Coffee Data Science

Can Vacuum Jars Remove Gas from Coffee Beans Faster?

A small study in taste and extraction

Robert McKeon Aloe
Towards Data Science

--

A few months ago, I read and then summarized an excellent research piece on how quickly or rather slowly, coffee beans degas. I thought, maybe a vacuum could accelerate degassing. I had trouble pulling an espresso shot within the first two weeks of roasting, and I attributed that mostly to CO2.

All images in this article are by the author

Coffee is usually shipped in vacuum sealed bags, and the amount of space inside the bag is relatively low. I thought I could accelerate degassing by using a vacuum jar where half the space would be empty. The hope would be that I could use the beans sooner rather than wait a few weeks.

The result: it didn’t help speed up degassing, but it did improve taste as I showed in some data I collected. However, the data for how quickly degassing occurred is still interesting even though I’m using cheap hardware to measure the weight. I was still able to see that vacuum jars degas only slightly faster than sealed jars.

Image by author from previous publication: https://link.medium.com/d6dmEwaQhbb

The other result of this study is that since the off-gassing between regular jars and vacuum jars is roughly the same, the taste improvement between vacuum jarred beans and non-vacuum jarred beans is probably more related to oxidation than any other variable

Degassing Data

After each roast, I split the roast into two containers (aside from the first roast of Ethiopian and El Salvador beans, in the first two columns). I weighted the beans before they went into the jar, and I weighted the jar with the beans and lid. The intention was to be able to quickly weigh the beans multiple times a day. For each roast, I was roasting two beans types separately, so each bean was usually paired during a shot.

I did a quick test to understand how the scale was affected by placement. As a result, I aimed for center placement to stay consistent.

The measurement was using a scale that had a +/- 0.03g accuracy. Based on the previous degassing work, the difference after a few days should be around 0.20g per 100g of coffee or 2 mg/g. This is around what I found, but the results between the days was a little noisy.

All images by author

It’s easier to summarize all the data for Jars vs Vacuum Jars. In the final comparison, Vacuum don’t seem to improve degassing.

Even looking at the data over time, it seems the amount of weight loss is pretty similar.

This one shows some of the issue with the scale’s precision because the loss goes negative which isn’t possible. Often, the time of day affected the measurement due to slight temperature changes of a few degrees in my house throughout the day.

We can look at the cumulative effect across time, and I did not do a great job on weighting the coffee each day past 5 days, so it shows the Jar average goes higher than the Vacuum. This average also assumes the amount of degassing for the beans not measured past 5 days stays constant. I think this could simply be part of the noise because a difference of 0.4 mg/g is around 0.04g. 0.04g is very close to the noise of the scale (0.03g).

Overall, these results suggest that vacuum sealing doesn’t make a a big difference in off-gassing as suggested by extraction times or the ability to pull a good shot before two weeks, but a previous study, using these beans, has shown that it makes a taste improvement. So that degradation in taste must be caused by the oxidation of the coffee beans rather than the off-gassing of CO2 and flavors.

--

--

No responses yet