On behalf of the entire world of coffee aficionados, I have a confession to make: We have absolutely no clue at all…
On behalf of the entire world of coffee aficionados, I have a confession to make: We have absolutely no clue at all…
… how much caffeine you actually consume when you enjoy your favorite coffee drink. This question is one that quite a few people wonder about, as it plays into another common question: “Is coffee healthy?”
- Jump to Infographic: The Amount of Caffeine in Coffee, by Preparation Method
- Jump to Infographic: Caffeine Content in a Serving of Coffee, by Preparation Method
- Download the Complete Infographic
Table of Contents
- How Did We Test Caffeine Content
- Caffeine in Coffee: Preparation Methods and Brew Ratios
- The Test Results Are In: Which Kind of Coffee Has the Most Caffeine? Infographic 1: The Amount of Caffeine in Coffee, by Preparation MethodInfographic 2: Caffeine Content per Serving of Coffee, by Preparation Method
- But What About…?
When we start dealing with the question of caffeine in coffee, we inevitably also have to deal with approximate figures and percentages, and right now I should issue the disclaimer that every coffee bean is different.
There are obviously enormous variations between Arabica and Robusta beans, but you can even notice significant differences between beans of the same variety, such as an Arabica bean from India and another Arabica bean from Guatemala.
Heck, you could even pick two beans growing next to each other on the same branch, and they each might contain different amounts of caffeine — and that variation will only grow more pronounced if you roast them differently.
Despite all that, you’ll still find all kinds of claims about the amount of caffeine in a cup of drip coffee or a shot of espresso, although nobody seems to know where these numbers actually come from. Many people take those claims at face value because they sound authoritative, and because we want our blogs and news sites to have “evidence” and “proof.”
I’ll even admit that I myself have worked with these inexact and unproven numbers — at least when there was a lack of better evidence. However, that’s all over now.
I didn’t want to just use ballpark figures or back-of-the-envelope calculations to figure out how much caffeine is in different kinds of coffees. Instead, I called upon a certified test lab to calculate the exact numbers for me, you and the whole world of coffee lovers.
From what I can tell, we at Coffeeness are so far the only ones out there who have taken this beyond a mere scientific curiosity, actually putting in the effort to run these kinds of tests and then making them available for anyone to see.
To set up our caffeine test, I was allowed to commandeer the Flying Roasters café and roasting operation in Wedding, a district of Berlin, for a whole day. I used the same master batch of beans to make coffee in 15 different ways — everything from using a classic pour-over, all the way to making cold brew coffee with different steeping times.
When I told Oliver Klitsch and Dr. Nadine Heymann about my idea, they were enthusiastic about it right away. Even though they are both experienced coffee roasters, they really had no idea (nor could they) how much caffeine was actually in the beans they were dealing with every day.
Obviously, my tests can’t change the fact that caffeine content will vary from bean to bean and from brew to brew. That’s (literally) the nature of coffee.
However, the results of these tests finally gave us some exact specifications, and they also clued us in to which preparation method is the best to give you the ultimate caffeine jolt.
As you can probably imagine, everyone was placing bets about which coffee preparation method or which kind of coffee would contain the most caffeine, and everyone seemed to have different predictions going into this. Yet it seemed like everyone was right.
That’s because the results of our huge test, which was backed with the scientific support of the Institut für Produktqualität (IFP) — the Institute for Product Quality — in Adlershof, Berlin, demonstrated several things:
- The overall, absolute caffeine content and the content in relation to serving size are two completely different things.
- The two most important parameters that affected the coffee’s ultimate caffeine content were time and volume. Temperature, pressure and the amount of ground coffee used were less relevant than one might think.
- We can finally answer the question of whether a ristretto really does give you more of a caffeine kick than an espresso.
- Super-automatic espresso machines are weak sauce — or at least they make weak sauce.
Read on to find out what drink took the Caffeine Crown. We’ve also put all of the test results and tables into convenient, downloadable PDF files.
How Did We Test Caffeine Content
Even though our test kitchen is already filled with all of the equipment and coffee beans that we would have needed for this caffeine test, I still felt that a change of scenery was called for. Right away, Flying Roasters was game for a bit of experimentation and fun, so we took the coffee party over to its place.
Flying Roasters was extremely accommodating, and it lent us its roastery, gear, baristas, café and coffee beans, as well as its deep expertise. Its crew also spent a lot of time thinking about which coffee beans and roasts would be best for our test. In the end, we decided on their Pezcadito micro-batch.
The Pezcadito is a so-called “omni-roast” from a cooperative in Honduras — it’s traded directly and is certified organic. It’s called an omni-roast because it works excellently with almost every coffee preparation method.
