Floating Light Ball Reactor: A Scalable and Flexible Photobioreactor Design for Microalgae Cultivation

Abstract

Bringing microalgae cultivation to industrial large-scale production needs photobioreactor designs suitable for economic scale-up to large volumes. Current designs using sunlight lack around-the-clock cultivation, are location dependent, and, when using artificial lighting, do not possess the ability for economic scale-up due to physical connection of lighting to the photobioreactor or thin structures such as tubes or plates. We present a photobioreactor design where floating light balls are used to bring light into the reactor and therefore allow for a flexible and easy scale-up in three dimensions. The physical separation of light balls and reactor housing allows for the transformation of empty containers into photobioreactors. Using a modular and flexible design with assemblies of light balls, we present the transformation of a metal cylinder and a brewery tank into floating light ball reactors for microalgae cultivation. The diatom Phaeodactylum tricornutum was successfully grown on 22-L and 234-L scale in our reactors up to 0.9 g/L dry biomass. Potential biofilm formation can be flexibly handled by quickly cleaning or replacing the light ball assemblies, while at the same time harvesting the biofilm biomass. Our concept is transferrable to all cultivation and reaction systems where light needs to be brought efficiently into a liquid.

Keywords

Algae technology - Scale-up - Photobioreactor - Floating Light Ball Reactor

Significance

We demonstrate a system for microalgae cultivation using floating light balls that are submerged in a cultivation medium. This design allows for the transformation of empty containers into photobioreactors. The light balls are not physically connected to the container and therefore can be easily added, removed, or cleaned. This design offers drastic advantages in the economic scalability of photobioreactors and the use of existing containers therefore. Successful algae cultivation on a 22-L and 234-L scale demonstrates the feasibility of this newly developed photobioreactor system.

Introduction

Microalgae cultivation yields carbon-rich biomass and value-added materials via CO₂ sequestration. This provides a sustainable alternative feedstock to crude oil for subsequent manufacture of a wide range of carbon-based products by microalgae.[1] In order to contribute hereto, it is necessary for microalgae technology to enter industrial large-scale production. However, current production systems do not achieve this.[1d] [2] In Malaysia, the world’s largest microalgae production facility currently has an annual capacity of 350 metric tonnes microalgae per year. In the 1950s, artificial industrial microalgae cultivation started in photobioreactors using sunlight[3] as free energy source with its natural limitations of sunshine duration, location dependency, and process regulation after sunlight intensity. All light-driven processes in solution face the challenge of a limiting penetration depth of light.[4] In case of a dense microalgae solution (ca. 1 g/L), this results in complete light adsorption on a pathway of typically 5–6 cm.[5] To account for this physical boundary, popular commercially available designs of photobioreactors enable an improved light distribution and efficient light input by being either one-dimensional (tubes), two-dimensional (plates, bags, or pond), or branched.[5c] [6]

Compared to non-light-controlled biotechnological cultivation, these designs pose additional challenges in terms of economical access to the large quantities required, low-cost savings in further scaling, land demand, and extensive cleaning efforts due to the large illuminated surface structure. While sunlight has been the predominant light source until now, recent advances in LED technologies have made high-power LED available as cheap mass good with efficiencies exceeding 50% and therefore less heat dissipation to cope with.[7] Allowing for round-the-clock cultivation and location independency, these advantages put us now in a position to surpass purely sun-based-driven microalgae cultivations. With artificial lighting, all the process parameters of microalgal cultivation can be actively controlled and optimized for the desired outcome for the first time. Therefore, the much-desired optimization that allows the maximum algal growth rates can be achieved with the application of artificial intelligence.[8] Flat-panel reactors are successfully used for the artificially illuminated cultivation of Phaeodactylum tricornutum and the production of algal omega-3-oil, e.g. by VAXA Technologies and other companies.[9] In 2012, Buchholz et al. were the first to introduce the concept of internal lighting by using dynamic freely moving illumination elements combined with wireless energy transfer by an electromagnetic field for the microalgae cultivation of Chlorella vulgaris. However, the system was limited to lab scale and, to best of our knowledge, has only been reproduced once by one other research group in 2017.[6b] [10] Besides using tubes and plates, alternative new designs of photobioreactors focusing purely on artificial lighting that transit from the laboratory scale and find a lasting adoption from industry or other research groups are rare.[1d] [11] An externally illuminated 1250-L system is commercially available from Industrial Plankton with worldwide customer adoption. The concept of submerged light rods has been successfully applied by companies such as Brevel and BrightWave.[12]

Although the existing photobioreactor technologies are being highly adopted for the economical cultivation of microalgae for obtaining high-priced products (e.g., omega-3 fatty acids, astaxanthin), the construction of these technologies does not allow reaching the mass market or potentially contributing to the replacement of crude oil as a raw material for the chemical industry.[2] According to our analysis, this is due to the fact that photobioreactor designs so far do not allow further significant savings of investment cost (CAPEX) upon scaling up to the larger industrial level.

