To bloom or not to bloom? (Part 1)

Experimentally measure the effect of the coffee bloom on dissolved CO₂ in coffee brewed using the AeroPress inverted method.

What is coffee bloom?

During the coffee roasting process, CO₂ is generated and partially released through macropores formed within the roasted coffee beans. A correlation exists between the degree of roast and both the released and residual CO₂ (1). A portion of the residual CO₂ is believed to remain trapped within the porous matrix of the beans as occluded gas, in equilibrium with CO₂ that is absorbed or adsorbed in oils, polysaccharides, and moisture (2).

After roasting, the beans gradually release these gases in a process known as “degassing,” which can continue for several weeks. When hot water comes into contact with the coffee grounds, it displaces and rapidly releases the retained CO₂, producing the phenomenon known as the “bloom.”

Why is bloom needed?

It is widely hypothesized that the bloom phase is essential, as insufficient release of CO₂ may hinder effective contact between water and the ground coffee. Limited wetting of the grounds can result in suboptimal extraction and a weaker brew. Additionally, it is proposed that if CO₂ is not adequately released, it can dissolve in the brewing water to form carbonic acid. This compound is detected by sour-sensitive taste receptors on the tongue, contributing to a sharp, acidic, and slightly effervescent sensation. In aqueous solution, dissolved CO₂ exists in equilibrium with carbonic acid (3):

Hypothetically, if the coffee is not allowed to bloom, a greater amount of CO₂ may dissolve into the brewing water, leading to the formation of carbonic acid and its dissociation products, including bicarbonate and carbonate ions.

A higher concentration of carbonic acid may negatively affect the taste of the coffee.

The present study is based on these two hypotheses and aims to evaluate them experimentally. However, quantifying CO₂ in brewed coffee presents a significant analytical challenge, which may explain the limited availability of scientific literature on this topic. Accurate measurement of dissolved CO₂—present as carbonic acid, bicarbonate, and carbonate ions—could provide a clearer answer to the question of how critical the bloom step is in coffee brewing.

Such an experiment would help determine whether CO₂ is effectively removed during the bloom phase when using different brewing techniques, such as the AeroPress, V60, or Chemex. In this context, one potential approach is to employ a Total Organic Carbon (TOC) analyzer as an indirect method for assessing dissolved CO₂ in the brewed coffee.

What is TOC?

Total organic carbon (TOC) is a measure of the carbon content of dissolved and undissolved organic matter present in various matrices. In this study, the matrix of interest is water. To properly interpret TOC measurements, it is necessary to distinguish between three related parameters:

Total carbon (TC) – the sum of all carbon present in the sample, including both organic and inorganic forms, as well as elemental carbon.

Total inorganic carbon (TIC) – the fraction of carbon present in inorganic forms, primarily including dissolved carbon dioxide (CO₂), carbonic acid, bicarbonate (HCO₃⁻), and carbonate (CO₃²⁻). While other inorganic carbon species such as carbon monoxide (CO), cyanide (CN⁻), cyanate (NCO⁻), and thiocyanate (NCS⁻) may be present, TOC analyzers typically quantify TIC mainly through CO₂ released from bicarbonate and carbonate species.

Total organic carbon (TOC) – the fraction of carbon bound in organic compounds, whether dissolved or suspended. In practice, certain non-carbonate inorganic species and particulate elemental carbon may interfere with or be included in TOC measurements, depending on the analytical method.

In a typical TOC analysis, both total carbon (TC) and inorganic carbon (IC or TIC) are measured. The TOC value is then obtained by subtracting inorganic carbon from total carbon (TOC = TC − IC) (4).

As a brief note for non-chemists, most of the compounds extracted from coffee during brewing are carbon-based. These include organic compounds (measured as TOC) and inorganic carbon species (measured as TIC). Only a relatively small fraction of extracted compounds—such as certain minerals—do not contain carbon.

This means that TOC analysis can capture the majority of carbon-containing substances present in brewed coffee, providing a useful, though not complete, representation of the extracted chemical composition.

This approach is more precise and informative than a simple total dissolved solids (TDS) measurement. A TDS meter estimates the concentration of dissolved ionic species based on electrical conductivity and does not account for undissolved material or non-ionic compounds such as oils. Moreover, it cannot distinguish between organic and inorganic carbon species.

In this experiment, a TOC-L analyzer from Shimadzu Corporation was used. The instrument operates on a high-temperature combustion principle. A portion of the sample is first acidified with phosphoric acid, converting all inorganic carbon species into dissolved CO₂ and shifting the carbonate equilibrium toward CO₂ formation:

The resulting CO₂ is then carried to a detector for quantification. The second half of the sample is introduced into a catalytic combustion chamber, where it is heated to approximately 600–700 °C. Under these conditions, all carbon present in the sample is oxidized to CO₂ in the presence of a catalyst and oxygen. The combustion gases are subsequently passed through a cooling unit and then directed to the detector.

Detection is performed using a non-dispersive infrared (NDIR) spectrophotometer, which quantifies CO₂ concentration. Total inorganic carbon (TIC) is determined from the acidified fraction, while total carbon (TC) is measured from the combusted fraction. Total organic carbon (TOC) is then calculated by subtracting TIC from TC (TOC = TC − TIC).

Sample preparation

A series of samples was prepared according to the following procedure:

1. 15 g of freshly roasted coffee (roasted 3 days prior to the experiment), ground to a medium-coarse size suitable for AeroPress;

2. Water temperature: 85 °C;

3. A total of 200 mL of water was used for all samples;

4. Total brewing time: 2:30 min for all samples;

5. AeroPress inverted method;

6. 30 s plunging time;

7. 10 gentle stirs at the 1:00 min mark.