As we were doing our testing, we were sure to taste every version of coffee that we made, and I must say that there wasn’t a single one that didn’t taste good. It tasted especially outstanding in a Chemex and as a cold brew steeped for eight hours.
Unfortunately, at the time of writing, the Pezcadito batch is currently sold out, but I’ll keep checking to see if they get more in stock in the future. I can also wholeheartedly recommend the other Flying Roasters products — it simply has awesome coffee!
We used the Pezcadito beans to make coffee 15 different ways, with each test encompassing precisely verifiable parameters (see Table 1). For each preparation method, we tested 50 milliliters in a sample jar. We also used the beans for a 16th test to use as a reference.
We then personally handed the samples over to the IFP in Adlershof, Berlin. Among its other functions, the IFP specializes in food analysis, and it makes itself available to anyone who wants to find out more about food. For example, if you want an analysis of the nutritional content of a newly developed food product, or if you want to know whether your salad contains pesticides.
We also received permission from the institute to publish the results of the tests, which apparently isn’t always a given in the world of commercial food analysis. Many companies fear their competitors and industrial espionage, which is the reason that we weren’t allowed to take any photos in the laboratory.
Caffeine in Coffee: Preparation Methods and Brewing Ratios
Below, you’ll find a table that lists which coffee test samples we submitted and how we prepared the samples. We made all of the coffees — including the cold brews, which we obviously needed to start quite a bit earlier — in a single, continuous test run, and then directly filled up the sample jars.
I also listed the grind coarseness, as well as the kinds of grinders and other gear that we used, so you can replicate our tests at home, if you’d like. For the best results, be sure to get Daniel, the barista from Flying Roasters, to join in on your tests.
We definitely relied on Daniel’s expertise, especially when it came to using the traditional espresso machines with a portafilter. He has obviously mastered his craft, and he skillfully handled the inevitable human factor that comes with doing a test like this.
Note that I haven’t specified the exact water temperature that we used. That’s obviously an important factor when it comes to brewing, but we omitted it because we based our tests on the concept of “what would the coffee drinker at home do?”
Generally, you won’t find most coffee drinkers standing in their kitchen, holding a thermometer in their water kettle. However, our electric water kettle did have a temperature display, which we did indeed use to ensure that our temperatures were consistent.
When it came to the drinks made with a super-automatic espresso machine, the variables were unavoidably inexact. That’s because our test machine, the Tchibo Esperto Caffe is — like most super-automatic espresso machines — a mostly self-enclosed device, and it reveals only a limited extent of what’s going on inside.
However, all of the other variables, such as the brew ratio and the other preparation parameters, were uniform. The laboratory even asked about this twice to verify that it was all uniform.
Table 1: Sample Selection and Brew Ratios
|Sample||Preparation Method||Brew Ratio||Grind Coarseness (Grinder Model)||Special Notes/
|KG01||Cold brew, 8 hours||100 g / 1 liter||3.5 (Mighty King Guatemala K22)||Cold, 8-hour steeping time|
|KG02||Cold brew, 24 hours||100 g / 1 liter||3.5 (Mighty King Guatemala K22)||Cold, 24-hour steeping time|
|KG03||Cold drip with ice||40 g / 0.5 liter*||3.5 (Mighty King Guatemala K22)||Made with a Dripster² using the recommended 2 drops per second|
|KG04||Cold drip without ice||40 g / 0.5 liter||3.5 (Mighty King Guatemala K22)||Made with a Dripster² using the recommended 2 drops per second|
|KG05||AeroPress||22 g / 0.15 liter||3.5 (Mighty King Guatemala K22)||Made using inverted method, undiluted: let steep 10 seconds, stirred, let stand 30 seconds, then pressed|
|KG06||Chemex||48 g / 1 liter||3.5 (Mighty King Guatemala K22)||-|
|KG07||French press||18.3 / 0.32 liter||3.5 (Mighty King Guatemala K22)||4-min. steeping time|
|KG08||American press||20 g / 0.355 liter||3.5 (Mighty King Guatemala K22)||30-sec. pre-infusion|
|KG09||Pour-over drip||30 g / 0.5 liter||3.0 (Mighty King Guatemala K22)||-|
|KG10||Espresso||18 g / 50 ml**||4.88 (CEADO E37T)||25-sec. flow time|
|KG11||Ristretto||18 g / 30 ml**||4.88 (CEADO E37T)||15-sec. flow time|
|KG12||Stove-top espresso maker||22.5 g / 300 ml||1.5 (Mighty King Guatemala K22)||Preheated the water|
|KG13||“Espresso” from super-automatic espresso machine||20 g / 30 ml||Smallest setting||Made using the Tchibo Esperto Caffe, with the “Intenso+” feature activated***|
|KG14||“Drip” coffee from super-automatic espresso machine||20 g / 200 ml||Smallest setting||Made using the Tchibo Esperto Caffe, with the “Intenso+” feature activated***; the machine made an Americano|
|KG15||Drip coffee machine||55 g / 1 liter||3.0 (Mighty King Guatemala K22)||Made using a Moccamaster machine, according to manufacturer’s instructions|
|KG16||Whole-roasted coffee beans||-||-||Flying Roasters Pezcadito, Honduras, washed; roasting date June 12, 2019; roasting level 1.5/3|
* 220 g ice cubes + 280 g water (conversion to liters, approx. 1:1)