In our efforts to contribute to overcoming the current limitations (i.e., light distribution and scaling constraints, high investment costs, biofouling and cleaning challenges, process control limitations and economic viability), we developed a cultivation system using floating light balls as illumination sources being submerged in the microalgae cultivation medium.[13] The physical separation of the light balls and reactor housing allows for the transformation of any container into a photobioreactor. Thereby, our overall objectives are the development of a new photobioreactor system based on artificial lighting with improved three-dimensional light introduction, scalability, applicability with the use of existing containers, low-cost serial manufacturing, flexible handling of possible biofilm formation, and the successful long-term cultivation at different scales.

Results and Discussion

Design of the Floating Light Ball Reactor

Our floating light ball reactor was realized using assemblies of light balls comprising LEDs and a connection to a power supply. To generate a photobioreactor, we put these assemblies of light balls into a suitable container ([Fig. 1]). As the assemblies are floating in the tank within the aqueous cultivation medium, they can be easily removed, added, or cleaned. Even the harvesting of the biofilm or deposits from the light balls outside the photobioreactor is possible. A balanced buoyancy of the light balls compared to the medium ensures minimal static forces compared to submerged cylindric light rods. Therefore, being a pluggable concept, the assemblies can be manufactured from a single light ball. This comes with advantages such as on-site assembly, easier transport, introduction in containers via available size-limited manholes, and the possibility of making longer assemblies using the same material thickness compared to cylindric light rods.

Fig. 1 Design of a Floating Light Ball Reactor and transformation of a brewery tank into a photobioreactor by adding assemblies of light balls. (a) One light ball submerged in a 22-L metal cylinder filled with water. (b) 300-L brewery tank used in this work. (c) CAD model of light ball assemblies realized in the brewery tank. (d) Top view after light ball installation. (e) Top view during cultivation of Phaeodactylum tricornutum.

By adjusting the size, number, and spatial arrangement of the light balls in the floating light ball reactor, the amount of light can be adjusted according to microalgae cultivation requirements following the growth curve. In this approach, the main part of the photobioreactor is the individual light ball. Scale-up of this technology to large cultivation volumes can be achieved easily by using more light balls, which are not physically attached to the cultivation containers, but floating in the medium and currently loosely fixated with a top suspension.

Thus, the advantages and cost savings of serial production come into play when scaling up this technology to large cultivation volumes – a unique selling point that no other microalgae photobioreactor technology currently offers. The physical separation of the light balls and the reactor housing allows for the transformation of any container into a photobioreactor. Existing reactors or containers, as well as commercially available brewery tanks, storage tanks, or international bulk containers (IBCs), which often anyway comprise all the necessary base elements (i.e., connections for charging/discharging, a circulation pump and/or inside mixing apparatus, a jacket for cooling, level sensor, temperature and pH-measurement, and a gas injection or gas exchange) for microalgae cultivation except for the light source are transformed into a working photobioreactor by the addition of the light-ball assemblies.

Construction of the Light Balls and Assemblies

The individual light ball is a low-component system comprising six dimmable high-power LEDs (each LED up to 25 W neat lighting power) being attached to an aluminum-cuboid, fixated in a water-proof transparent PVC housing with electrical wiring, and a cooling circuit attached to the aluminum-cuboid (see [Fig. 2]). Light intensity can be varied using a frequency dimmer. Several light balls can be combined with a rod spacer in variable lengths to form an illumination assembly comprising two or three light balls ([Fig. 1c]). In our case, a 10-cm tubular spacer was used to account for the maximum penetration depth by the light of 5 cm. The individual light ball has a diameter of 13 cm and a height of 17 cm, a total outer illumination surface of 0.080 m², and a displacement volume of 1.2 L with an oval spheroid shape.[14] We were also able to include sensors such as a temperature sensor into the light ball.

