Development of a low-cost real-time monitoring system for biomass concentration and environmental factors in microalgae Limnospira fusiformis cultivation

doi:10.21203/rs.3.rs-4704238/v1

Download PDF

Research Article

Development of a low-cost real-time monitoring system for biomass concentration and environmental factors in microalgae Limnospira fusiformis cultivation

https://doi.org/10.21203/rs.3.rs-4704238/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

With increasing demand for microalgae, there is a need to reduce operational production costs and develop stable growth prediction methods. In this study, we have developed a low-cost and user-friendly monitoring and biomass auto-recovery system using a microcomputer (Raspberry Pi) and a sensor. The microalgal monitoring sensors (turbidity, light, pH, and temperature) designed for real-time measurements and remote monitoring were validated using standard instruments. The monitoring system was implemented in a culture of the filamentous and spiral microalgae Limnospira fusiformis. The turbidity sensor showed a strong correlation with optical density (R² = 0.943–0.986) and dry weight (R² = 0.954–0.975). The sensors for light, pH, and temperature demonstrated average percentage errors of 0.50%, 0.58%, and 2.52%, respectively, indicating their accuracy in measuring the intended parameters (p < 0.05). The developed auto-recovery system effectively maintained biomass within the desired threshold range (OD₇₅₀ = 0.74–0.67). The threshold value for the operating biomass density was adjustable with data available in real time and logged with time stamping on a Google spreadsheet. This cost-effective system, priced at approximately $330, offers a practical solution for the real-time monitoring and control of biomass density in microalgal cultures.

Real-time monitoring

Raspberry Pi

Sensor

Auto-recovery system

Limnospira fusiformis

Microalgae cultivation is attracting significant interest because of its potential across a wide range of applications spanning diverse economic sectors, including biomedicine, pharmaceuticals, cosmetics, nutraceuticals, and biofertilizers (Mutanda et al. 2020; Khavari et al. 2021; Joy and Anju 2023). The upward trend is expected to continue in the global microalgae market with a projected growth of 7.6% in 2024, increasing the market value from $1.03 billion in 2023 to $1.11 billion (Research and Markets 2024). This substantial growth highlights the critical need for the implementation of advanced, reliable, and low-cost technologies for microalgal cultivation to meet the increasing demand. Microalgal productivity is influenced by several factors such as light intensity, temperature, pH, nutrient availability, and biomass density (Esposito et al. 2017). For example, improper light conditions can affect photosynthesis and growth (Yu et al. 2022), whereas suboptimal temperatures can slow microalgae metabolism (Ganuza et al. 2008). A deviant pH disrupts nutrient uptake and photosynthesis (Yu et al. 2022). High biomass density can lead to self-shading and nutrient depletion, thereby reducing microalgal growth (Ganuza et al. 2008). Turbidostat system, traditionally used to control biomass density, is often associated with high operational costs, which make it less feasible for large-scale applications (Thoré et al. 2021). Therefore, cost-effective monitoring and control of the cultivation conditions are crucial for optimal growth and productivity.

The algal biomass density is a particularly difficult parameter for accurate monitoring. Biomass density monitoring methods fall primarily into two categories: offline and online (Uyar 2013). Offline methods, such as cell dry weight determination and microscopic cell counting, are known for their accuracy but are time-consuming (Thoré et al. 2021; Schagerl et al. 2022). However, online monitoring methods are more efficient (Uyar 2013), including the measurement of optical density (OD), in situ microscopy, flow cytometry, fluorometry, and infrared spectroscopy (Bluma et al. 2010). Among these, OD is widely used to estimate algal biomass. However, OD monitoring is primarily limited to the laboratory scale and is expensive (Ferrando et al. 2011). The ideal method for measuring microalgal biomass concentration should be cost-effective, user-friendly, accurate, reliable, and provide real-time data. Several promising techniques are currently being investigated (Schagerl et al. 2022). Red-green-blue (RGB) image analysis, for instance, is cost-effective and easy to apply, but its accuracy can be affected by camera quality and lighting conditions (Sarrafzadeh et al. 2015; Salgueiro et al. 2022). Machine-learning algorithms can be used to monitor biomass density and analyze large-scale data; however, they require extensive calibration and validation (Ning et al. 2022).

