Introduction

Climate change is an increasing concern for humanity. The anthropogenic carbon dioxide (CO2) emissions are one of the key contributors to this global phenomenon. Despite the Paris climate agreement of 2015 to limit global warming well below 2 °C increase, global emissions of anthropogenic CO2 into the atmosphere are still far too high to be compatible with this scenario. The main consequences of global warming are rising sea levels, disruption of ecosystems, and negative impacts on human health and the economy (Field et al. 2014; Matsumoto 2019).

Different strategies are pursued to reduce the atmospheric CO2 concentration. For this purpose, it is necessary to actively remove CO2 from the atmosphere (carbon capturing) and to sequester it permanently (López-Pacheco et al. 2021; Choi et al. 2022). Technologies that remove CO2 from the atmosphere are called negative emission technologies (United Nations, Department of Economic and Social Affairs, Population Division 2019; Renforth and Wilcox 2020). They were proposed as one of three main strategies by the Conference of the Parties (COP21) held in Paris (Pires 2017; Kremer et al. 2022).

The routes for CO2 capture and sequestration comprise geological (e.g., biochar, oceanic carbon storage), physico-chemical (e.g., ab/adsorption, cryogenic-based capture and membrane-separation processes), and biological approaches (e.g., energy crops, coalbed methanogenesis and bacterial/algal approaches) (Nanda et al. 2016; Song et al. 2019; Kammerer et al. 2022). Here, we focus on the biological approach using a cyanobacterium.

Microalgae or cyanobacteria can be used as a biological strategy (Bhola et al. 2014; Roy et al. 2022; Onyeaka and Ekwebelem 2022) because they can fix 10–50 times more CO2 per area and time than most plants. They can use carbon from various sources:

  1. 1.

    Fossil-based carbon in the form of soluble carbonates (e.g., NaHCO3 and/or Na2CO3) (Thieme 2000; Eggeman 2011; Agbebi et al. 2022),

  2. 2.

    CO2 from industrial exhaust gases (e.g., flue gas and flaring gas). However, such gases must be purified when the biomass is used for humans (Nguyen et al. 2021) because flue gas has been found to contain dozens of substances other than CO2 (such as SO2, nitrogen oxides (NOx), and fine dust particles (Huang et al. 2016)), and their effect on microalgae cells has not been fully explored (Cheng et al. 2018),

  3. 3.

    CO2 from the atmosphere.

For optimization of biomass yield and better control of carbon utilization, commercial AP producers use high concentrations of NaHCO3 added to the cultivation medium. This, however, leads to substantial outgassing of CO2 into the atmosphere. By contrast, using atmospheric CO2 as the sole carbon source for AP production has not been established yet.

Although the atmospheric CO2 has already increased from 0.028% at the beginning of the industrial revolution to currently 0.04% CO2, the mass transfer is described to be so low that the cyanobacterial growth is too slow to be applied for economic use (Chelf et al. 1993). Therefore, in big-scale production, soluble carbonates and or a mixture of air/CO2 from industrial suppliers are conventionally used as carbon sources (Araújo et al. 2021).

Among cyanobacteria, Arthrospira platensis (AP) have several advantages that have led to the production of these species on a big-scale worldwide (Araújo et al. 2021; Razaviarani et al. 2022): a very high protein content of up to 75% and further valuable ingredients like Phycocyanin. Moreover, AP has been considered a safe food by the European Food Safety Authority (EFSA) (Chacón-Lee and González-Mariño 2010). Interestingly, AP grows at pH values between 9–10, a pH range at which no toxic accompanying flora was found (Heussner et al. 2012). By contrast, cyanotoxin contamination might be of grave concern, especially in cyanobacterial strains grown in uncontrolled and unmonitored environments, such as outdoor ponds or lakes (Grosshagauer et al. 2020).