Fig. 2 Explosion drawing showing the nine main components of one light ball as a 3D model (left) and after building (right).[14]

Transformation of a metal cylinder and a brewery tank into photobioreactors (cultivation volumes: 22 L and 234 L), and operation and cultivation of P. tricornutum therein

22-L Scale (FLBR-1)

A stainless-steel metal cylinder (diameter 30 cm and height 40 cm) was equipped with one light ball, a pH-probe, and a gas inlet.[14] The so-obtained photobioreactor was operated in an air-conditioned environment (22 °C) leading to cultivation temperatures of 20.0–25.0 °C. Air injection at the bottom with 240 L/h provided circulation and mixing. Neat CO₂ from a gas cylinder was injected into the airflow inlet using a solenoid valve, controlled by a pH probe and a PID system to keep pH stable at 8.0–8.2. Microalgae cultivation was performed using P. tricornutum in 22 L F/2-type medium.[14] Starting with an inoculum concentration of 0.05 g/L, a typical reasonable growth curve was obtained, and after 19 days, the stationary phase was reached with an microalgal dry biomass concentration of 0.63 g/L.[14]

234-L Scale (FLBR-2)

To further demonstrate the feasibility of our concept on a larger scale, a commercially available 300-L brewery tank was equipped with five illumination assemblies (four assemblies with two light balls and one assembly with three light balls; [Fig. 1] c,d), a pH probe, a gas inlet, and a circulation pump. The total of eleven light balls were spatially arranged to mimic a cubic close packaging with an assumed maximum light-penetration depth of 5 cm. Light balls comprising LEDs with a warm white spectrum of 3500 K, total illumination surface of 0.88 m², and neat lighting power of 450 W were used. The light balls were connected to an external cooling device to prevent overheating of the LED. An additional external cooling device was applied to the brewery’s tank jacket to maintain the cultivation temperature at 20.0–21.5 °C. Air injection at the bottom with 300 L/h provided gas exchange. Neat CO₂ from a gas cylinder was injected into the airflow inlet using a solenoid valve, controlled by a pH probe and a PID system to keep pH stable at 8.0–8.2. P. tricornutum was successfully grown on a 234-L scale comparable to FLBR-1. Starting with a microalgal inoculum concentration of 0.2 g/L after 28 days, an algal dry biomass concentration of 0.87 g/L was obtained (see [Figs. 1e] and [Fig. 3]).[14]

Fig. 3 Growth curve of Phaeodactylum tricornutum in FLBR-2 at the 234 L scale.[14]

Cleaning and Handling of Deposits and/or Biofilm

Deposits and/or biofilm formation on the light balls were observed during cultivation. After three weeks of cultivation, the light balls were removed from the FLBR-2. The light balls were cleaned outside, and the biofilm and deposits on the light balls were harvested. Simple rinsing off using a lab spray bottle and wiping with a plastic scraper were sufficient for cleaning and re-introduction into the FLBR-2 within four hours to continue the cultivation. This quick cleaning outside of the setup and the possibility of simultaneous biofilm harvesting of algal deposits demonstrate the flexibility of the floating light ball reactor design. In the case of P. tricornutum, the biofilm was not adhering stubbornly to the light balls, allowing for both easy cleaning and the simultaneous harvesting outside the photobioreactor.[14] Therefore, the floating light ball reactor design offers advantages compared to the existing photobioreactor systems, in that the biofilm formation does not result in a mandatory pause in the cultivation; furthermore, extensive chemical or mechanical cleaning efforts with long production downtime are not required. The cultivation takes place as a manageable, even harvestable side stream, with minimal cultivation downtime. Depending on the specific microalgal cultivation objective, different operation modes are possible within the floating light ball reactor: (1) cultivation in the medium and on the surface simultaneously, (2) focus on cultivation in the medium with regular replacement of biofilm-covered light ball assemblies with clean ones, or (3) biofilm-focused culture on the light ball surfaces.

Further Scale-up to 757 L: Assessment of Economic Scalability, Modularity, and Flexibility

For a small series of 50 light balls, the material costs per light ball are already less than 40 EUR, providing 150 W neat illumination power per light ball on a 0.080 m² surface.[14] When manufacturing a higher number of light balls, these costs will be significantly reduced due to the benefits of serial manufacturing. We consider this an important advantage for further economic growth in cultivation volume, whether through larger individual containers or by increasing the number of standard container systems. Applying the floating light ball reactor concept to a standard stainless steel 1000-L international bulk container (IBC) with jacket, and using the same type of light balls as for the 22- and 234-L setup would require, e.g., 35 light balls ([Table 1]). With such a design as shown in our CAD model ([Fig. 4]), an internal illumination surface of 2.8 m² with a maximum possible neat lighting power of 5.3 kW would be realized inside the so-obtained photobioreactor, resulting in an effective cultivation volume of 757 L. This case study demonstrates the versatility of the floating light ball reactor approach. By adjusting, e.g., size, material, and shape of the individual light balls as well as the spatial arrangement in a certain container, great flexibility can be achieved for a given algal cultivation task. Easy and temporal transformation of containers, storage tanks, and reactors of differing volumes and dimensions into photobioreactors will facilitate further adoption of algae cultivation in different industries and by stakeholders (i.e., winery, brewery industry, and farmers).

