Sustainable Utilization of CO₂ from Exhaust Gases for the Autotrophic Cultivation of the Biohydrogen-Producing Microalga Tetraselmis subcordiformis

Abstract

The aim of the study was to evaluate the feasibility of using exhaust gases as a CO₂ source in the cultivation of Tetraselmis subcordiformis microalgae for biomass and hydrogen production. It was shown that the growth rate of T. subcordiformis biomass and its biochemical composition depended on the CO₂ source. The highest growth rate of 286 ± 15 mgVS/L-d and a final biomass concentration of 2710 ± 180 mgVS/L were achieved in the variant where exhaust gases from a coal and biomass supplementary combustion plant were the CO₂ source (V2). The highest CO₂ reduction efficiency of 90.3 ± 3.2% was achieved in the case where waste gases from biogas combustion were the CO₂ source (V1). In V2, the highest CO₂ utilization efficiency was achieved (CO₂UE = 46.7 ± 2.4%). Analyzing the biomass composition confirmed differences in total carbon content (TC) and polysaccharide fraction. The highest H₂ production efficiency and rate, which were 70.9 ± 2.7 mL/gVS and 2.27 ± 0.08 mL/gVS·h, respectively, were obtained in V2. The results obtained indicate the possibility of integrating fuel combustion processes with the cultivation of T. subcordiformis and photobiological H₂ production, which is a promising solution in the context of climate neutrality and the implementation of circular economy postulates. This approach demonstrates a sustainable strategy for linking industrial CO₂ emissions with the production of renewable biohydrogen and thus contributes to climate protection and the promotion of circular economy concepts.

1. Introduction

The rising concentration of greenhouse gases in the atmosphere, particularly carbon dioxide (CO₂), represents one of the major environmental challenges of the 21st century. In 2023, average CO₂ levels reached 421 ppm, approximately 50% higher than in the pre-industrial era [1]. In light of the need to reduce greenhouse gas emissions and the ambition to achieve carbon neutrality by 2050, technologies enabling efficient utilization of CO₂ streams as feedstocks for industrial processes are increasingly important [2]. Approaches that simultaneously reduce emissions and generate renewable energy, in line with circular economy principles, are particularly promising [3,4]. One such strategy involves using CO₂ from exhaust gases to cultivate microalgae, which can then be applied in the photobiological production of biofuels [4], including hydrogen [5], methane [6], biodiesel [7], or biomass for thermal energy conversion [8]. Biogas, once purified and combusted, typically produces an exhaust stream containing 5–15% CO₂ by volume, along with water vapor and trace amounts of NO_x and SO₂ that can be removed or neutralized. In comparison, exhaust from fossil fuel or coal-biomass co-combustion plants can contain up to 16–17% CO₂ [9]. The CO₂ from biogas is part of the short carbon cycle, and its use in microalgae cultivation enhances the overall energy efficiency of the system [10]. In contrast, for co-combustion of coal and biomass, only CO₂ from biomass contributes to the short carbon cycle, while fossil carbon emissions remain a net burden [11].

Microalgae such as the green alga Tetraselmis subcordiformis exhibit high photosynthetic rates. Maximum biomass productivity can reach 0.4–0.6 g/L·d [12], and the species tolerates elevated CO₂ levels, often up to 10–15%, without significant growth reduction [13]. T. subcordiformis is also capable of photobiological hydrogen production under anaerobic conditions. Hydrogenases catalyze the reduction in protons to hydrogen molecules, utilizing electrons from water photolysis or cellular metabolites like NADH [14]. These enzymes are activated under sulfur-limited conditions, which inhibit photosystem II, enhance cyclic electron flow, and initiate water photoreduction [15]. Under optimal conditions, immediate hydrogen production rates of 4–6 mmol H₂/g TS·d have been observed, with total yields exceeding 50 mL H₂/L culture in long-term experiments [13].

The species combines high photosynthetic efficiency, rapid biomass growth, and resilience to variable environmental conditions, making it a promising candidate for biohydrogen production [16]. For example, cultivation in a medium containing effluent from soilless tomato production achieved a biomass concentration of 3030 ± 183 mg VS/L and a chlorophyll-a content of 67.9 ± 3.5 mg/L [17]. The growth rate during the logarithmic phase was 270 ± 16 mg VS/L·d, and hydrogen production reached 161 ± 8 mL, with an average rate of 4.67 ± 0.23 mL/h [17]. In another study, using CO₂ from the anode chamber of microbial fuel cells (MFC) as the carbon source resulted in optimal growth at 220 ± 8 mg CO₂/L. Under these conditions, the biomass growth rate reached 0.52 ± 0.03 g VS/L·d, chlorophyll-a content was 11.54 ± 0.42 mg/L·d, final biomass concentration was 2.68 ± 0.10 g VS/L, and H₂ production totaled 312 ± 38 mL, corresponding to 62.4 ± 6.1 mL/d [18].

Previous studies integrating CO₂ biosequestration from waste gases with microalgae cultivation mainly focused on biomass production for fertilizers [19], animal feed [20], or biodiesel from lipid accumulation [21]. Research on photobiological biohydrogen production is less common, particularly when using CO₂ directly from power plant exhaust streams [22]. To date, the scientific literature has not provided comprehensive analyses on the use of T. subcordiformis for capturing CO₂ from real exhaust gases, whether from biogas combustion or conventional municipal heating plants. In particular, studies integrating the assessment of CO₂ assimilation efficiency with simultaneous evaluation of biomass growth dynamics, changes in biochemical composition, and the impact of these parameters on secondary biohydrogen production via water photoreduction are lacking. Most previous research has focused on laboratory-scale cultivation using pure CO₂ streams, which does not reflect the complexity of real exhaust gases containing toxic components, variable gas concentrations, and particulate matter.

This study investigates for the first time the utilization of CO₂ from biogas combustion and the co-combustion of coal and biomass as an inorganic carbon source in the autotrophic culture of T. suncordiformis for biohydrogen production. The growth of the biomass, the macromolecular composition and the H₂ production were analyzed under laboratory conditions with exhaust gas injection and provide insights for integrated energy–biotechnology systems that combine biohydrogen with CO₂ emission reduction.