To ensure that sufficient carbon is available for the buildup of AP biomass in a culture medium without fossil-based C, atmospheric CO2 must be constantly pumped into the culture medium. Here, CO2 is converted into \({\text{HCO}}_{3}^{ - }\) due to the CO2-\({\text{HCO}}_{3}^{ - }\) equilibrium (see Hägg diagram) in the pH range around 9 (Andersen 2002) (see Eq. 1).

$$ {\text{CO}}_{{2}} + {\text{H}}_{{2}} {\text{O}} \rightleftharpoons {\text{H}}_{{2}} {\text{CO}}_{{3}} \rightleftharpoons {\text{HCO}}_{3}^{ - } + {\text{H}}^{ + } $$
(1)

The bicarbonate ion (\({\text{HCO}}_{3}^{ - }\)), which enters the cell by active transport (Shibata et al. 2002), is the only carbon source for AP at this pH value. Intracellularly, \({\text{HCO}}_{3}^{ - }\) + H+ are enzymatically converted into CO2 and H2O by carbonic anhydrase. The \({\text{HCO}}_{3}^{ - } { }\) taken up by AP cells must be continuously balanced by a sufficiently large CO2 transfer to avoid phototoxicity due to lack of C availability within the cell. In case of C deficiency, light energy would no longer be used to build biomass, and reactive oxygen species (ROS) are formed, causing cell damage and even cell death (Cooke et al. 2003). To avoid cell damage, the air flow must be high enough so that the extremely low content of CO2 of 0.04% in the airflow must be enough so that CO2 can be dissolved in the culture medium (see Eq. (1)). This requires optimized conditions in the bioreactor (airflow, air bubble diameter, time of the bubbles in the medium, etc. (Jung et al. 2019) to ensure a sufficient \({\text{HCO}}_{3}^{ - }\) concentration.

In this experiment, we show for the first time that it is possible to produce comparable amounts of AP biomass by growing in a culture medium supplied with ambient air as the sole source of carbon in comparison with a culture medium (Zarrouk medium) containing NaHCO3 as carbon source.

Materials and methods

Microalgae strain and growth medium

AP used for cultivation was obtained from the “Culture Collection of Algae at Göttingen University” (strain: SAG21.99).

For the propagation of backup cultures of AP, Zarrouk medium (Zarrouk 1966) without the addition of NaHCO3 was used to cultivate AP consisting of (per liter): 0 g NaHCO3, 0.5 g K2HPO4, 2.5 g NaNO3, 1 g K2SO4, 1 g NaCl, 0.2 g MgSO4 · 7H2O, 0.04 g CaCl2, 0.01 g FeSO4 · 7H2O, 0.08 g Na2EDTA and 1 mL of trace metal solution. The trace metal solution consisted of (per liter): 2.86 g H3BO3, 1.81 g MnCl4 · 4H2O, 0.222 g ZnSO4 · 4H2O, 0.0177 g Na2MoO4, 0.079 g CuSO4 · 5H2O. The growth medium was initially sterilized at 121 °C in an HV-50 autoclave (SYSMEX VX-95, Sysmex, Norderstedt, Germany) for 15 min.

Experimental design

For the experiments, AP from the same stock solution was cultured in six 2 L bubble column glass bioreactors with 1.8 L working volume for 28 days. All bioreactors were thermostatted to 30 °C, illuminated by fluorescent lamps, and mixed up using magnetic stirrers (Jung et al. 2021). Photon flux densities of 50 μE/(m2/s) were applied at a 24/0 h photoperiod using fluorescent lamps.

Three bioreactors contained Zarrouk medium with 16.8 g/L NaHCO3. These bioreactors were aerated with 0.236 vvm air with 2% CO2 (0.5 L/h 100% CO2 in 25 L/h atmospheric air, described hereafter as “NaHCO3/CO2-based”). Conventionally, this additional CO2 aeration should compensate for the carbon loss due to CO2 outgassing. Three other bioreactors contained Zarrouk medium but without the addition of NaHCO3 and were aerated with 0.926 vvm of atmospheric air flow (100 L/h of atmospheric air flow, described hereafter as “air-based”). All measurements were performed as triplicates.

