue (H), which measures the ratio of yellow color to red color for assessing browning, is however the same, which means that the barrel sample is developing a bit more brown color as would be expected in barrel wine. As mentioned above, these differences are so small that they are not visible to the naked eye.

The barrel sample also measured a higher Total Phenol Index (TPI) suggesting that the oak wood is still contributing some tannins.

VOLATILE ACIDITY

As for volatile acidity (VA), it is greater in the barrel sample and increasing at a faster rate. This is likely due to acetic acid bacteria still capable of oxidizing ethanol in the ullage due to evaporative loss in the barrel, this in spite of the barrel being topped up every two weeks. The drop in ethanol would be too small to measure to make a definitive assessment, however, a measurement of acetaldehyde (for which I’m not equipped to do) could have given some clues.

ACCESSING THE WINE IN THE FLEXTANK

In trying to minimize headspace (the space between the wine and the lid) and preventing leakage via the lid threads caused by wine expansion during aging, I have found it easier to access the wine (for sampling) via the port on the lid. I dabbed the lid gasket with a couple of drops of olive oil to ensure a super-tight fit on the Flextank. Then I placed a solid (i.e., no airlock) green cap with the threads wrapped in Teflon tape. I found the green cap to be easy to remove while allowing the wine to reach the very top.