Pressurised Cold Brewing

The two brewers

I read quite a lot of blogs. Actually that is probably a half truth. I subscribe to a huge number of blogs and often skip through postings that don’t immediately grab me. This one I read, and it lead me to this post.

I’m sure it is no leap to see where I am going with this, and also clear that I am not claiming this as original thinking!

I began to wonder if pressurised brewing could be applicable to more than espresso. Cold brewing (not icing hot coffee to dilute, but brewing with ambient/cool/cold water) is often a little disappointing. While you can get a balanced and tasty brew you sometimes miss out on some of the interesting aromatics and flavours, as well as the acidity, of some coffees – due to a lack of heat/energy to extract them.

Could pressure add the necessary energy – perhaps even allow rapid cold brewing?  I loaded a syphon with 15g of coffee and 250ml of water, pressurised it, shook vigorously and then waited a minute and released the pressure.  I strained it through a V60 and tasted it.  Disgusting, barely any extraction.

“More time….” I thought – so I set up a longer brew.  The results were very tasty.  I needed a control sample, so last night I experimented again.

The two brewers

I would use the same amount of coffee, water and use the same brew time.  One brew would be in a french press, left to steep.  The other would be pressurised.

Matching doses within 0.1g

For the coffee I used a test batch of the washed lot from Finca Killimanjaro, from Aida in El Salvador.  We’d already cupped it and I knew it to be juicy, tasty and not an obvious coffee to (traditionally) cold brew.

There is a lot less bloom with cold water.

I steeped them for 12 hours – the french press lidded and two charges into the cream whipper.  (If anyone knows how to calculate how much pressure the liquid was under I’d be very grateful!)

This morning I strained the two coffees in preparation for serving to the rest of the team at the roastery.  The french press brew looked fairly normal:

Straining the press pot

Straining the pressurised liquid was very different.  Overnight the gas had gone into solution and releasing the pressure meant that it began to bubble and fizz out (this is important – more on this later!).  It doesn’t look very appealing to pour a mass of fizzing coffee slurry into a V60!

Fizzing coffee slurry

The coffee also looks a little odd initially as I think it was still giving up a little of the dissolved gas.  Very quickly it just looked like paper filter coffee again.

Pressurised cold brew coffee

I served the two coffees blind to everyone, though it was clear there was an obvious winner.  Everyone picked the pressurised brew as being more delicious.  I checked the extractions with the Mojo and the preference made sense.  The ambient pressure cold brew struck out at a lowly 14% extraction.  The pressurised brew (with identical time, brew temp and grind size) came out at 18%.  This was a pretty significant change to the brewing process.

Now – I know what you are thinking:  Is the pressure speeding up the extraction.  My initial thoughts are:  probably not.

I think that the aggressive fizzing upon depressurisation and during the pouring process was effectively and noticeably agitating the coffee grounds increasing the extraction.

What I’d like to try next is to compare an ambient cold brew that gets up to 18-19% extraction (into the zone of deliciousness) to a pressurised brew of equal extraction.  I’ll probably need to agitate the coffee in the ambient brew to get it to do that without introducing a new variable between them of brew time.  This should also give a slightly clearer impression of the role of pressure outside of its agitating effect.

The coffee:  pretty damned tasty – and some nice acidity/juiciness in the cup which makes me hopeful.  We did it with one of the Kenyas we have (the Tegu AA lot) and it was seriously delicious.  If any of you have a cream whipper and some spare time and want to experiment too I’d love to hear about it.  Thoughts, suggestions and gentle berating for silly experiments always welcome in the comments!

35 Comments

  1. I wonder if you ran an experiment utilizing straight pressurization against a high agitation in a regular cold brew steep.

    Logic says that the agitation will up the extraction rate anyway, so I wonder if it would be a more controlled way of upping it rather than the pressure, which may be hard to replicate.

    In any case, I’m about to write up an article about doing a hot bloom for cold brew, something that Jesse Kahn recommended.

  2. Hot bloom sounds interesting – might give that a try.

    As I said in the penultimate paragraph – I do need to run another experiment to try and rule out agitation a little bit. Might do it on Monday!

  3. I am new to the cold brew game. I’ve tried cold brewing in a french press- double the extraction time of regular, up to 10 minutes total- but was not impressed. I guess, judging from your experiment here, that much more time is needed? I wonder, would this cold brewing have useful application in an AeroPress? I think I’ll try it. Any suggestions as to brew time, grind size, etc. would be appreciated. Great post. Thought provoking.

  4. James-
    Something I’m not clear on – you mentioned a “fizzing” – were you using CO2 or nitrogen cartridges for the whipper?

    And it seems that your concern about agitation is the most plausible explanation for the difference as most of the solid materials in the press would have settled to the bottom after 12 hours. I’m presuming that a stirring of the pressed coffee would result in similar Mojo readings since the processes are essentially the same.

    Also, did you utilize the same grind for both samples?

    Is the test here to see if the pressure will somehow shorten the brewing time?

  5. I am, as usual, inspired by your innovative work. In this case, however, I have to say I am most pleased by the all-too-rare proper use of “penultimate” in your comment. I mean, it’s not every day “penultimate” gets thrown around, and correctly to boot!