The air was pumped through a membrane filter (Millipore; 0.45 µm pore size, 10 cm diameter) and moistened by passage through distilled water.

On every second day of cultivation, optical density, dry weight, pH value, nitrate, and phosphate concentrations were measured. In addition, phycocyanin, allophycocyanin, phycoerythrin, chlorophyll-a, carotenoids, and fatty acids were measured 28 days after inoculation.

A pre-study showed that the ratio of gas flow and illumination intensity must be balanced. At an illumination of 50 μE/(m2/s), an airflow of 0.926 vvm was sufficient for the AP to grow well. The AP died at 100 μE/(m2/s) during prolonged exposure times (320 h, see Figure S1 of the Supplement). Therefore, an airflow of 0.926 vvm was selected for all following experiments.

Analytical methods

The dry weight of biomass was determined after centrifugation with 17,000 g, washing the cell pellet, and second centrifugation (again at 17,000 g), drying, and then weighing. Using the mass balance, the formation of 1 kg of bio-dry matter results in a calculated consumption of 1.8 kg of CO2 (Doucha and Lívanský 2006). This factor was used to assess the quantity of CO2 fixed into AP dry mass. The pH values were monitored continuously over the cultivation time (EGA 133, Sensortechnik Meinsberg, Meinsberg, Germany).

Elemental analysis (elemental analyzer Euro Vector, Pavia, Italy) was used to determine the content of nitrogen (N), carbon (C), hydrogen (H), and sulfur (S).

The productivity of AP was calculated according to Eq. (2):

$$ {\text{Px}} = \left( {{\text{Xm}} - {\text{Xi}}} \right){/}\left( {{\text{tc}}} \right) $$
(2)

with Px: productivity (g/L per day), Xi: initial biomass concentration (g/L), Xm: maximum biomass concentration (g/L), and Tc: cultivation time related to the maximum biomass concentration (days).

$$ {\text{R}} = {\text{Px}}\; \cdot \;X_{{{\text{Cbm}}}} \cdot{\text{M}}_{{{\text{CO2}}}} /{\text{M}}_{{\text{C}}} $$
(3)

The CO2 fixation rate R was calculated according to Eq. (3) with:

XCbm: C content of the microalgae cell, MCO2: Molecular weight of CO2, and MC: Molecular weight of C.

In addition, the specific growth rates µ for both processes were calculated according to the Eq. (4):

$$ \mu = ({\text{ln }}\left( {{\text{Xm}}} \right) - {\text{ln}}\left( {{\text{Xi}}} \right))/({\text{t}_\text{m}} - {\text{t}_\text{i}}) $$
(4)

where Xi is the initial biomass concentration (g/L), Xm the maximum biomass concentration (g/L), ti the initial cultivation time (days), and tm the cultivation time related to the maximum biomass concentration (days).

The pigments phycocyanin, allophycocyanin, and phycoerythrin were photometrically determined according to Bennett & Bogorad (Bennett and Bogorad 1973). The nitrate concentration was determined according to the Association of Official Analytical Chemists (AOAC) methods (Latimer 2023). Phosphate was quantified by using the Spectroquant® Phosphate Reagent Test (Merck, Darmstadt, Germany). Total carotenoids and chlorophyll-a were spectrophotometrically estimated at 450 nm and 660 nm, respectively (Lichtenthaler and Buschmann 2001). Fatty acids present in the cells in lipid form were converted to fatty acid methyl esters (FAME) and determined according to Bligh & Dyer (Bligh and Dyer 1959). These were analyzed with a gas chromatograph (Agilent 7820A, GC column SP2560 from Supelco, use of biscyanopropyl polysiloxane).

On each experimental day, samples were taken from the bioreactor and were examined via bright field and phase contrast microscopy (Axio Scope, Zeiss Microimaging GmbH, Jena, Germany; BZ-X810, Keyence, Japan). Samples were further studied by laser scanning microscopy (Axio Observer.Z1/7, Zeiss Microimaging GmbH, Jena Germany). Geometry measurements of the trichomes were carried out with ImageJ (National Institute of Health, Bethesda, MD, USA) (Collins 2007).