2. Materials and Methods

2.1. Research Organization

The research was conducted on a laboratory scale. The experiments were divided into two stages (S). In the first stage (S1), T. subcordiformis was cultured using different carbon dioxide (CO₂) sources. The research focused on the evaluation of CO₂ biosequestration efficiency, carbon dioxide utilization efficiency and monitoring the yield and kinetics of microalgae biomass production. The produced T. subcordiformis biomass was then separated from the culture medium by sedimentation. In the second stage (S2), the T. subcordiformis biomass was transferred to anaerobic respirometric bioreactors, which were fed with a new culture medium (for biohydrogen production) and stimulated to produce H₂. The experimental work carried out in S2 focused on evaluating the yield and kinetics of H₂ production. Each S was divided into three variants (V) whose isolation criterion was the source of CO₂ supplied to the culture system. In variant 1 (V1), the CO₂ source consisted of waste gases from biogas combustion in a gas furnace. In variant 2 (V2), desulphurized waste gases from the co-combustion of coal and biomass in a municipal heating plant were used. In variant 3 (V3), the T. subcordiformis culture was fed with atmospheric air. The organizational scheme of the experimental work is shown in Figure 1.

2.2. Materials

2.2.1. T. subcordiformis Biomass

The microalgal strain T. subcordiformis SAG 161-1a was obtained from the University of Göttingen algae culture collection (SAG, Goettingen, Germany) [23]. To obtain sufficient biomass for the experimental procedures, the culture was initially propagated under controlled laboratory conditions. Microalgae were first grown in sterile 50 mL glass tubes (Biospace, Poznań, Poland) and subsequently transferred to 1.0 L glass bioreactors (Duran Bottle System, Mainz, Germany) for further scale-up. All laboratory glassware was sterilized in a 2840 EL-D autoclave (Tuttnauer, New York, NY, USA) at 121 °C for 15 min prior to use.

2.2.2. Culture Medium Used in S1 (Biomass Production)

The composition of the synthetic medium in S1, as described by Guan et al. [24], comprised: 100.00 mg/L NaNO₃, 20.00 mg/L NaH₂PO₄, 1.30 mg/L FeCl₃, 33.60 mg/L H₃BO₃, 45.00 mg/L EDTA, 0.36 mg/L MnCl₂, 0.21 mg/L ZnCl₂, 0.20 mg/L CuSO₄, 0.20 mg/L CoCl₂, 0.09 mg/L (NH₄)4Mo₇O₂₄, and vitamins 1.00 μg/L VB1 and 0.10 μg/L VB12. The salt content in the medium was 30–33 ppt, while the pH was 8.00–8.20.

2.2.3. Culture Medium Used in S2 (Biohydrogen Production)

For all experiments in S2, the medium used for hydrogen production was prepared using deionized water and analytical-grade chemical reagents, with the following composition: 27.23 mg/L NaCl, 5.079 mg/L MgCl₂, 1.123 mg/L CaCl₂, 0.667 mg/L KCl, 0.196 mg/L NaHCO₃, 0.098 mg/L H₃BO₃, 0.098 mg/L KBr, 0.024 mg/L SrCl₂, 0.003 mg/L NaF and 0.002 mg/L CuCl₂. The pH value of the medium was kept in the range of 7.90–8.00. The composition of the medium in S2 was based on literature data and previous studies by the authors.

2.2.4. Carbon Dioxide (CO₂) Sources

In variant V1, exhaust gases derived from the combustion of biogas produced at a commercial agricultural biogas plant (53°36′03.31427″ N, 020°36′12.86527″ E) were cooled to 20 °C before use. The biogas was produced in a fermentation reactor fuelled with cattle manure, maize silage and grass, operating under mesophilic conditions. The basic technological parameters for the operation of the biogas plant were as follows: Organic compound load (OLR) ≈ 2.4 kgVS/m³·d, hydraulic retention time (HRT) ≈ 40 days, process temperature approx. ≈ 39 °C, substrate dosing twice daily with a frequency of every 12 h, complete mixing with vertical axis mixers and total plant power ≈ 30 kWe. The composition of the purified biogas fed to the Bentone BG Biogas ~30 kW gas furnace (Bentone AB, Enertech Group, Ljungby, Sweden) contained 63.1 ± 1.2% CH₄; 32.9 ± 1.7% CO₂; 70 ± 5 ppm H₂S; 1.2 ± 0.3% O₂ and 3.4 ± 0.6% N₂. The average CO₂ content in the flue gases was 13.4% ± 1.2%. In V2, exhaust gases cooled to 20 °C from a coal and biomass auxiliary combustion plant of the municipal thermal power plant in Olsztyn (53°74′77.64492″ N, 20°44′62.22156″ E) were used. The average CO₂ content in the exhaust gases was 14.9% ± 1.7%. In V3, the T. subcordiformis culture was supplied with atmospheric air in which the CO₂ concentration was approximately 420 ppm.

For variants V1 and V2, exhaust gases were collected daily and stored in sealed 70 L Tedlar bags. In all experimental setups, the CO₂ mass flow supplied to the V-PBR was maintained at 8.0 mg CO₂/min. The exhaust gases were introduced into the V-PBR using a peristaltic pump (FASTLoad Programmable Control Peristaltic Pump, VWR, Darmstadt, Germany) for V1 and V2, whereas in V3, a diaphragm pump (Mistral 200, Aqua Medic, Brentwood Essex, UK) with a flow rate of 10.8 L/min was employed. In V3, the air was discharged from the V-PBR reactor after passing through the culture medium, while in reactors V1 and V2 the exhaust gases were recirculated using a peristaltic pump (VWR Germany) with a capacity of 10.8 L/min. Therefore, the volume flow of gases through the culture medium was the same in all photobioreactors.