Recent studies have used analog turbidity sensors as a proxy for microalgal density to measure light scattering, which can vary based on particle characteristics such as size, shape, and reflectance (Berberoglu and Pilon 2007; Joubert et al. 2010; Ferrando et al. 2011; Kandilian et al. 2016; Defa et al. 2019). Therefore, an empirical correlation between turbidity values and actual biomass is needed for the accurate monitoring of specific microalgal species. In a previous study, a similar system was effectively used to monitor the growth of Synechocystis sp., which is known for its small size, spherical shape, and single-cell state (small colonies) (Anderson and Mcintosh 1991; Nguyen et al. 2022). However, the potential of this system for use in filamentous and spiral algae, such as Limnospira fusiformis, which potentially influences the light-scattering signal of the sensor, has not yet been thoroughly evaluated.

This study aims to develop a novel and low-cost real-time monitoring system that incorporates light, temperature, and pH sensors along with an auto-recovery system. The auto-recovery system was designed to maintain a specific range of microalgal biomass in algal cultures without manual intervention. This system uses a Raspberry Pi microcontroller unit (MCU) to orchestrate the active feedback loop. This MCU was equipped with a turbidity sensor that continuously monitored and controlled the density of the algal biomass inside the reactor. The system was programmed with a predefined threshold range to establish the desired levels of microalgal biomass. The integration of low-cost sensors for light, temperature, and pH with an auto-recovery systems has not been comprehensively explored in previous studies (Nguyen and Rittmann 2018; Hermadi et al. 2021; Nguyen et al. 2022). To the best of our knowledge, the light-sensor system employed in this study represents a novel approach that has yet to be utilized in scientific research.

Microalgae strain and pre-culture

The blue-green alga Limnospira fusiformis LA-08 was isolated from Lake Arenguade, an Ethiopian Soda Lake (Woldie et al. 2024). Limnospira, a recently established genus, was reclassified from Arthrospira (Nowicka-Krawczyk et al. 2019). A modified Spirulina-Ogawa-Terui (SOT) medium (Kishi and Toda 2018) with the following compositions (g L^− 1) was used: H₃BO₃, 2.86; MgSO₄.7H₂O, 0.20; CaCl₂, 0.03; FeSO₄.7H₂O, 0.01; Na₂EDTA.2H₂O, 0.08; NaNO₃, 2.50; Na₂SO₄, 0.82, NaCl, 1,000; Na₂CO₃, 12.19; NaHCO₃, 9.66; K₂HPO₄, 0.50; MnSO₄∙5H₂O, 0.022; ZnSO₄∙7H₂O, 0.022; CuSO₄∙5H₂O, 0.079; Na₂MoO₄∙2H₂O, 0.021. The stock culture was maintained in 500-mL Erlenmeyer flasks at a temperature of 25°C and a light intensity of 160 µmol m^− 2 s^− 1 with a 12h light:12h dark cycle. The cultures were continuously aerated at a flow rate of 0.2 L min^− 1.

For the pre-culture, cells from the stock culture were transferred into a 0.7 L glass bottle reactor. The light intensity was increased to 250 µmol m^− 2 s^− 1, while all other parameters remained consistent with the stock culture.

Experimental design of a microalgae monitoring system under indoor conditions

Cells from the pre-culture were transferred into a medium-sized glass bottle with a working volume of 0.9 L under indoor conditions, aerated with a flow rate of 0.2 L min^− 1 at a temperature of 30°C and a light intensity of 300 µmol m^− 2 s^− 1 with a 12h light:12h dark cycle. The experiment started with an initial inoculum of 0.05 OD at 750 nm and lasted for a month. A semi-continuous feeding mode was employed, with a hydraulic retention time (HRT) of 5 days (20% daily dilution rate).