Statistical analysis

For all samples, mean values with standard deviation are given. Gaussian distribution and variance of the samples were tested by Kolmogorov–Smirnov test and F-test. The null hypothesis could not be rejected (p > 0.05), so that two-sided t tests for two sample problems could be performed with comparable variance as well.

Results and discussion

Here, we compare two different processes for AP biomass production. The traditional one is based on fossil-based NaHCO3 added to the AP culture medium (i.e., Zarrouk full medium). As mentioned in Materials and Methods, an additional CO2 aeration should compensate for significant carbon loss due to outgassing in conventional bioreactors with high-concentrated fossil-based NaHCO3. The new procedure does not require the addition of NaHCO3 but relies on aeration with ambient air providing atmospheric CO2 as the sole carbon source.

The AP productivity in the fossil-based bioreactor was 70 mg/L per day, and it was 77.14 mg/L per day in the air-aerated bioreactor.

Biomass

Figure 1 shows the growth of AP biomass over time for both conditions. The growth of AP in both bioreactors did not differ (p = 0.168). In the bioreactor lacking fossil-based carbon, a biomass of 2.16 g/L was grown (biomass at day 28 minus inoculated biomass) and in the bioreactor with fossil-based carbon and 2% CO2, a biomass of 1.96 g/L was grown (biomass at day 28 minus inoculated biomass). The specific growth rate in the bioreactor lacking fossil-based carbon was 0.028 and in the bioreactor with fossil-based carbon 0.024.

Fig. 1
figure 1

Arthrospira platensis biomass production over 28 days for the bioreactor without fossil-based carbon with air ambient flow only (□: open squares) and for the bioreactor with the addition of fossil-based carbon (NaHCO3) plus airflow with 2% CO2 (■: black squares)

Full size image

According to Doucha and Lívanský (2006), the general C-conversion factor is 1.8, which means that 1.8 g of CO2 had to be taken up and be metabolized per gram of AP biomass produced. Accordingly, 3.89 g/L CO2 was taken up in the air-based condition, and 3.53 g/L was taken up in the NaHCO3/CO2-based condition. The amounts of gCO2/L fixated in AP can be calculated using the measured C concentration (as weight percent of AP), and the ratio of MCO2/MC (2.16 g/L · 0.4773 gC/gAP · 44/12 MCO2/MC) is 3.78 gCO2/L for the air-based condition and 3.42 gCO2/L for the NaHCO3/CO2-based condition. This is the amount of CO2 fixated in AP cells during the growth of 28 days per L of bioreactor volume. The calculated data correlate well with those calculated by Doucha (Doucha and Lívanský 2006).

Elemental analysis

The proportion of different elements in AP for the two cultivation conditions is shown in Table 1. Except for the sulfur content in AP, which was significantly greater in the airflow bioreactor (p = 0.00011), the elemental composition did not differ. From the nitrogen content, the percentage of protein can be estimated. The conversion factor for AP is 6.25 (Piorreck et al. 1984). This results in protein content of 69% for both production types. By using these data, the CO2 fixation rate R can be calculated. R in the air-based cultured bioreactors was 0.123 g/L per day, while the C content of the fossil-based cultured algae was 0.110 g/L per day.

Table 1 Proportion of different elements [nitrogen (N), carbon (C), hydrogen (H), and sulfur (S)] in Arthrospira platensis for the cultivation in the air-based and the NaHCO3/CO2-based bioreactors
Full size table

Acidity of the culture medium

Figure 2 shows the course of the pH values in the culture medium for both conditions. While the pH values in the air-based bioreactors remained constant over cultivation time, the pH values of the NaHCO3/CO2-based reactors slightly dropped after inoculation and then remained constant.

Fig. 2
figure 2

pH values of the culture medium over 28 days of cultivation for the air-based (□) and the NaHCO3/CO2-based (■) bioreactors

Full size image

Nitrate and phosphate

Figure 3 shows the course of nitrate and phosphate concentrations in the respective media during the cultivation for both conditions. The nitrate, as well as the phosphate concentration, decreased comparably for both procedures over time. This is also evident when looking at yields in terms of decreases in the substrates nitrate and phosphate ions in the nutrient medium (see Table 2).