Table 1 Realization of floating light ball reactors at different scales using the same light balls

Scale	# Light ballsa	Lighting surface Light powera	Status
22 L	1	0.0036 m2/Lb 6.8 W/L	Algae cultivation demonstrated
234 L	11	0.0038 m2/L 7.1 W/L	Algae cultivation demonstrated
757 L	35	0.0037 m2/L 6.9 W/L	Case study ([Fig. 4])

^aEach light ball is of 13 cm diameter, 17 cm height, 0.080 m² surface equipped with 6 × 25 W LED neat lighting power.[14]

^bEffective cultivation volume.

Fig. 4 Design model for transformation of a 1000 L stainless-steel IBC into a 757-L Floating Light Ball Reactor using 35 light balls.

Conclusions

We have developed a versatile photobioreactor design based on assemblies of light balls that are floating in the cultivation medium, without the need for a physical connection to the bottom or wall of the cultivation system. This allows for the three-dimensional light entry into algal-cultivation systems, promising economic scale-up to large cultivation volumes. The low component system is modular and adaptable to different light-ball sizes and spatial arrangements. A metal cylinder of 22-L and a brewery tank of 234-L scale were transformed into floating light ball reactors. Successful cultivation of the diatom P. tricornutum with artificial lighting was demonstrated in these systems. We believe this approach to be pushing the boundaries of algal-cultivation systems based on artificial illumination and could further expand the technology of any light-driven reaction system.

Contributors’ Statement

Timo Gehring: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Visualization, Writing - original draft, Writing - review & editing. Patrick Maurer: Conceptualization, Funding acquisition, Methodology, Supervision, Writing - review & editing. Mutlu Yildirim: Formal analysis, Investigation, Methodology, Validation, Visualization, Writing - review & editing. Maurice Siegfried Lierse: Formal analysis, Investigation, Methodology, Validation, Visualization, Writing - review & editing. Richard Ickes: Investigation, Methodology, Validation, Visualization, Writing - review & editing.

Conflict of Interest

Timo Gehring and Patrick Maurer filed patent applications on the presented photobioreactor technology (EP24190886.2, PCT/EP2025/070090, PCT/EP2025/066640 and EP24182686.6). Timo Gehring holds a registered trademark for the abbreviation FLBR® (EUIPO Wordmark 019090215) and a picture mark (EUIPO 019142975). The authors declare no other conflict of interest.

Acknowledgment

We are grateful for support and discussions with Louisa Rau, Christina Karhan, and Elias Friedrich. Thank you Sarah Leis for drawing the 3D models, Maria Reinhard and Anas Boumia for medium preparation and inoculation, and Nour Albaali for strain management and growing of precultures.