  6. I think that it is the pressure speeding up the extraction… the technique is called ‘Nitrogen Cavitation’.

    We are playing around with it to infuse flavours into spirits quickly (we’re talking 5 minutes) and for making flavoured syrups. What’s great is that it’s cold and so you don’t ‘stew’ things like fresh fruit during the infusion period.

    My understanding is that the high pressure forces the liquid (water in this case) into the cavities of whatever you’re infusing, coffee in this case. Agitating will probably speed it up even more, but when you release the pressure everything rushes out and the inside of the ISI gun returns to ambient pressure.

    Infusion is complete.

  7. @Jay I used nitrogen. The grind was the same for both samples.

    The test was to see what pressure did to cold brewing. Unfortunately this experiment didn’t sufficiently isolate pressure and introduced agitation so I will have to go again…

  8. I wonder if the gases inside the grounds keep prevent extraction but pressurization helps to overcome this resistance and get more water into them.

  9. interesting idea. i have been experimenting with a similar idea brewing in a vacuum (not a vac-pot). the idea is the same as marinating meat in a vacuum; forcing any gases out and replacing the cavities with water, creating better infusion before extraction. good results so far, and much shorter brew time, as an added bonus.

  10. I am no chemist but I wonder if using a pressurized gas that’s less soluble than the coffee chemicals would prevent the slurry becoming fizzy and allow an easier measure on whether or not pressure or agitation is causing the increased extraction.

    I would, however, think that the pressure is causing more extraction than the agitation as the agitation is happening in a rather small window as opposed to the pressure being over a much longer time frame.

    Perhaps it would be worth pressurizing a canister with brew with a lot of gas forcing the liquid to take on the gas much quicker, providing nitrogenated slurry without the long steep, as a test to the effect?

  11. Wow. Very interesting and inspiring idea. Can’t wait to try this. If anyone could expand on the mecahnics/practicalities of how to introduce pressure into a cold brew system (without specialised equipment), I would be most grateful!

  12. What was the ratio for this? I’ll try and replicate this experiment tomorrow.
    Also did you refrigerate the brews or just do it room temp?

  13. James-
    I don’t know if the induced agitation is necessarily a bad thing – you did say that the coffee tasted great and isn’t that the idea: achieving great tasting coffee?

    I’m interested to see how your experiments progress. I’ll try to make some time this week to do some alternate testing. We regularly brew cold drip tower iced coffee and I’ll see what, if any affect, NO pressure might have on the brew results.

    I’ve been theorizing on sous vide brew coffee but these comments cause me to wonder the possibilities of cold brewing under vacuum – and I’ve got a chamber vac at the ready…

  14. @Collin – the dose was 15g to 250ml of water.
    Ground roughly where a 2 V60 would be for how we brew them, which isn’t that useful to yoU!

    @Jay – I don’t think the agitation is a bad thing either – just an unexpected part of the experiment.

  15. You haven’t really given enough information to compute the pressure, so I’ll just make up what’s missing, and probably embarrass myself by simultaneously failing to remember high school chemistry.

    A mole of gas has a volume of 24.3L at 20°C and atmospheric pressure. You say the gas was nitrogen (N₂), but I thought cream whippers usually used nitrous oxide (N₂O) or carbon dioxide (CO₂). I’ll go with N₂O. A mole of that weighs 2 × 14 + 16 = 44 grams. You said two charges were placed in the whipper. A bit of googling says that a charge is usually 8 grams. At one atmosphere, the volume would be (16g / 44g) × 24.3L = 8.8L. Assuming the whipper is 500mL, of which 250mL is taken up by coffee, the partial pressure of N₂O is 8.8 / (1 / 4) = 35 atmospheres. Adding in the 1 atmosphere already in the canister, you get 36 atmospheres. This seems about 4x too high, so something probably went wrong. I could go on to compute the pressure after the N₂O dissolves in the water/coffee, but I’ll wait for confirmation that the above numbers aren’t completely bogus first.

  16. does the entire solution become permeated by the pressurised nitrous?

  17. Some of the nitrous dissolves in the coffee. With such a small amount of liquid, it’s safe to assume that the amount dissolved is the same throughout, not, say, more at the bottom than top.

    The only number I could find for solubility 0.15g/100mL at 15°C (not 20°C). In our 250mL, we could dissolve 0.375g. Solubility is proportional to the partial pressure of the gas, so at 35 atmospheres, 35 * 0.375 ≅ 13g could be dissolved. Of course, then there would only be 3g left as a gas, so the pressure wouldn’t be 35 atmospheres anymore.

    Cutting to the chase by skipping the algebra, at equilibrium, 7.2g would be dissolved, and 8.8g would be a gas. The pressure would be (8.8g/16g) × 35atm = 19.3 atmospheres. Double checking, 0.375g × (19.3atm/1atm) = 7.2g dissolved. This could be checked by weighing the solution as the canister is opened, and as it outgases.

    I have no idea how long it takes to reach equilibrium.