2.3. Photobioreactors for T. subcordiformis Biomass Cultivation (S1)

The experiments were carried out in vertically orientated V-PBR tubular reactors made of transparent acrylic glass (polymethyl methacrylate). The active volume of the culture medium was 10.0 L, while the gas space was 5.0 L. CO₂ was introduced into the medium through a valve at the bottom of the reactor and the gas was vented through the top of the system. Each V-PBR was equipped with a pH probe that allowed continuous measurement. Daily measurements were performed and the data were subsequently compiled, averaged, and stored in the pH meter memory for further analysis. The vertical photobioreactor (V-PBR) was illuminated with 10 W white LED lamps (Leddy Slim, Aquael, Warsaw, Poland) with a color temperature of 900 K. The lighting system provided a luminous flux of 720 lm, corresponding to photosynthetically active radiation (PAR) of 270–300 μmol·m⁻²·s⁻¹ within the 400–700 nm wavelength range. The photoperiod was set to 14 h of illumination followed by a 10 h dark phase each day, ensuring consistent light–dark cycles throughout the cultivation period. The temperature of both the incoming gases (air and exhaust) and the culture medium was carefully maintained at 20 ± 2 °C throughout the experiment.

2.4. Hydrogen Production Bioreactor (S2)

The biomass of T. subcordiformis, separated by gravity sedimentation, was transferred into 1.0 L respirometric reactors (OxiTop^®-IDS, WTW, Weilheim, Germany) and incubated at a controlled temperature of 25 ± 1 °C. The reactor contents were continuously stirred at 60 rpm using magnetic stirrers, while changes in gas pressure were recorded every 10 h. Pressure measurements were subsequently converted to volumetric gas production under standard conditions using the ideal gas law. The biomass concentration in each experimental variant was determined by the amount of culture harvested from the S1 system. The experiment was conducted over 120 h, comprising a 30 h dark period followed by 90 h of illumination. The respirometers were illuminated with a 58 W linear fluorescent lamp (Philips Lighting MASTER TL-D Super 80, Philips Lighting, Eindhoven, The Netherlands) providing a color temperature of 6500 K.

2.5. Analytical, Computational and Statistical Methods

Samples for volatile solids (VS) and chlorophyll a (Chl-a) were taken at 48 h intervals. The VS content was determined gravimetrically by calcining the biomass in a muffle furnace at 550 °C (LAC L, Dąbrowica, Poland) and weighing the remaining ash (DanLab AX423, Białystok, Poland). The chlorophyll a concentration was determined using a fluorescence technique (Algae Online Analyser, bbe Moldanke, Schwentinental, Germany). Taxonomic analyses were performed using an MF 346 optical microscope equipped with an Optech 3MP camera (Eduko, Warsaw, Poland). The basic physicochemical parameters of the culture medium were determined using a DR 5000 UV/VIS spectrophotometer (Hach Lange, Düsseldorf, Germany). Salinity was measured with a Marine Control Digital device (Aqua Medic, Janików, Poland) and pH with a VWR meter (International, Radnor, PA, USA). Light intensity was measured with a HI 97500 Luxmeter (Hanna Instruments, Woonsocket, RI, USA). Total nitrogen (TN), total carbon (TC) and total organic carbon (TOC) were determined in the biomass samples dried at 105 °C using a Thermo Flash 2000 elemental analyser (Thermo Scientific, Waltham, MA, USA). Total phosphorus (TP) was determined colourimetrically after digestion in a mixture of H₂SO₄ and HClO₄ using ammonium metavanadate (V) and ammonium molybdate (λ = 390 nm, DR 2800, Hach-Lange, Berlin, Germany). The total protein content was calculated as TN × 6.25. Lipids were extracted using the Soxhlet method with a Büchi B-811 device (BÜCHI Labortechnik, Flawil, Switzerland). Reducing sugars were determined colourimetrically with Anthron reagent (λ = 600 nm, DR 2800, Hach-Lange, Germany). Gas samples (CO₂, O₂, H₂) were taken from the respirometers with a gas-tight syringe and analyzed with a GC 7890 A gas chromatograph (Agilent Technologies, Santa Clara, CA, USA). The exhaust gas composition was determined using a Testo 340 analyser (Testo Ltd., Alton, UK, EN 50379 certificate of conformity). The hydrogen production rate (r) and the kinetic constants (k) were determined using iterative non-linear regression. The carbon dioxide utilization efficiency (CO₂UE) was calculated according to [25] using Equation (1):

$$CUE={\displaystyle \frac{1.88\mathrm{ }·(M_{PBR\mathrm{ }}-\mathrm{ }{M}_0)}{M_{CO2}}}\mathrm{ }$$

(1)

where the coefficient 1.88 results from the adopted summary formula for microalgae biomass (CO_0.48H₁₈₃N_0.11P_0.01) [26]; (M_PBR − M₀) denotes the increase in dry biomass in the V-PBR and M_CO2 is the mass of CO₂ supplied. The M_CO2 value was calculated as the total gas volume multiplied by the density ρCO₂ = 1.8 g/L (28–30 °C, 105 Pa). The total CO₂ volume (V_CO2) was determined based on the efficiency of the pumps supplying CO₂-containing gases (flow rate R_GAM) and the gas analyser (molar fraction n_CO2) (2):

$$V_{CO2}=n_{CO2}·R_{GAM}·(t_2-t_1)$$

(2)

All experiments were performed in quadruplicate. Statistical analyses were performed with Statistica 13.3 (Statsoft, Inc., Tulsa, OK, USA). The significance of differences was assessed with p = 0.05. Normality of distribution was tested using the Shapiro–Wilk test, homogeneity of variance was tested using the Levene test and differences between means were analyzed using a one-way analysis of variance (ANOVA). Tukey’s HSD test was used to identify significant differences between the groups.

3. Results and Discussion

3.1. T. subcordiformis Culture and Properties of the Obtained Biomass

Analysis of biomass concentration changes in T. subcordiformis over the cultivation period revealed a significant (p ≤ 0.05) effect of the CO₂ source on growth dynamics. In variant V1, which was supplied with exhaust gases from biogas combustion, a 5-day lag phase was observed, followed by a 7-day exponential growth phase (Figure 2a). During this phase, the biomass growth rate reached 264 ± 11 mg VS/L·day, resulting in high biomass concentrations of approximately 2010 ± 69 mg VS/L at the end of exponential growth (Figure 2b). The subsequent 3-day stationary phase saw the biomass peak at 2660 ± 123 mg VS/L, followed by a short die-off phase with stabilization of the biomass (Figure 2c). These results indicate efficient utilization of CO₂ for cellular biosynthesis, as reflected by both the growth rate and final biomass concentration (Figure 2d).