Eight samples were prepared in total: four with a blooming step and four without. In the blooming group, coffee grounds were first pre-infused with 50 mL of water and allowed to bloom for 40 s, after which the remaining 150 mL of water was added. In the non-blooming group, the full 200 mL of water was added as rapidly as possible without a defined blooming phase.

Immediately after brewing, the coffee was transferred into 50 mL centrifuge tubes, which were filled completely and tightly sealed to minimize CO₂ exchange with the headspace and to reduce further degassing. The samples were then stored at 4 °C. Lowering the temperature shifts the CO₂ equilibrium toward dissolved CO₂ and carbonic acid formation, thereby reducing gas release. Samples were kept refrigerated until analysis.

For TOC analysis, 1 mL of each sample was transferred into a 50 mL centrifuge tube and diluted with 49 mL of ultrapure water. The solutions were gently mixed 2–3 times and then analyzed using the TOC instrument. An injection volume of 300 µL was used for each measurement, based on both total carbon (TC) and total inorganic carbon (TIC) determination. Each sample was measured in triplicate to improve statistical reliability.

On the following day, the same measurements were repeated; however, in this case the samples were analyzed immediately after preparation without prior cooling.

This experiment does not require the construction of a calibration curve, as it is designed as a comparative study rather than an absolute quantitative analysis. In addition to the coffee samples, tap water, filtered water, and ultrapure water were also analyzed to provide reference baselines and improve the interpretability of the results.

Since the objective is comparative evaluation, peak areas were used directly as the response variable, expressed in arbitrary area units (AU), rather than converting signals into absolute concentrations.

The surprising results

Immediately after obtaining the first results, the outcome was somewhat unexpected. Prior to the experiment, it was assumed that there would be a clear difference between the two coffee preparation methods, and the study was designed with this hypothesis in mind.

However, repeated analyses consistently confirmed the same results.

No measurable difference was observed in the concentration of inorganic carbon across all samples. The mean value across the four replicates for coffee brewed with a bloom was 666 AU, while the corresponding mean for coffee brewed without a bloom was 716 AU (Table 1).

Table 1

To pre-empt interpretation of a clear effect, it is important to consider the reference measurements of different water samples shown in Table 2. Tap water 1 and tap water 2 originate from the same source but were collected at different times. The difference in inorganic carbon content between these two measurements is 68.8 AU, which exceeds the observed difference between coffee brewed with and without blooming.

A difference of approximately 50 AU was also observed between tap water and the same water after filtration through a household water filter jug. In the city where the measurements were performed, the water is classified as very soft, with a hardness below 2.5 °dH (German degrees).

Furthermore, differences between ultrapure water and tap water were in the range of approximately 160 AU (Graph 1).

Table 2

Graph 1

Measuring a new batch of samples on the following day immediately after preparation yielded consistent results. Coffee brewed with a bloom and coffee brewed without a bloom differed in inorganic carbon content by approximately 70 AU (Table 3).

When compared to the total carbon signal of approximately 150,000 AU, this difference represents only about 0.047% of the total extracted carbon, indicating a very small relative effect.

Although a measurable difference exists, it is unlikely to be perceptible without highly sensitive analytical instrumentation such as TOC analysis. The instrument used has a detection limit of approximately 4 ppb (µg/L), and the observed difference of ~70 AU falls within the low-concentration (ppb-range) sensitivity of the method.

Table 3

Conclusions

All experimental results indicate that the difference in CO₂ content—represented as dissolved carbonic acid and measured as Total Inorganic Carbon (TIC)—between coffee brewed with and without a bloom is negligible. This suggests that, under the controlled conditions of this study, CO₂ removal is comparable in both methods when all other variables (temperature, total brew time, brew ratio, stirring, and plunging) are kept constant.

Based on these findings, the effect of blooming on reducing dissolved CO₂ appears to be minimal within the sensitivity and scope of this experimental setup. Consequently, the results do not support the hypothesis that blooming significantly enhances extraction by reducing dissolved carbonic acid under the tested conditions.

The observed differences are so small that they are only detectable using highly sensitive and specialized laboratory instrumentation.

These results may be met with skepticism, as the role of blooming is widely accepted within the specialty coffee community and has been frequently discussed by experts and practitioners. As a result, its perceived importance is often taken as given rather than experimentally verified.

This study therefore aims to contribute empirical data to a topic that is largely based on established practice and anecdotal evidence rather than quantitative measurement.

Whether these findings will be widely accepted remains uncertain. However, as a scientist, I trust the experimental results.

One thing is certain: this research will continue, as I plan to extend it to the standard AeroPress method, V60, Chemex, and other drip brewing techniques. At the time of writing, I am already working on these additional samples. However, I have the impression that the overall trend in the results is already becoming clear.

During the experiment, an additional and potentially more interesting observation emerged: the Total Organic Carbon (TOC) content. This parameter may provide insight into extraction efficiency, repeatability, overall extraction yield, and potential differences between brewing with and without a bloom.

However, that is a story for another time…

If you like what I'm doing, please support me on Patreon following the link.

---

Literature:

(1) Effect of Roasting Conditions on Carbon Dioxide Degassing Behavior in Coffee XIUJU WANG, LOONG-TAK LIM* Department of Food Science, University of Guelph, Guelph, On, N1G 2W1, Canada

(2) Illy and Viani 1995; Schenker 2000; Clarke and Vitztbum 2001.

(3) Greenwood, Norman N.; Earnshaw, Alan (1997). Chemistry of the Elements (2nd ed.). Butterworth-Heinemann. p. 310. ISBN 978-0-08-037941-8.

(4) Lenore S. Clescerl; Arnold E. Greenberg; Andrew D. Eaton (1999). Standard Methods for Examination of Water & Wastewater (20th ed.). Washington, DC: American Public Health Association. ISBN 0-87553-235-7. Method 5310A