Guarana and yerba mate: caffeine in different packaging?

The can says guarana. The tea tin says mate. Both promise "natural energy" over a leafy picture, and both quietly hand you the same famous molecule: caffeine. Before either goes into your routine, it helps to know that they are not twins. Guarana is a seed with a high (and very variable) caffeine percentage; yerba mate is a leaf you steep in hot water, with less caffeine per gram and a few safety questions of its own.

Put plainly, both are caffeine sources first. What you actually swallow depends on the product, the extract and the serving, not on the plant's name. The extra compounds in each plant are real, and the one trial that compared guarana with matched caffeine found the two performing alike (details below).

Guarana: a seed with a high caffeine percentage

Guarana (Paullinia cupana) is an Amazonian plant whose seeds are ground into powder or turned into extracts for capsules, tablets and soft drinks. A 2013 review in the Journal of Ethnopharmacology describes the seeds as having up to 6% caffeine and notes that most of the crop goes to the soft and energy drink industry, with the rest sold as powder for capsules or mixing in water.

"Up to 6%" is a ceiling, not a label. A University College London trial that measured its own guarana powder by HPLC found 3.99% caffeine. A gaming trial that used 500 mg of a guarana ingredient reported 40-44 mg of caffeine in that portion, which works out at roughly 8-9% for that particular material. Different materials, different percentages. That is the whole story of guarana labelling in two data points.

What else is in the seed

Researchers usually list theobromine and theophylline (two chemical relatives of caffeine), tannins, saponins and catechins as the supporting cast. The idea that these work alongside caffeine is, in the words of the UCL trial's authors, something that is speculated, not a demonstrated mechanism.

A measurement of one commercial guarana supplement shows the proportions. A 2009 Japanese analysis found 139 mg caffeine per tablet against 5.48 mg theobromine and 4.45 mg theophylline. That is one product tested once, so treat it as an illustration of the ballpark rather than a rule. The authors also warned that taking four tablets as directed could mean overdosing, which is a good reminder to do the multiplication on any label.

What the human trials tested

• Haskell and colleagues, 2007. In a double-blind, counterbalanced, placebo-controlled study, 26 adults received 37.5, 75, 150 or 300 mg of a standardised guarana extract. The extract improved secondary memory performance and alert and content mood ratings, and the two lower doses gave better cognitive results than the higher ones. The authors concluded the effects could not be attributed to caffeine alone, though one small study cannot say what else was responsible.
• Gurney and colleagues, 2023. Twenty-five physically active adults (eighteen men, seven women, average habitual caffeine intake 122 mg a day) took guarana, caffeine at 5 mg per kg, or placebo before reaction-time tests and a maximal cycling test. The guarana dose was calculated to match the caffeine in the caffeine condition. Guarana beat placebo on some reaction-time and alertness measures, while caffeine did not reach a significant difference from placebo on them. But guarana and the matched caffeine dose did not differ significantly from each other, so the trial cannot say guarana did anything the caffeine did not. The authors reported no funding and no conflict of interest.
• Leonard and colleagues, 2023. A trial in 61 experienced gamers tested a microalgae extract combined with guarana, so it cannot tell us what guarana did on its own. The sponsor was the microalgae company; its researchers are co-authors, and the paper says they did not collect or analyse the data. It shows why a "guarana plus something" result is not evidence for guarana.

Put together: guarana can behave like the caffeine it contains in short, controlled tests, and the one trial that compared it with matched caffeine found the two performing alike. The popular claim that guarana's caffeine is released more slowly or lasts longer than the same amount of plain caffeine has no human trial behind it in the sources linked here. If a product page says so, it is a claim, not a result.

Yerba mate: a leaf you steep

Yerba mate is the dried leaf of Ilex paraguariensis, traditionally drunk from a gourd in South America and increasingly sold in Europe as loose leaf, tea bags and bottled drinks. IARC describes mate as an infusion of the dried leaves, traditionally drunk very hot but also drunk warm or cold.