Previous studies on biogas combustion gases as a CO₂ source have been conducted with Chlorella vulgaris [27] and Arthrospira platensis [28]. In these cases, biomass growth rates during the first 9 days of the logarithmic phase were 77.8 ± 3.1 mg VS/L·day, with concentrations reaching 745 ± 42 mg VS/L. Maximum biomass (754 ± 45 mg VS/L) was observed on day 13, after which the culture entered the death phase. For cyanobacteria, growth rates during the logarithmic phase were 148 ± 12 mg VS/L·day with air supply and 209 ± 17 mg VS/L·day when CO₂ was supplied from exhaust gas [28].

In variant V2, supplied with waste gases from co-combustion of hard coal and biomass, the lag phase was slightly shorter (4 days, Figure 2a). The linear growth phase lasted 7 days, with the highest growth rate among all variants at 286 ± 9 mg VS/L·day (Figure 2b), reaching a biomass concentration of 2340 ± 42 mg VS/L. The stationary phase extended for 4 days, with a maximum biomass of 2710 ± 91 mg VS/L. The higher CO₂ content in this variant enhanced growth dynamics and final biomass compared to V1 (Figure 2c,d). Similar results were reported for Chlorella sp. ABC-001 cultivated with CO₂-enriched exhaust gases [29]. At 1% CO₂, biomass reached 2330 mg VS/L, with no significant increase at 5% or 10% CO₂ (2320 mg VS/L and 2200 mg VS/L, respectively). However, increasing the gas flux to 0.6 vvm at 10% CO₂ raised biomass to 2900 mg VS/L, while 15% CO₂ produced 2730 mg VS/L after 7 days. These findings highlight that efficient gas delivery, rather than higher CO₂ concentration alone, plays a key role in optimizing biomass production [29].

Variant V3, supplied with atmospheric air, exhibited the longest lag phase of 7 days (Figure 2a) and the shortest linear growth phase of 5 days, with a growth rate of 230 ± 11 mg VS/L·day. Exponential phase biomass reached 1760 ± 123 mg VS/L, peaking at 2240 ± mg VS/L during the stationary phase (Figure 2b,d). Using atmospheric air as a CO₂ source generally leads to lower growth rates and biomass productivity compared to CO₂-enriched sources, and full photosynthetic potential typically requires CO₂ concentrations between 400 and 1000 ppm [30]. In V1, a typical growth curve of T. subcordiformis was observed, which was characterized by the concentration of chlorophyll a in the culture medium. The adaptation phase (lag) lasted 4 days (Figure 3a), indicating a moderate period of time required for the physiological adaptation of cells to new environmental conditions [31]. During this time, the chlorophyll a concentration reached 24.9 ± 2.1 mgchl-a/L (Figure 3b). The intensification of chlorophyll a biosynthesis occurred in the exponential growth phase between day 4 and 11 of the culture (Figure 3a,b). During this period, an increase of over 62 mgchl-a/L was recorded, and the average growth rate was 9.9 ± 0.9 mgchl-a/L∙d. The stationary phase lasted 3 days and the maximum pigment concentration of 94.8 ± 4.2 mgchl-a/L was reached on day 15 (Figure 3c,d).

The total increase in chlorophyll a concentration reached 85.19 ± mg chl-a/L, with an average growth rate of 9.94 mg chl-a/L·d throughout the cultivation period. Similar trends were reported by [18], where T. subcordiformis exhibited a daily biomass increase of 0.52 ± 0.03 g VS/L·d, accompanied by a chlorophyll a rise of 11.54 ± 0.42 mg chl-a/L·d, resulting in a total pigment content of 92.32 mg chl-a/L, comparable to the values obtained in our study.

The strongest chlorophyll a accumulation was observed in V2. Starting from 7.2 ± 0.1 mg chl-a/L (Figure 3c), the maximum reached 98.5 ± 3.1 mg chl-a/L on day 14. The adaptation phase was the shortest among all variants, lasting only 3 days (Figure 3a). A 6-day exponential phase followed, with an average growth rate of 12.4 ± 0.9 mg chl-a/L·d, about 25% higher than in V1. The stationary phase lasted 5 days, during which chlorophyll a stabilized near 100 mg chl-a/L, while the dieback phase caused minimal pigment loss (<0.3%) (Figure 3c).

In comparison, Ref. [32] investigated the effect of CO₂ on Spirulina platensis PCC 9108. The highest chlorophyll a content (21.86 mg chl-a/L) was observed at 8% CO₂. Despite a similar initial concentration (7.9 ± 0.1 mg chl-a/L), V3 showed the lowest chlorophyll a accumulation. The lag phase extended to 5 days, indicating a longer adaptation period, likely due to sub-optimal conditions (Figure 3a). During the exponential phase (days 5–10), chlorophyll an increased by ~49 mg chl-a/L at an average rate of 9.9 ± 0.2 mg chl-a/L·d (Figure 3b). The maximum pigment concentration on day 15 was 78.57 ± mg chl-a/L, about 20% lower than in V1 and more than 25% lower than in V2 (Figure 3c,d). The stationary phase lasted 5 days, but pigment growth was limited, and the final chlorophyll a concentration (78.21 ± 4.0 mg chl-a/L) remained significantly lower (p ≤ 0.05) than in the other variants. The overall average growth rate was 9.6 mg chl-a/L·d, the lowest among all tested conditions.

For comparison, Ref. [33] cultured local Scenedesmus sp. and Chlorella sp. strains on raw sewage with access to air containing 0.04% CO₂. The highest chlorophyll a concentration was 19.04 mg chl-a/L for Scenedesmus sp., and 13.10 mg chl-a/L for Chlorella sp. on day 8. After the logarithmic growth phase ended (day 12), pigment content decreased to approximately 8.6–8.9 mg chl-a/L [33]. Analysis of the elemental and biochemical composition of T. subcordiformis biomass showed relatively stable values across the three experimental variants, although some differences indicate that gas supply conditions influenced microalgal metabolism (

Table 1. Characteristics of T. subcordiformis biomass depending on the experimental variant.

