Why I stopped packaging our coffee in one-way valves

Yesterday I decided to do away with one-way valves on our roasted coffee bags when packing with a vacuum. It’s been something that I’ve been testing for many years now and time again I’ve come to the conclusion that they really don’t make that much difference to the coffee. Indeed, too often they provide an access point for oxygen to enter the bag and degrade the coffee.

Good for sniffing the beans, letting out aromas and bugger all else.

One-way valves are supposed to let CO2 gas out of the bag while preventing O2 getting into the bag. Oxygen is the main culprit behind the staling of coffee and packaging is an attempt to extend the vitality of the beans.

Coffee is supposed to de-gas (aka out-gas) after roasting. This gas is mostly CO2 and the amount claimed in Espresso Coffee: the science of quality – edited by Andrea Illy – is up to 10L of CO2 per kg of roasted coffee. If this were the case, fresh roasted coffee that is immediately packaged (best practice to minimise oxidation) would explode the bag unless there were a gas release valve.

One of the problems with the valves, even when working correctly, is that while they let out CO2, they also let out volatile aromatics. The claim in Espresso Coffee is that it’s a direct correlation: lose 50% of the CO2 and you lose 50% of the volatiles. And we at the more discerning end of the spectrum love our volatile aromatics. However, if this prevents bags exploding in the back of the car while negotiating midday traffic, it is an unfortunate, but fair trade off.

But I’ve never had a bag explode. I have read of roasters who have and this has kept me wary of doing away with the one-way valves before proper testing. However, we have roasted and packaged hundreds of thousands of kilograms of coffee. Sometimes the valves have been faulty and not allowed gas out. In the UK, these bags usually puff up, but haven’t exploded. In Malaysia, we use a vacuum sealer and quite often the one-way valves have not prevented gas entering the bags. The leak is audible. And it is in Malaysia that I’ve decided to do away with the valves. I also contend that, although CO2 is a by-product of the roasting process, there is no 10L CO2/kg roasted coffee released. Rather, I suggest that sealing the bags without using a vacuum results in trapped oxygen that reacts with carbon compounds within the beans producing CO2 gas. Ie 1kg of roasted coffee may produce 10L of CO2 gas if exposed to oxygen, but this amount of CO2 is not within the beans.

There is a theory that vacuum packing is not really necessary as if 10L of CO2 is produced, a kilogram of coffee packed right after roasting will push out most, if not all, oxygen through the de-gassing valve. This presumes that there is indeed 10L of CO2 hiding within the cells of the coffee. If it is in fact the oxidisation of carbon compounds that produces the majority of the CO2, then one is better of using a vacuum packer to remove as much of the air as possible without damaging the beans. The bags I’ve packed with a vacuum packer only puff up if the valve is faulty. Bags without valves puff up a little.

Part of the reason that I took so long to make this decision is that medium roasted coffee when put through an espresso machine while too fresh has voluminous, albeit short-lived, crema and it is difficult to get a good balanced brew. Indeed the taste can be quite carbonated. Presumably the extreme crema is a result of excess CO2 and the coffee needs to de-gas (we usually wait at least a week) before going through the espresso machine. However, if I take a bag without a one-way valve and vacuum-pack it the beans seems to go through exactly the same process of settling down. Perhaps there is some CO2 within the beans that needs to escape. But nowhere near enough to warrant using fault-prone valves. In fact, I’ve found that the beans taste better when vacuum packed without the valves. However, I’ve yet to do this blind.

Another reason for my tardiness is that every decent roaster seems to be using them. People I respect have spent hours glueing these valves into buckets (although there might still be a case for the valves here as the buckets cannot be vacuum-packed). So, every time that I’ve come to the conclusion that the valves aren’t worth it, I’ve been too cowardly to act, thinking that it’s better to be safe than order 25000 bags and find them exploding like Elektra machines en-route to the cafes. But (fuck it) I reckon I’ve tested this enough. And we’re ordering 60000 bags so, if I’m still here in a month, you’ll know it’s gone well.