A 2026 narrative review in Nutrients gathers composition figures for the dry leaf:

• Constituent: Caffeine; Reported range in the dry leaf: 8.7-17.5 mg per gram (about 0.9-1.75%); Notes: Main methylxanthine
• Constituent: Theobromine; Reported range in the dry leaf: 0.5-1.5 mg per gram; Notes: Much smaller than caffeine
• Constituent: Chlorogenic acid; Reported range in the dry leaf: 28.2-44.5 mg per gram; Notes: Plus other caffeoylquinic acids
• Constituent: Saponins; Reported range in the dry leaf: Present, not consistently quantified; Notes: The review says standardisation is limited

Ranges differ by cultivar, leaf-to-stem ratio and processing, and they describe dry leaf, not a brewed cup.

What the human trials tested

The same review says controlled human studies on cognition are currently insufficient, and that much of the evidence for mate is preclinical, with human studies limited, short and varied in dose and formulation. The nearest thing to a focus-relevant test is a 2026 crossover study in 29 adults in which two commercial brewed mate brands did not change energy use or heart rate on average during cycling, compared with water or a 135 mg caffeine drink. That is an exercise study, not a focus test, but it fits the general picture: mate is a pleasant caffeinated drink with plenty of interesting chemistry and not much proof of anything extra.

Doing the sums: from "500 mg of guarana" to milligrams of caffeine

The word "guarana" on a label is not a caffeine amount. Here is how far apart the examples above sit once you do the arithmetic (ours, from the cited sources, not product claims):

• Example: Guarana powder in the UCL trial; What the source states: 3.99% caffeine; Caffeine implied: about 20 mg per 500 mg of powder
• Example: Guarana ingredient in the gaming trial; What the source states: 500 mg portion, caffeine reported by the study; Caffeine implied: 40-44 mg per 500 mg
• Example: Yerba mate dry leaf, review range; What the source states: 8.7-17.5 mg per gram; Caffeine implied: about 9-18 mg per gram of leaf, before brewing

Two more questions matter as much as the percentage: how many capsules, tablets or grams make a serving, and how many servings a day the label allows. A product that says "guarana extract" with no caffeine figure leaves you doing detective work; one that states caffeine per serving is easier to add to your day's total. Our guide to reading supplement labels covers the general habit.

For drinks, the Finnish Food Authority summarises the EU rule in Regulation 1169/2011 (Annex III, point 4): a drink containing caffeine, whatever its source (guarana included), above 150 mg per litre, must show "High caffeine content" with the amount per 100 ml and a note that it is not recommended for children or for pregnant or breast-feeding women, while coffee- or tea-based drinks that name "coffee" or "tea" are exempt. For scale, the Dutch Voedingscentrum puts a 250 ml energy drink can at about 80 mg of caffeine.

Caffeine limits, pregnancy and sleep

Whichever plant it comes from, the caffeine adds up with your coffee and tea. The Dutch Voedingscentrum's current advice (read on 8 October 2026) is no more than 400 mg of caffeine a day for healthy adults, and no more than 200 mg at one time. For pregnancy it gives the Dutch Health Council's advice of no more than 200 mg a day, and it gives lower figures for children. Those are national guidance figures for the general population, not a prescription for any one person; people with heart rhythm problems, anxiety or other conditions may be told something different by their own doctor.

Sleep is the cost most people notice first. A 2023 systematic review and meta-analysis of 24 studies found that caffeine reduced total sleep time by about 45 minutes and sleep efficiency by about 7% on average. The authors estimated that a coffee (107 mg per 250 ml) should be finished at least 8.8 hours before bedtime, and a pre-workout serving (217.5 mg) at least 13.2 hours before, to avoid losing sleep time. Those are pooled estimates, not a personal cut-off, which is worth remembering before a late session on a "natural" energy drink. Our sleep hygiene guide has the wider context.

If you take prescription medication, particularly for the heart, thyroid or mood, or you are pregnant, check caffeine-containing supplements with a pharmacist or doctor first. Our supplement and drug interaction guide explains the general approach.

The mate cancer headline, read properly

Yerba mate has a reputation for a cancer link, and the reality is more specific than the headline.