Parametr	Unit	V1	V2	V3
Volatile solids (VS)	[%FM]	89.3 ± 1.7 a	88.7 ± 3.7 a	88.3 ± 2.0 a
Total nitrogen (TN)	[mg/gTS]	47.6 ± 2.5 a	46.3 ± 1.3 a	48.1 ± 4.1 a
Total phosphorus (TP)	[mg/gTS]	8.7 ± 0.9 a	9.1 ± 1.4 a	9.0 ± 2.3 a
Total carbon (TC)	[mg/gTS]	539 ± 21 a	577 ± 27 b	554 ± 47 ab
Total organic carbon (TOC)	[mg/gTS]	470 ± 17 a	493 ± 12 b	483 ± 29 ab
Protein	[%TS]	29.7 ± 1.5 a	28.9 ± 0.9 a	30.0 ± 3.2 a
Lipids	[%TS]	7.4 ± 0.4 a	7.2 ± 1.1 a	8.1 ± 1.9 b
Saccharides	[%TS]	36.5 ± 2.2 a	37.2 ± 1.4 a	36.7 ± 3.1 a

Different superscript letters (a, b, ab) within the same row indicate statistically significant differences between the analyzed variables. The same letters indicate no significant difference.

).

The content of volatile solids (VS) ranged from 88.3 ± 2.0% (V3) to 89.3 ± 1.7% (V1), confirming a high proportion of organic matter in the dry biomass and indicating good quality for energy conversion (Table 1). Differences between variants were statistically insignificant (p > 0.05), suggesting that organic composition remained stable regardless of growth conditions. Total nitrogen (TN) ranged from 46.3 ± 1.3 to 48.1 ± 4.1 mg/g TS, with the highest value in V3, corresponding to a protein content of 28.9 ± 0.9% TS (V2) to 30.0 ± 3.2% TS (V3) (Table 1). Although differences were not significant (p > 0.05), a trend toward higher protein accumulation was observed in V3, the variant with lower CO₂ availability, possibly reflecting a metabolic shift toward nitrogen fraction synthesis under limited carbon conditions [34]. This aligns with previous observations that the C/N ratio in microalgae cells depends on inorganic carbon availability [35,36]. Specifically, when the C/N ratio is below 12:1, protein content in Chlorella vulgaris remains high (>57.3%), whereas ratios around 19:1 significantly reduce protein synthesis (p < 0.05) [36].

Total phosphorus (TP) ranged from 8.7 ± 0.9 mg/g TS (V1) to 9.1 ± 1.4 mg/g TS (V2), with differences within the measurement error, indicating that CO₂ availability had little effect on phosphorus metabolism. The largest differences between variants were observed in total carbon (TC) and organic carbon (TOC). Variant V2 had significantly higher TC (577 ± 27 mg/g TS) and TOC (493 ± 12 mg/g TS) compared to V1 (539 ± 21 and 470 ± 17 mg/g TS) and V3 (554 ± 47 and 483 ± 29 mg/g TS) (Table 1). This suggests that higher CO₂ availability promoted greater carbon accumulation and enhanced polysaccharide biosynthesis. Similar trends were reported in previous studies [27], where TC and TOC were highest in reactors supplied with 13.0 ± 1.0% CO₂ and lowest under atmospheric air conditions.

Key biochemical fractions were largely comparable across variants (Table 1). Lipid content ranged from 7.2 ± 1.1% TS (V2) to 8.1 ± 1.9% TS (V3), suggesting that limited CO₂ may have favored lipid synthesis as an alternative energy reserve. Sugar content was highest in V2 (37.2 ± 1.4% TS), consistent with the higher TC and TOC, reflecting intensified photosynthesis and carbohydrate accumulation under increased CO₂ availability [37,38,39]. Overall, T. subcordiformis biomass composition was relatively stable. Significant differences (p < 0.05) were observed in TC, TOC, and the proportions of sugars and lipids. Variant V2, with more CO₂-rich gas, favored carbon accumulation and polysaccharide synthesis, whereas limited CO₂ in V3 slightly enhanced protein and lipid content (Table 1). These results indicate that inorganic carbon availability is a key determinant of the metabolic profile of microalgae, with direct implications for CO₂ biosequestration efficiency and the potential use of biomass in bioenergy and value-added products.

3.2. CO₂ Biofixation and pH Changes

Analysis of the composition of the gases fed to the V-PBR showed that the T. subcordiformis culture effectively bioassimilates CO₂ and completely eliminates the pollutants contained in the exhaust gases. The concentrations of the main pollutants present in the tested exhaust streams at the inlet and outlet of the V-PBR are shown in

Table 2. Characteristics of the gas composition supplied to the V-PBR used for the cultivation of T. subcordiformis.

Component	Unit	Variant 1		Variant 2		Variant 3
		Inflow	Outflow	Inflow	Outflow	Inflow	Outflow
CO2	% obj.	13.4 ± 1.2	1.3 ± 0.2	14.9 ± 1.7	3.3 ± 0.2	0.04 ± 0.002	0.029 ± 0.003
N2	% obj.	75.8 ± 2.3	77.4 ± 0.8	73.7 ± 1.9	76.9 ± 1.1	78.2 ± 0.1	78 ± 0.1
O2	% obj.	8.2 ± 0.9	21.3 ± 0.4	6.9 ± 2.1	20.9 ± 0.4	21.1 ± 0.1	21.3 ± 0.1
CO	ppm	141 ± 14	0 ± 0	91 ± 31	0 ± 0	0 ± 0	0 ± 0
NOx	ppm	145 ± 33	0 ± 0	140 ± 19	0 ± 0	7 ± 1	0 ± 0
SOx	ppm	83 ± 11	0 ± 0	139 ± 11	0 ± 0	9 ± 1	0 ± 0
CO2UE	%	43.1 ± 3.1		46.7 ± 2.4		37.5 ± 1.2

.