• References

• 1a Xu P, Shao S, Qian J. et al. Bioresour Technol 2024; 398: 130528
• 1b Posten C, Schaub G. J Biotechnol 2009; 142 (01) 64-69
• 1c Stephens E, Ross IL, King Z. et al. Nat Biotechnol 2010; 28 (02) 126-128
• 1d Guieysse B, Plouviez M. Front Bioeng Biotechnol 2024; 12: 1359755
• 2 “Why are we still so far away from mass market” Round Table Discussion, 1st International Congress on Algae Biotechnology 2025 Lisbon, Portugal https://algaebiotechnology.pt/ (accessed 2025-08-30)
• 3 J Agric Food Chem. 1953; 1 (09) 596-597
• 4 Gilmore K, Seeberger PH. Chem Rec 2014; 14 (03) 410-418
• 5a Cho C, Nam K, Seo YH. et al. Sci Rep 2019; 9 (01) 1723
• 5b Benner P, Meier L, Pfeffer A, Krüger K, Oropeza Vargas JE, Weuster-Botz D. Bioprocess Biosyst Eng 2022; 45 (05) 791-813
• 5c Posten C. Eng Life Sci 2009; 9 (03) 165-177
• 6a de Vree JH, Bosma R, Janssen M, Barbosa MJ, Wijffels RH. Biotechnol Biofuels 2015; 8: 215
• 6b Heining M, Buchholz R. Biotechnol J 2015; 10 (08) 1131-1137
• 6c Legrand J, Artu A, Pruvost J. React Chem Eng 2021; 6 (07) 1134-1151
• 7 Kusuma P, Pattison PM, Bugbee B. Hortic Res 2020; 7: 56
• 8a Yeh Y-C, Syed T, Brinitzer G. et al. Bioresour Technol 2023; 390: 129882
• 8b Syed T, Krujatz F, Ihadjadene Y. et al. Comput Biol Med 2024; 172
• 8c Zhang F, Li Z, Chen C. et al. Adv Mater 2024; 36 (03) 2303714
• 9a Neumann U, Derwenskus F, Flaiz Flister V, Schmid-Staiger U, Hirth T, Bischoff SC. Antioxidants 2019; 8: 6
• 9b Bashan O, Bashan O, Drummey S. US Patent 11,912,966 B2 2024
• 10a Buchholz R, Heining M, Lindenberger C, Manstetten P. European Patent EP2719753B1; 2015
• 10b Murray AM, Fotidis IA, Isenschmid A, Haxthausen KRA, Angelidaki I. Algal Res 2017; 25: 244-251
• 11 Deprá MC, Dias RR, Zepka LQ, Jacob-Lopes E. Processes 2025; 13 (01) 51
• 12 Golan I. US Patent 2023051997A1 2023
• 13a Gehring T, Maurer P. European Patent EP24190886.2 filed 2024
• 13b Gehring T, Maurer P. European Patent EP24182686.6 filed 2024
• 14a Experimental section (Supporting Information (SI) for additional information)
• 14b Microalgae strains: Phaeodactylum tricornutum (SAG-1090a and SAG-1090b) were used in this work. Axenic precultures were maintained in a F/2-type medium (see SI) in 500 mL flasks on an orbital shaker at 22 °C, 100 μmol/m²/s illumination, light/dark cycle 18 h/6 h, Niello QB1000 LED with 3500 K light color. Light balls: PVC-U transparent housing, each light ball contains 6 × 50 W full-spectrum COB LEDs with 3500 K spectrum (model DOB4075 with average efficiency 2.9 J/μmol). For detailed component listing and manufacturing of the light balls, see SI. LED intensity can be varied using a frequency dimmer. Light intensity measurements were conducted using the ITC Parwise Smart Submersible Quantum (PAR) Sensor. Setup and microalgae cultivation in FLBR^®-2 (234 L): Five assemblies with a total of eleven light balls were used and arranged in a spatial arrangement as shown in Fig. 1c. Starting with a microalgal inoculum concentration of 0.2 g/L (OD₇₀₀ = 1.1) after 28 days an algal dry biomass concentration of 0.87 g/L was obtained (experiment MY-028). Light/dark cycle was L/D = 18 h/6 h. Air injection via a sterile filter (Midisart 2000 Sartorius, type 17804) at the bottom with 300 L/h provided gas exchange. A circulation pump with 7 m³/h was used for repumping algal suspension from the bottom valve to the container lid. pH was measured in the recirculation loop using InPro 3253I/SG/120 sensor (Mettler-Toledo) and neat CO₂ injection via a solenoid valve into the airflow inlet stream from a gas cylinder was controlled using Mettler-Toledo M800 PID controller to keep the pH stable at 8.0–8.2. A Huber Unichiller was used to keep the cultivation temperature at 20.0–21.5 °C, and a second minichiller (Huber Olé) was attached to the five light ball assemblies to prevent aluminium cuboid temperatures inside the light ball exceeding 50 °C (where the LEDs are attached). After four weeks of cultivation, the five assemblies were cleaned outside and after centrifugation (5 min, 5000 × g), a pellet of 245 g wet biomass of P. tricornutum was obtained (in addition to the growth in the medium; Fig. 3). Analytics: Optical density was measured at 700 nm (UviLine 9400). 100 mL samples from the cultures were centrifuged, rinsed twice (resuspending in water and centrifugation), and dried (50 °C overnight) to determine microalgal dry biomass concentration. All microalgae concentrations are given as dry biomass. See SI for further details on FLBR^®-1.

Publikationsverlauf

Eingereicht: 02. September 2025

Angenommen nach Revision: 09. Oktober 2025

Artikel online veröffentlicht:
07. November 2025