  18. Just got my A-level results on Thrusday was an A in maths and a B in chemistry so I will give the maths ago, jesse you may have got it correct, they don’t seam to teach much on atmospheres these days, but I will just use metric.

    You are correct pretty much all N20 chargers are 8g all the ones I used for cream where. I will assume for now that it doesn’t dissolve and that again 250cm3 is taken up by coffee and water. The chargers are 10cm3 so to be accurate we will use that in the maths.

    When the gas charges into the whipper some will be left in the charger, so the actual amount in the whipper after the first charge is

    8 x (250/260) = 7.69g (3 sf)

    On the second charge its

    (7.69+8) x (250/260) = 15.1g (3 sf)

    Mr of N20 = 44

    n = m/Mr = 15.1 / 44 = 0.343 mol

    Using the Idea Gas Equation

    Pv = nRT

    R = 8.31 (3 sf)
    V = 250cm3 = 0.00025 m3
    T = 20C = 293K
    n = 0.343 mol

    Rearrange to make Pressure the subject

    P = nRT/V = 0.343×8.31×293/0.00025 = 3.34×10^6 Pa

    3.34×10^6 Pa = 3340 Kpa = 484 Psi = 33.4 Bar = 33 Atm

    We don’t need to worry about vapour pressure here, when the N2O gas gets to 5150 Kpa (@ 20C) it starts to condense into a liquid, so theoretically there will be a limit to the maximum pressure you will be able to charge your whipper to, one more of those 8g chargers and it won’t increase in pressure any more

    So I think Jesse’s Maths was perfectly fine.

    That is about 3 to 4 times higher than that of espresso extraction.

    From my lessons on equilibrium chemistry, this probably all has to with le chatelier’s principle. Where by if you change the pressure or change the temperature the equilibrium shifts to contract this change.

    For example

    2A + 2B C + D Enthalpy change +100Kj mol-1

    In this situation there are 4 moles of gas of reactants and 2 moles of gas of products if we increase the pressure the equilibrium moves towards the products to try and decrease the pressure. Also if we increase the temperature because the forward reaction is endothermic The equilibrium again moves towards the products to try and decrease the temperature. This is in fact probably a good example if any equilibrium reaction is taking place of the extraction of coffee.

    Pressure is analogues for concentration if we are talking about solutions, coffee is a solution not a gas, so increasing pressure won’t increase the concentration.

    I’m not so sure why it would have tasted better, blind fold or not. With espresso extraction we need the high pressure so that the water is forced quickly through the fine grounds, but from what I can understand I don’t know how it actually effects the taste directly other than the fact that it causes the water to actually be able to get through.

    Please expand onto what the pressure is actually doing, my chemistry A-level unfortunately didn’t go this in depth into coffee science.

  19. I hope your able to read my dyslexic mistakes and in the equilibrium example my arrows got removed.

  20. Nitrogen Cavitation (cells are placed under high nitrogen pressure, then when the pressure is suddenly released they burst as nitrogen bubbles form in them, typically used to prepare samples for fractionation iirc) would explain the effect with fruit, but are there really analogues to the cells to burst open to release aromatics in the coffee’s case?

  21. josh (et al),

    in the middle of musing about the effects pressure during brewing on taste, i came across this in gimme! coffee’s training manual:

    “Standard drip brewing relies on gravity to extract flavor from the bean and exposes only those flavor oils just under the surface of the grounds to the brewing water. Espresso brewing forces water in to the interior cell wall of the tightly packed grounds, extracting by pressure what gravity alone can’t: emulsified oils, sugars and lipids that give espresso the intense and complex flavor profile for which it is renowned.”

    i suppose i’d always sort of assumed this but never thought much about it or had it verified (although, it’s kind of a no brainer, given the effects of pressure profiling and such).

    perhaps, since it’s still a cold brew, rather than speeding up the extraction process too terribly much, the added pressure is simply increasing the effectiveness of the extraction, thereby bringing out more and more varied aromatics and oils.

    this of course makes me wonder what would happen if you did a pressurized brew with the added energy of heat…but then i’m guessing all that pressure would lower the temp?

  22. @JL Salgado

    There is no way the internal pressure of the canister is going to ever reach 72.45 bar. That’s 1051 PSI. Think about it – the contents of the N2O cartridge is allowing to expand inside the canister – the pressure inside the cartridge is much higher than the pressure inside the canister. If N2O was a shock sensitive explosive inside a cartridge, they would never be shipped.

  23. As with others, I’m not sure how good my high school chemistry is but…

    Couldn’t it just be the temperature that is increasing the extraction quality? By pressurizing it, we are necessarily heating it up right? (PV=nRT, only pressure and temperature can change).

    Once jesse figures out the math we could calculate the temperature difference.

  24. Definitely something i’ll try at home, feel free to post more experiments like this :-)

  25. Incidently, if freshly roasted coffee beans are pressurised under nitrogen to about 9-10 atmospheres and left 48 hours, teh flavour molecules are forced into the coffee oils and after SLOW DEPRESSURISATION, the “freshness” of the coffee is prolonged… well..  at least to several years.
    STEVE

Submit a comment