In V1, where biogas combustion gases were used, the CO₂ concentration in the inflow was 13.4 ± 1.2%. In the effluent, it decreased to 1.3 ± 0.2%, corresponding to a removal efficiency of 90.3 ± 3.2% (Figure 4). The carbon dioxide utilization efficiency reached 43.1 ± 3.1% (Table 2). Oxygen levels increased from 8.2 ± 0.9% to 21.3 ± 0.4%, while nitrogen slightly rose from 75.8 ± 2.3% to 77.4 ± 0.8%. Initial concentrations of other monitored pollutants were CO—141 ± 14 ppm, NO_x—145 ± 33 ppm, and SO_x—83 ± 11 ppm; in the exhaust gas, these fell below the detection limit, indicating complete removal (100%) (Table 2). The high efficiency of CO₂ removal may be partially explained by the potential role of the Target of Rapamycin (TOR) signaling pathway in regulating microalgal metabolism [40]. TOR is a key regulator of cell growth and resource allocation in eukaryotic cells, including microalgae, integrating signals related to nutrient availability, energy status, and environmental stress [41]. In the context of autotrophic cultivation of T. subcordiformis, the presence of elevated CO₂ concentrations in the medium may stimulate the activity of photosynthetic enzymes, leading to enhanced carbon assimilation. By modulating protein synthesis and carbohydrate metabolism, the TOR pathway may facilitate efficient redistribution of carbon within the cell from photosynthetically fixed CO₂, through cellular biomass, to energy-related metabolites, including those associated with H₂ production [42]. In this way, TOR could support both high CO₂ sequestration efficiency and the optimization of biomass and energy metabolite production under conditions of increased carbon availability, highlighting its potential role in the adaptation of microalgae to CO2-rich waste gas streams [43].

The decrease in CO₂ concentration strongly correlated with a rapid pH increase in the medium from 7.20 ± 0.04 to 8.9–9.0 (Figure 4 and Figure 5), reflecting intensive CO₂ uptake by the growing microalgae population. Biomass growth dynamics support this mechanism: the fastest increase occurred on days 5–10, coinciding with the sharpest CO₂ decrease and pH rise, highlighting the link between photosynthetic CO₂ assimilation and medium alkalisation [44].

Alkalinisation primarily occurs through uptake of CO₂ and HCO₃⁻, but additional processes contribute. During growth, microalgae secrete extracellular polymeric substances (EPS) that bind cations (e.g., Ca²⁺, Mg²⁺, Na⁺), altering the local ion balance. This reduces free ion concentrations and shifts the carbonate equilibrium toward alkaline forms (HCO₃⁻ and CO₃²⁻) [45]. Furthermore, enzymatic activity, particularly of carbonic anhydrase on the cell surface, promotes the conversion CO₂ ↔ HCO₃⁻. Since photosynthetic CO₂ consumption predominates, additional protons are removed from the solution, further increasing pH. Under air-saturated conditions, with low CO₂ availability (~0.04%), these effects are amplified, as microalgae activate carbon concentration mechanisms and enhance EPS secretion, together causing pronounced medium alkalisation [45].

In V2, where gases from coal combustion plants were used, the CO₂ concentration in the inflow was 14.9 ± 1.7% (Table 2). At the V-PBR outflow, CO₂ decreased to 1.8 ± 0.3%, corresponding to a removal efficiency of 77.8 ± 2.4% (Figure 4), while the carbon dioxide utilization efficiency reached 46.7 ± 2.4% (Table 2). During cultivation, the medium pH gradually rose from 7.01 ± 0.08 to 8.68–8.90 (Figure 4), and biomass increased from 250 ± 21 mg/L to approximately 2700 ± 62 mg/L (Figure 2c). The correlation between decreasing CO₂ in the off-gas and rising pH was strong (R² = 0.9421), though the slope was less steep than in V1 (Figure 5). Similarly, in [28] with Arthrospira platensis, CO₂ biostorage ranged from 87 to 96%, the pH reached 9.33 during growth, and biogas feeding enabled complete removal of CO, NO_x, and SO_x. In that study, O₂ increased from 6.9 ± 2.1% to 20.9 ± 0.4%, and N₂ from 73.7 ± 1.9% to 76.9 ± 1.1%, while CO, NO_x, and SO_x concentrations fell below detection limits.

The intensive conversion of CO₂ to HCO₃⁻ during the cultivation of T. subcordiformis can be largely attributed to enzymatic activity, particularly that of cell-surface carbonic anhydrase (CA) [46]. This enzyme catalyzes the reversible reaction CO₂ ↔ HCO₃⁻, facilitating the rapid transformation of CO₂ captured from the medium [47]. Under autotrophic conditions, where photosynthesis is the primary mechanism of CO₂ consumption, this reaction leads to efficient carbon fixation and the simultaneous removal of protons from the solution, resulting in the observed increase in medium pH [48]. This mechanism not only supports high CO₂ capture efficiency but also contributes to the stabilization of the bioreactor environment, creating favorable conditions for further biomass growth and the production of energy-related metabolites [49]. In other microalgae, such as Chlorella vulgaris and Scenedesmus sp., CA also accelerates the hydration of CO₂, but the enzyme can be localized both in the chloroplasts and on the cell surface. In these species, CO₂ uptake often leads to proton release, resulting in a decrease in the medium pH during intensive photosynthesis [50]. In marine microalgae, such as Dunaliella salina and Phaeodactylum tricornutum, cell-surface CA enables efficient CO2 conversion even at low dissolved concentrations, enhancing carbon assimilation and promoting biomass growth [51]. In higher plants, such as wheat and tobacco, CA is predominantly localized within chloroplasts and supports the Calvin cycle, i.e., intracellular CO₂ assimilation. In this case, the enzyme does not directly affect apoplastic pH, and its primary role is to improve photosynthetic efficiency rather than alter the external environment [52].

In V3, supplied with atmospheric air, the initial CO₂ concentration was 0.04 ± 0.002%, decreasing to 0.029 ± 0.002% in the effluent, corresponding to a CO₂ utilization efficiency of 37.5 ± 1.2% (Table 2). Other gas components changed minimally: O₂ increased slightly from 21.1 ± 0.1% to 21.3 ± 0.1%, and N₂ decreased from 78.2 ± 0.1% to 78.0 ± 0.1%. Traces of NO_x (7 ± 1 ppm) and SO_x (9 ± 1 ppm) were completely removed during cultivation of T. subcordiformis (Table 2). Previous studies confirm that CO₂ removal efficiency decreases as feed concentration rises. For example, Ref. [53] reported efficiencies of 81.82%, 45.10%, 44.04%, and 38.73% at CO₂ feed concentrations of 2%, 5%, 10%, and 15%, respectively. This trend reflects the limited capacity of microalgae systems to assimilate higher CO₂ concentrations [53]. Our results showed a similar relationship: increasing CO₂ concentration in the feed stream corresponded with a decline in relative capture efficiency.