** Due to the large sample size needed, a double shot was prepared according to the barista’s specifications.
*** If this feature was not used, then the results were much too watery to work with as an “espresso” sample from a super-automatic espresso machine.
The Test Results Are In: What Kind of Coffee Has the Most Caffeine?
Perhaps many readers will say, “Of course, it’s obvious that the ristretto was going to have the most caffeine.” However, only one person on our team (me!) bet on the ristretto. We took about 10 minutes to sort out the results, and I could just feel it in my bones that I was going to be the winner.
The ristretto was interesting because it contained by far the most caffeine, even though it’s made with less water, and despite the fact that the ground coffee is in contact with the water for less time. However, it doesn’t deliver the most caffeine at the same time.
If you wanted to get your caffeine hit all at once, you’d need to chug 4.43 ristrettos in one go. Only then would you get the same amount of caffeine as you’d get from a single glass of cold brew that had been steeped for 24 hours.
Incidentally, the rest of the team had bet on the cold brew, so I guess you could say that everyone was a winner, depending on how you look at it!
Still, although you could quibble about the serving sizes that we used for our standards, it doesn’t change the fact that nobody drinks more than a double ristretto at once. Meanwhile, it’s surprisingly very easy to down a huge glass of cold brew coffee — especially if the weather is as hot as it was when we were doing our testing.
Our analysis also revealed some other interesting findings that you might not necessarily notice at first glance.
- The different brewing methods were unequally efficient at releasing the caffeine from the coffee beans. Even our champion, the ristretto, only managed to release about 31 percent of the original caffeine. On average, all the techniques only release about 10 percent.
The two methods at the bottom of our test rankings were the “drip” coffee made with a super-automatic espresso machine and the American press. That’s noteworthy because the two methods are completely different; one immerses the ground coffee in water, while the other uses a variation on a pressurized espresso method.
Why would that be true, especially when comparing the vastly different results between an American press and a French press? The explanation is that with an American press, the ground coffee is concentrated in a small internal container, while with a French press, the grounds can float freely in the water throughout the whole press.
This means that with a French press, there’s much more contact area between the water and the ground coffee, which leads to faster and more large-scale extraction. Additionally, the contact time itself is longer.
- If you want a proper caffeine jolt in a nice, compact package, you’ve got to try a traditional espresso machine.
Otherwise, if you’ve already got a super-automatic espresso machine, you should at least set it so it uses more ground coffee to make your drinks. Admittedly, the model we used for our test, the Tchibo Esperto Caffe, is not necessarily the be-all and end-all when it comes to automatic espresso machines. We had to use the “Intenso+” feature because, without it, the coffee was much too weak and watery.
One advantage of a super-automatic espresso machine, however, is that it can make larger servings than you’ll get from a traditional espresso machine. In the end, the “drip” coffee and “espresso” settings on the automatic machine ended up containing around the same amount of caffeine, once you factor in the differing serving sizes.
- When it comes to cold brew coffee or the cold drip coffee that we made with the Dripster, the steeping time or method hardly made any difference at all.
That’s because, at some point, the coffee simply becomes saturated and reaches its maximum caffeine level. It’s still interesting to note, however, that using ice definitely slows down the reaction, or at least makes it more difficult for the beans to release their caffeine.
- We can also finally declare a winner in the classic fight between pour-over and drip coffee machines: the pour-over coffee maker wins.
That’s true even if your drip fighter is the Moccamaster, which is the excellent machine that we used in our test. The only disadvantage here of using a pour-over coffee maker like the Chemex is that, because the serving size and brew ratio were larger than they would normally be, the amount of time that the water and ground coffee were in contact was also affected.