Fig. 3
figure 3

Course of nitrate and phosphate concentrations in [mg/L] during the cultivation time for the air-based (□) and the NaHCO3/CO2-based (■) bioreactors

Full size image
Table 2 Phosphate and nitrate yields in the Arthrospira platensis cultivation medium
Full size table

Phycobiliproteins

The effect of processing conditions on different pigments in AP is shown in Table 3. Phycobiliproteins as phycocyanin (PC, p = 0.819) and allophycocyanin (APC, p = 0.113) did not differ in the two production variants, while a significant but not relevant difference appeared for phycoerythrin (PE, p = 0.149).

Table 3 Influence of cultivation in the air-based and the NaHCO3/CO2-based bioreactors on pigments of Arthrospira platensis
Full size table

This was different for chlorophyll-a. Here, significantly more chlorophyll-a had formed in the airflow bioreactor with 18.24 ± 0.71 µg/mgbiomass (p = 0.0002) than in the bioreactor with fossil-based carbon and 2% CO2 (13.36 ± 2.00 µg/mgbiomass).

The ratios concerning carotenoids were similar. Again, there were significantly more carotenoids in the airflow bioreactor with 4.45 ± 0.24 µg/mgbiomass (p = 0.0001), than in the bioreactor with fossil-based carbon and 2% CO2 (2.86 ± 0.55 µg/mgbiomass).

In line with these findings are those reported by, e.g., Park et al. (2013), Gordillo and coworkers (Gordillo et al. 1998), and Chunzhuk et al. (Chunzhuk et al. 2023). The studies revealed that the gassing of CO2 into the culture medium can influence the chlorophyll and carotenoid content in the AP biomass. Particularly interesting for the comparison with our data are the results of Park et al. (Park et al. 2013a), since the CO2 gassing protocol (2%) and the composition of the standard Zarrouck medium are comparable to our study. They showed that the complete omission of CO2 gassing resulted in an increased chlorophyll-a and carotenoid content in the AP biomass compared to all higher CO2 levels applied. Moreover, with increasing CO2 concentration (up to 8% CO2 gassing), the amounts of the pigments were decreasing (e.g., about 0.6–0.7% at 0% CO2 gassing, below 0.3–0.4% chlorophyll-a of the biomass at 8% CO2 gassing) (Gordillo et al. 1998; Park et al. 2013a).

The mechanisms behind this are not clear. One hypothesis describes the accumulation of inorganic carbon in the culture medium in addition to the already supplemented sodium hydrogen carbonate. This results from the CO2 level being above the atmospheric concentration, and the initial pH (9.0–9.5) as a driving force for the proton-generating conversion of the gassed CO2 into hydrogen carbonate and carbonate (carbonic acid route). The pH is reduced and, thus, can inhibit the photosynthetic activity of AP because enzymes and pigments involved in photosynthesis are sensitive to pH changes.

It was further suggested that these excessive levels of CO2/inorganic carbon in the culture medium might promote the degradation of light-harvesting pigments. Either because they are not required for energy generation (Gordillo et al. 1998), or they are affected by the generation of toxic metabolites such as reactive oxygen species as suggested by Ganesh and Park for A. maxima and A. platensis (Ganesh et al. 2007; Park et al. 2013a).

Comparable to our result, pH values decreased initially and stabilized at around 8.5 during the experiments (Park et al. 2013a). A further increase in the CO2 concentration (8%) did not lead to a further decrease in the pH, which can be attributed to the pH stabilizing properties of the sodium hydrogen carbonate and the associated carbonate equilibrium (Kishi et al. 2020; Chunzhuk et al. 2023).

In the present study, the medium of the air-aerated AP culture was adjusted to a pH of 10 at the beginning of the experiment. In the further course, this pH level did not significantly change despite the absence of NaHCO3 and its buffering capacity (see Fig. 2).