The implementation of CO₂ capture technologies from exhaust gases of varying origin and composition, based on the use of microalgae, including T. subcordiformis, involves multiple challenges that must be considered when assessing its industrial potential. The primary barriers relate to high capital and operational costs [54] as well as technological and process complexity [55]. It has been repeatedly demonstrated that the construction and subsequent operation of photobioreactors (PBRs) require substantial financial investment [56]. Similarly, the implementation and functional integration of modules for exhaust gas purification, targeting NO_x, SO_x, or particulate matter, are highly resource-intensive. Studies have shown that, in many cases, without highly efficient pre-treatment of exhaust gases, the long-term viability of microalgal cultures cannot be maintained [57].

Operational costs have been identified by many researchers as a critical limitation [58]. Cultivation of T. subcordiformis requires stable light and temperature conditions. In temperate climates, intensive artificial lighting is necessary, which can account for 40–50% of total system costs [59]. Regular supplementation with nutrient-rich growth media, containing nitrogen, phosphorus, and trace elements, is also required, further increasing financial demands [60]. Biomass harvesting processes—separation, dewatering, and concentration—are particularly energy-intensive; at low cell concentrations (<3.0 g VS/L), they can consume 20–30% of operational expenses [61]. Consequently, 60–70% of total operating costs may be attributed to auxiliary operations (lighting, nutrient supply, harvesting) rather than to the CO₂ bio-assimilation itself.

From a technological standpoint, maintaining the purity and stability of microalgal cultures over extended periods remains a challenge [62]. Cultures are susceptible to bacterial contamination, invasion by other microorganisms, and biofilm formation, leading to decreased photosynthetic efficiency and potentially requiring costly interventions, such as inoculum replacement. It should be noted that most studies to date have been conducted under laboratory conditions, which do not fully reflect practical challenges. In particular, there is a lack of data on the long-term effects of complex exhaust gas components on the physiology and stability of T. subcordiformis. Scaling up processes to large volumes also presents engineering challenges, including uneven light distribution, mixing, and gas exchange [63]. The absence of comprehensive pilot- or demonstration-scale studies currently prevents reliable assessment of long-term performance and stability of microalgal systems under industrial conditions [64]. Overcoming the barriers outlined above requires further research into the resilience of T. subcordiformis strains to gas contaminants, optimization of photobioreactor design, reduction in biomass harvesting costs, and integration of cultivation systems with renewable energy sources. Advancements in these areas are essential to enable the technology to become a feasible industrial solution, simultaneously addressing CO₂ emission reduction and the production of valuable biomass.

3.3. Hydrogen Production

V1 was characterized by an efficiency of hydrogen production of 68.4 ± 3.7 mL/gVS (Figure 6a,d). The reaction rate constant was 0.030 l/gVS·h, and the H₂ production rate was 2.05 ± 0.11 mL/gVS·h. The H₂ content in the biogas was 58.2 ± 1.1%, CO₂ 40.6 ± 1.2% and oxygen 1.2 ± 0.2% (

Table 3. Comparison of the efficiency and kinetics of H₂ production and the composition of the biogas produced by T. subcordiformis as a function of the experimental variant tested.

Parameter	Unit	Value
		V1	V2	V3
H2	mL/gVS	68.4 ± 3.7	70.9 ± 2.7	66.9 ± 3.0
H2	mL/L	181.9 ± 9.8	192.1 ± 7.3	149.8 ± 6.7
k	1/gVS·h	0.03	0.032	0.028
r	mL/gVS·h	2.05 ± 0.11	2.268 ± 0.08	1.87 ± 0.06
H2	%	58.2 ± 1.1	59.1 ± 2.3	58.9 ± 1.7
CO2	%	40.6 ± 1.2	39.7 ± 2.0	39.5 ± 1.1
O2	%	1.2 ± 0.2	1.2 ± 0.1	1.6 ± 0.3

). V2 showed the highest H₂ production efficiency among the analyzed variants and reached 70.9 ± 2.7 mL/gVS (Figure 6b,d). At the same time, the highest reaction rate constant was measured at 0.032 1/gVS·h and the maximum hydrogen production rate was 2.27 ± 0.08 mL/gVS·h (Table 3).

The qualitative composition of the biogas showed the highest proportion of hydrogen with 59.1 ± 2.3% and 39.7 ± 2.0% CO₂. In turn, in the study [18], the highest biohydrogen production efficiency achieved was 58.2 ± 6.7 mL/g VS with a carbon dioxide content in the culture medium of 220.0 ± 8.0 mg/L. The percentage of biohydrogen in the biogas was 38.2 ± 1.4% [18]. Variant V3 was characterized by the lowest process efficiency (Table 2, Figure 6d).

The total H₂ production was 66.9 ± 3.0 mL/gVS (Figure 6c), and the production rate reached 1.87 ± 0.06 mL/gVS·h, which was significantly lower compared to the other variants. The reaction rate constant was also the lowest at 0.028 1/gVS·h. The gas composition (58.9 ± 1.7% H₂; 39.5 ± 1.1% CO₂; 1.6 ± 0.3% O₂) was comparable to the other variants (Table 3). To evaluate the potential of different microalgae strains for biohydrogen production, experiments were conducted with Chlorella pyrenoidosa, Scenedesmus obliquus and Chlorella sorokiniana [65]. The maximum hydrogen production was 45.50 ± 0.32 mL/g VS for C. pyrenoidosa, 38.43 ± 0.42 mL/g VS for S. obliquus and 34.83 ± 1.82 mL/g VS for C. sorokiniana [65]. The results obtained in our own studies were higher than those reported above but did not deviate from the frequently reported values typical of T. subcordiformis. Among the microalgae analyzed in the literature, T. subcordiformis stands out for its relatively high potential for biohydrogen generation. However, the efficiency of the process is strongly dependent on cultivation conditions as well as on the initial experimental parameters, particularly the inoculum density. This is confirmed by Ji et al. [66], who observed that increasing the biomass concentration from 0.5 g TS/L to 3.2 g TS/L resulted in nearly a tenfold increase in the H₂ evolution rate. In practical terms, this translated into an increase in hydrogen yield from 16 mL H₂/g TS to 49 mL H₂/g TS. A detailed quantitative comparison of hydrogen production efficiencies among different microalgal species, including the relative advantages of T. subcordiformis, has already been presented in our earlier review [67].