In 2016 an IARC Working Group evaluated coffee, mate and very hot beverages. Its press release and Q&A say:

• Drinking very hot beverages, defined as above 65 °C, was classified as probably carcinogenic to humans (Group 2A) for cancer of the oesophagus, on limited human evidence.
• Drinking mate that is not very hot was not classifiable as to its carcinogenicity to humans (Group 3), because the evidence was inadequate. That means "we can't tell", not "proved safe".
• There is no separate evaluation of very hot mate; it falls under the very hot beverage evaluation.
• In a pooled analysis of studies with about 1,400 oesophageal cancer cases, risk rose with temperature independently of how much mate people drank. Very hot mate showed a statistically significant increase; warm mate did not. In rats and mice, very hot water at 65 °C promoted oesophageal tumours and mate did not.
• A classification describes how strong the evidence is that something can cause cancer, not how large the risk is.

So the headline is mostly about temperature. The same evaluation covers tea drunk at scalding heat.

Smoke is a separate question, and the IARC Q&A does not discuss it. Some mate leaf is dried over smoke, which can leave polycyclic aromatic hydrocarbons (PAHs), a group that includes known carcinogens. A 2019 review of ten studies found that nearly all measured very high levels in commercial dry mate, with benzo[a]pyrene mostly above 25 ng per gram and up to 600 ng per gram in some samples, while the share transferring into the drink varied from 1% to 50% depending on the method. A 2018 study of 244 adults in Brazil found that urinary PAH metabolites rose with recent mate intake, in some cases as much as for smokers, and the authors suggested the PAH content may contribute to the higher oesophageal cancer risk seen in mate drinkers. A 2021 risk assessment in Uruguay found significant differences between brands and concluded that heavy drinkers of mate with high benzo[a]pyrene content are the ones at risk.

Reading those together: heat is the best-supported factor, and smoke-derived PAHs are a plausible additional one, mostly discussed for heavy long-term drinkers in South America. The research does not show that an occasional mug of non-scalding mate carries a measurable risk, and it does not tell us which products on a shop shelf are low in PAHs. A label that says how the leaf was dried is the useful detail to look for.

Rules in the Netherlands and EU (read 8 October 2026)

Laws change, so check current rules before relying on any of this.

• Dutch herbal preparations decree. The annex to the Warenwetbesluit Kruidenpreparaten (version in force from 1 July 2022) lists plants and fungi that herbal preparations may not contain under Article 4(2); it names kava and yohimbe, for example. Paullinia cupana and Ilex paraguariensis are not named there. That is not an approval: not being on a list of banned ingredients says nothing about whether a product is allowed on other grounds.
• EU herbal medicine monographs. The European Medicines Agency has a monograph for guarana seed (published 25 March 2013) covering relief of symptoms of fatigue and weakness, for adults only, and advising against use in people with stomach and gut ulcers, hyperthyroidism or cardiovascular problems such as hypertension and arrhythmias. It also has a monograph for mate leaf (published 28 January 2011) with indications listed as fatigue and weakness and urinary tract and genital disorders. These are herbal medicine documents, not food rules and not clinical trials.
• Novel food status. The EU novel-food status of guarana and yerba mate isn't covered here; the Commission's Novel Food Status Catalogue is the place to check it before buying.

Seed, leaf, and what the label owes you

Guarana and yerba mate are worth knowing as ingredients, but neither is a secret upgrade to caffeine. Guarana's caffeine content varies a lot by material, and the human trials are small. Yerba mate is a milder, more social drink with little focus research and a temperature (and possibly smoke) caveat. Both count towards your daily caffeine, and both deserve a look at the serving size before the marketing copy.

If this is part of a wider question about focus or energy, Smart Supplements 101 and the label guide above are good places to carry on.

Written by Smart Supplements with AI assistance from the sources linked in the text. It has not been reviewed by a clinician.

Last updated: 12 October 2026

This article is for informational purposes only and is not intended as medical advice. Always consult a healthcare professional before starting any new supplement, especially if you take prescription medication.