When the culture is gassed with ambient air only, the CO2 dissolves into the medium, similarly to when the standard culture medium is gassed with CO2 directly. As mentioned above, this can lower the pH of the medium. The AP cells can use the CO2 or the formed hydrogen carbonate for photosynthesis. It appears conceivable that AP increases photosynthetic capacity and, thus, chlorophyll and carotenoid content to use these molecules. Particularly in the absence of other inorganic carbon sources. This raises the pH again due to the consumption of these molecules and the production of oxygen. The balancing of both processes—the intake of air-born CO2 and the photosynthetic activity—might result in the relative stability of the pH under these culture conditions. The results of Park et al. confirm that these processes are associated with the gassed CO2 and not with the supplemented sodium hydrogen carbonate. They have shown that in standard AP cultures without CO2 gassing, the pH raised from 8.5–9 up to 12, despite the presence of sodium hydrogen carbonate in the culture medium (Park et al. 2013).

Thus, a central role in the generation of chlorophyll and carotenoids in AP plays the concentration of the gassed CO2. Relatively low amounts appear to support the generation of the pigments while supplementing higher amounts results in their reduction in the AP biomass.

Fatty acids

The effect of cultivation conditions on different fatty acids in AP is shown in Table 4. In the airflow bioreactor, more fatty acids were detected than in the bioreactor with fossil-based carbon (p = 1.43E-06). The fatty acids present in AP dry biomass in greater proportions were palmitic acid (C16: 0) and γ-linolenic acid (C18: 3n6). All other fatty acids showed very low concentrations.

Table 4 Influence of cultivation in the air-based and the NaHCO3/CO2-based bioreactors on the total fatty acid concentration as well as of specific fatty acids in [mg/mgbiomass] of Arthrospira platensis
Full size table

Morphology

Figure 4 shows two representative images at the end of the linear growth phase (29 days after the inoculation) for the fossil-based bioreactor (Fig. 4B) and the air-based bioreactor (Fig. 4A). The diameter, as well as the length of the AP spirals, did not differ for both processes at this time point.

Fig. 4
figure 4

Representative images of Arthrospira platensis spirals at the end of the linear growth phase (29 days after the inoculation) for A the air-based bioreactor and B the fossil-based bioreactor (Primary magnification of 1:20)

Full size image

Carbon dioxide balance of both processes

The massive emissions of carbon dioxide, mainly originating from the consumption of fossil fuels, have a tremendous impact on climate change (Lenton et al. 2019) and have become a huge challenge for sustainable human development. Besides the dramatic reduction in CO2 emissions, all technologies that can make some contributions to removing CO2 from the atmosphere should be investigated to identify the best technologies that can contribute to the mitigation of climate change (Field et al. 2014; Maheshwari et al. 2022).

The growth of AP depends on the availability of \({\text{HCO}}_{3}^{{ - { }}} ,{ }\) on macronutrients, pH, temperature as well as light color, light quantity, and light distribution (Jung et al. 2019; Daneshvar et al. 2022). High concentrations of oxygen—which is generated by the photolysis of water during photosynthesis—have to be reduced in the culture medium to avoid photorespiration of the AP culture, a side reaction that is also catalyzed by ribulose-1,5-bisphosphate carboxylase-oxygenase (RubisCO) leading to the fixation of oxygen (oxygenation) and the release of CO2 (Veaudor et al. 2020). To prevent the accumulation of high oxygen concentrations and to supply a sufficient amount of CO2, a rather high airflow has to be applied within the bioreactor.

The study revealed that aeration with a high-volume ambient airflow into the cultivation medium can lead to a growth of AP biomass comparable to the growth of AP cultured in a medium with an artificial and fossil-based carbon source (NaHCO3) together with an airflow of 2% CO2 in the culture medium.