Based on the biomass concentrations of T. subcordiformis introduced into the growth medium, the hydrogen production yields per unit reactor volume were 181.9 ± 9.8 mL/L (V1), 192.1 ± 7.3 mL/L (V2), and 149.8 ± 6.7 mL/L (V3) (Table 3). These values were statistically significant (p < 0.05). Previous studies on T. subcordiformis indicate a wide range of achievable biohydrogen yields, reflecting the high sensitivity of this species to environmental conditions and process strategy. Ji et al. (2011) demonstrated that nitrogen limitation combined with CCCP treatment promoted hydrogen accumulation of 55.8 mL/L, whereas sulfur deprivation resulted in only 15.2 mL/L [68]. These results clearly show that nitrogen deficiency is a stronger factor driving metabolism towards biohydrogenogenesis than sulfur limitation, correlating with observed reorganization of protein and photosynthetic metabolism.

Significant improvements have also been achieved by integrating biohydrogen production with an alkaline fuel cell (AFC). Guo et al. (2016) reported that in a photobioreactor integrated with an AFC and using CCCP, hydrogen yields of 78 ± 5 mL/L were achievable [69]. Furthermore, subsequent optimization of the light regime demonstrated that a light/dark cycle of 9/6 h increased production to 126 ± 10 mL/L [70], highlighting the critical importance of tuning photosynthetic and mitochondrial respiration dynamics to a redox balance favorable for hydrogenase activity. Recent studies focusing on the use of agricultural waste and wastewater as growth media indicate the practical applicability of the technology. Under respirometric conditions, T. subcordiformis produced total H₂ volumes of approximately 150–170 mL/L [17]. The results obtained in the present study fall within the upper range of hydrogen production yields reported in the literature for T. subcordiformis (Figure 5, Table 3).

It should be emphasized that, despite demonstrating the technological feasibility of a system integrating fuel combustion, CO₂ sequestration using Tetraselmis subcordiformis, and subsequent biohydrogen production, this study does not include a detailed economic analysis. The absence of estimates for capital expenditures (construction of photobioreactors, exhaust gas purification systems, lighting infrastructure), operational costs (growth medium, electricity, biomass separation and dewatering), and potential revenue streams (biofuels, feed, biohydrogen) currently represents a significant limitation for the practical assessment of the proposed technology.

It is important to emphasize that the results presented here were obtained under controlled laboratory conditions, where factors such as illumination, gas flow distribution, and culture stability were tightly regulated. While this allowed for high CO₂ removal efficiencies (up to 90.3 ± 3.2%) and biohydrogen yields of 70.9 ± 2.7 mL/gVS, translating these outcomes to industrial-scale systems is far more complex. Large photobioreactors are subject to uneven light penetration, mass transfer limitations, higher risk of contamination, and significantly greater operational costs related to lighting, mixing, and biomass harvesting. Consequently, the laboratory-scale findings should be regarded as a proof of concept, whereas the evaluation of their industrial feasibility requires pilot- and demonstration-scale studies combined with comprehensive techno-economic analyses. It should also be noted that results obtained under laboratory conditions cannot serve as a reliable basis for estimating economic performance, due to scale-up challenges. Parameters characterizing small experimental systems, such as gas and liquid flow rates, working pressures, mass and energy transfer, and long-term stability of productivity, differ substantially from those encountered in full-scale installations. Consequently, it is necessary to advance the technology readiness level (TRL) to include pilot- and demonstration-scale systems operated under conditions close to real-world scenarios. Only at this stage will it be possible to reliably identify key process- and cost-driving factors and perform a comprehensive techno-economic assessment.

Future research should prioritize the integration of technological findings with detailed economic modeling, enabling an evaluation of the real competitiveness of the proposed solution relative to other CO₂ mitigation technologies and alternative energy sources.

4. Conclusions

Cultivation of T. subcordiformis in V-PBRs fuelled with different gas sources showed a high CO₂ removal efficiency, exceeding 90% in the biogas off-gas variant. These results indicate that microalgae can play a significant role as biological absorbers of exhaust gases, even in mixtures containing pollutants typical of fuel or biofuel combustion.

The growth analysis of T. subcordiformis revealed significant differences between the variants. The highest growth rate and final biomass concentration was achieved in the V-PBR fuelled with exhaust gases from the co-combustion of coal and biomass. However, the use of atmospheric air resulted in the lowest yields and the lowest dynamics of microalgae biomass growth, confirming that low CO₂ concentrations could limit the intensity of photosynthesis and biosynthesis.

The biochemical profile of the microalgae varied considerably between the variants. The highest total carbon and carbohydrate contents were observed when the V-PBR was fuelled with exhaust gases from the co-combustion of coal and biomass. When the system was fed with air, a higher proportion of lipids and proteins was observed, which could indicate a shift in metabolism towards nitrogen and lipid fractions under carbon stress conditions.

The highest efficiency and hydrogen production rate were observed in the variant using the exhaust gases of a municipal energy company, while significantly lower values were observed when air was used as a CO₂ source. The results confirm that carbohydrate-rich biomass is the most effective substrate for biohydrogenogenesis, while the presence of the lipid fraction can limit the process rate. The composition of the biogas remained stable regardless of the variant.

The results of the study demonstrate the feasibility of integrating off-gas systems with photobioreactors for simultaneous CO₂ biostorage and biohydrogen production. The use of such systems promotes the reduction in greenhouse gas emissions, improves the energy balance of the plant and supports the achievement of climate neutrality and the goals of circular economy.