I was also quite surprised that the AeroPress came out so far ahead in both of our analyses. I wonder if it has to do with the fact that we used the inverted method? Any physics majors are welcome to chime in with a plausible explanation.
Infographic 1: The Amount of Caffeine in Coffee, by Preparation Method
Copy the code below to embed the infographic in your own website:
<a href="https://www.coffeeness.de/en/how-much-caffeine-in-coffee/"><img style="max-width:1000px; width:100%; height:auto;" alt="" src="https://www.coffeeness.de/wp-content/uploads/2019/12/amount-caffeine-by-preparation-method.png" /></a>Used with permission from <a href="https://www.coffeeness.de/en/">Coffeeness.de</a>
Infographic 2: Caffeine Content per Serving of Coffee, by Preparation Method
Copy the code below to embed the infographic in your own website:
<a href="https://www.coffeeness.de/en/how-much-caffeine-in-coffee/"><img style="max-width:1000px; width:100%; height:auto;" alt="" src="https://www.coffeeness.de/wp-content/uploads/2019/12/caffeine-content-per-serving.png" /></a>Used with permission from <a href="https://www.coffeeness.de/en/">Coffeeness.de</a>
Table 2: Test Results
|Sample||Preparation Method||Caffeine Content|
[mg / 100 ml]
|Percentage (Rounded) of Caffeine Extracted from Ground Coffee||Caffeine Content per Serving (Average Serving Size) in mg|
|KG01||Cold brew, 8 hours||95.2||7||238.00 (250 ml)|
|KG02||Cold brew, 24 hours||112||8||280.00 (250 ml)|
|KG03||Cold drip with ice||72.9||5||182.25 (250 ml)|
|KG04||Cold drip without ice||98.6||7||246.50 (250 ml)|
|KG05||AeroPress||136||10||204.00 (150 ml)|
|KG06||Chemex||68.6||5||171.50 (250 ml)|
|KG07||French press||89.1||7||222.75 (250 ml)|
|KG08||American press||58.3||4||145.75 (250 ml)|
|KG09||Pour-over drip||74||6||185.00 (250 ml)|
|KG10||Espresso||273||21||68.25 (25 ml)|
|KG11||Ristretto||421||31||63.15 (15 ml)|
|KG12||Stove-top espresso maker||164||12||49.2 (30 ml, Angabe Bialetti)|
|KG13||“Espresso” from super-automatic espresso machine||224||17||67.2 (30 ml)|
|KG14||“Drip” coffee from super-automatic espresso machine||57.9||4||115.8 (200 ml)|
|KG15||Drip coffee machine||67.9||5||169.75 (250 ml)|
|KG16||Whole-roasted coffee beans||1319||-||-|
Our data about the caffeine content in a typical serving has some variation from the labratory’s test results, which you can download here as a PDF. The laboratory technicians used the brew ratio as a benchmark.
However, the brew ratio is somewhat different from an average serving size, as I explained above. That’s why we extrapolated the results based on the normal cup, mug or glass volume.
But What About…?
If you see these results and get the urge to grumble or point out the shortcomings of our tests — like, for example, that we haven’t tested whether an Americano contains more or less caffeine than a “drip” coffee from an automatic espresso machine, or that we relied too heavily on our barista’s standards for making an espresso — then you are surely correct in your observation.
You may also call into question the scientific soundness of our approach. Although I’m more than happy to pose in a lab coat for photo ops, I’m still definitely not a scientist. What’s more, every coffee bean is different, and like everyone else, I’m only human, and I may make mistakes or be inconsistent when using various coffee preparation methods.
However, I think that’s also why it’s perfectly OK for me to add my own voice to this topic. When it comes to the caffeine debate, there’s no definitive, final answer or key piece of evidence, and some of the voices in the debate don’t always come from a neutral place. For example, look at the conversation around chlorogenic acid in coffee.
Nevertheless, with our test results in hand (or, on screen), you now have at least a “study” that can help you better understand your daily coffee/caffeine consumption, and whether you might do well to cut back a bit on your intake. As a reference, the German government considers a single dose of 200 milligrams of caffeine to be safe and puts the ideal daily maximum limit at around 400 milligrams.
This means that if you prefer 24-hour cold brew, you might want to consider drinking a smaller glass of it, or at least taking your time to savor it. As for you ristretto lovers, you’d have to drink more than six servings of ristretto per day to hit your limit.
For everyone else, you might as well try out other kinds of coffee drinks to see which ones you like best. A little scientific experimentation never hurts!
I always love to read your comments. What surprised you about our test, and did it confirm what you always knew in your heart to be true? What did we miss or get wrong? Please feel free to leave a comment.