While in the air-based bioreactor, 3.78 gCO2/L were taken up, the value was only slightly lower in the NaHCO3/CO2-based bioreactor (3.42 gCO2/L). The difference seems small, but it might have a significant effect on the atmosphere. In the air-based process, the biomass is built up using CO2 that has been exclusively extracted from the atmosphere. In contrast, the biomass buildup in the NaHCO3/CO2 -based process occurred with fossil- or industry-based CO2, which is often added in excess (Yahya et al. 2020; Nguyen et al. 2021). Even if the biomass buildup were to occur solely from the CO2 supplied in the air/CO2 stream, 3.42 gCO2 would be fixated into AP per L bioreactor volume. However, with the NaHCO3/CO2-based bioreactor, a total of 336 L of pure CO2 are pumped into the culture medium over 28 days. With a CO2 density of 1.98 g/L, this corresponds to a weight of 665.28 g CO2.

The calculation of the CO2 data per bioreactor with 1.8 L working volume resulted in the following values: from the NaHCO3/CO2-based bioreactor, 659.12 g of CO2 (665.28 gCO2–3.42 gCO2/L ⋅ 1.8 L) were released unused to the atmosphere. In contrast, 6.80 g of CO2 were fixated from the air-based bioreactor without releasing additional CO2 into the atmosphere. Extrapolation of our data to the global AP wet biomass (live weight) production—e.g., 86,000 tons in 2014 (Food and Agriculture Organization of the United Nations (FAO) 2018)—showed that the NaHCO3/CO2-based method would release 181,055.7 tons of CO2 into the atmosphere. With the ambient air-based method, 1867.9 tons of CO2 can be removed from the atmosphere.

This does not yet account for the amount of CO2 released to the atmosphere from the NaHCO3 of the nutrient medium (until the equilibrium is reached with the 2% CO2 gas stream). The conversion of the added CO2 to the AP biomass is very low as 0.926% (6.16 g/665.28 g), which means that more than 99% of the added CO2 escapes unused to the atmosphere under the present conditions.

The culture medium pH range for both processes was in the range of 9.0–10.0, which is optimal for AP (Klok et al. 2013). Klok et al. found that nitrogen depletion during the cultivation of microalgae can lead to a decrease in biomass productivity (Klok et al. 2013) and lower chlorophyll concentrations which did not occur during the study (Rosa et al. 2015; Liu et al. 2019).

Valuable ingredients such as phycocyanin or chlorophyll-a and carotenoids (Yahya et al. 2020)—which can inhibit the growth of cancer cells (McQuistan et al. 2012; Braune et al. 2021; Russo et al. 2022; Jung et al. 2022)—are present in comparable or even increased amounts in AP produced on the air-based method compared to the conventional one. In addition, the protein content produced by AP was comparably high with 69% of the cell mass.

To the best of our knowledge, there is only one study in which a modified Zarrouk medium without fossil-based carbon in form of NaHCO3 was performed (Rosa et al. 2015). However, in this study, even the control cells were aerated with additional CO2, and a control using an ambient airflow was not included. Thus, a direct comparison to our results is not possible. Former studies showed that there is a clear dependence of the productivity from the CO2 concentration from additional CO2 treatment (de Morais and Costa 2007; Zeng et al. 2012; Ma and Gao 2014; Kim and Lee 2018). Though the different strains of AP show extremely different productivities, it seems that about 2% of CO2 gas flow might be optimal. Therefore, this CO2 concentration was used as a comparison to the air-based process.

Conclusion

The study demonstrates for the first time that cultivation of AP with Zarrouk medium without NaHCO3 using an adequate flow of ambient air as the only C source is possible. Biomass yields are nearly identical to those obtained by cultivation of AP with fossil carbon-supplemented growth medium. The particular advantage of this novel process is that no CO2 from fossil C sources escaped from the bioreactor into the atmosphere, while the conventional method of AP production with NaHCO3 in excess and additional CO2 leads to a considerable burden on the atmosphere with CO2. Valuable ingredients—such as phycocyanin or chlorophyll—are present in equal or even higher quantities in the air-based process.

The results improve our understanding of how AP production can be decarbonized using atmospheric CO2 as the sole source and represents the proof of principle.