A microalga monitoring system with the aforementioned culture conditions was established using sensors for turbidity, light, pH, and temperature. These sensors were equipped in the same three reactors for triplicate cultures (System 1, System 2, and System 3). A turbidity sensor (Seno189, Dfrobot, China) was used to measure microalgal biomass in a flow-through system operated by a Raspberry Pi (4b, Sonny, United Kingdom). The culture medium was continuously circulated between the reactor and the measuring section of the turbidity sensor using a 12 V DC peristaltic pump (AE1207, Gikfun, China) with a 2 mm inner diameter and a flow rate of 0.55 mL s^− 1. The analog output signal of the sensor was converted to a digital signal using an analog-to-digital converter (ADC) (MCP3008, Microchip Technology, USA). The ADC was powered with a 3.3 V reference voltage, while the turbidity sensor was operated with a 5 V power supply. A color sensor (AE-S11059, Hamamatsu, Japan) was used to measure light intensity. The sensor was connected to a Raspberry Pi using the system management bus (SMBus) library. The red, green, and blue (RGB) light components were recorded and converted into photosynthetic photon flux density (PPFD). HomeLab pH software interfaces with a pH sensor (Meter Kit V2, Dfrobot, China), providing measurements of the culture’s pH. Similarly, a digital temperature sensor (DS18B20, Dallas Semiconductor, USA) that utilizes a 1-wire communication protocol was used to monitor the temperature conditions of the culture.

The sensor data were logged into a Google sheet worksheet in the cloud. This strategic setup ensured that the data could be accessed and scrutinized from any location if there was an Internet connection. Each sensor monitored the culture every 10 min.

Development of biomass auto-recovery system under outdoor conditions

A biomass auto-recovery system was developed for outdoor conditions using a flat-plate reactor with a working volume of 4.5 L (Fig. 1). This experiment started with an initial inoculum of 0.2 OD and lasted for a month. A 12V DC peristaltic pump was used in order to circulate the medium between the sensor and reactor. This method prevented water interference from the sensor head and bubble formation in the tube. The sensor was oriented vertically to prevent cells from settling directly on the sensor wall. The sensor was wrapped in black tape to reduce the impact of the ambient light. The turbidity sensor was cleaned every three days to minimize biofouling effects. A dual-channel peristaltic pump head (AC-2110, Atto Corporation, Korea) with an internal diameter of 2 mm was used to simultaneously feed fresh medium into the reactor and to collect the culture from the reactor at a uniform flow rate (19.35 mL min^− 1).

A turbidity sensor was used to measure transparency every 10 min. When the transparency reaches a maximum threshold (2.12V: equal to 0.74 OD₇₅₀ and 0.67 gL^− 1, dry weight (DW)), the software in a Raspberry Pi activates a relay. The activated relay switched on the power supply for the dual-channel peristaltic pump for 2 min. The pump then starts moving to add fresh medium and harvest biomass. After 2 min, the Raspberry Pi software deactivates the relay, cuts off the power supply, and stops pumping. This cycle was repeated every 10 min. depending on the biomass in the culture, thereby reaching the maximum threshold. The system was turned on from 8:00 to 18:00 and turned off during the remaining hours. The maximum threshold was determined by considering the stationary phase of Limnospira fusiformis in semi-continuous feeding mode. The minimum threshold (equal to 0.67 OD₇₅₀ and 0.61 gL^− 1, DW) was set at 90% of the maximum threshold and mainly controlled by the flow rate and flow duration of the dual chamber pump. In addition, light, pH, and temperature sensors were installed along with the biomass auto-recovery system.

Analytical parameters

The OD at 750 nm and dry weight were measured to validate the turbidity sensor. The OD was measured using a UV-visible spectrophotometer (UV-2450, Shimadzu, Japan). The dry weight was determined by filtering the cell culture using combusted pre-weighed Whatman GF/F filter paper (pore size of 0.7 µm). The cells were then washed with distilled water and dried in an oven at 80°C for 8 h. After drying, the weight was measured using a weighing scale (UMX2; Mettler Toledo, USA).

A light meter (QSL-2100, Biospherical Instruments, USA), pH meter (D-51, Horiba, Japan), and Bluetooth temperature data logger (TR42, TNDD, Japan) were used to validate the light, pH, and temperature sensors, respectively. The RGB value of the light sensor was converted into lux using Eq. (1) and photosynthetic photon flux density (PPFD).

$$\:Lux=\frac{Red}{0.846\times\:0.8}+\frac{Green}{1.606\times\:0.7}+\frac{Blue}{2.154\times\:0.9}$$

Statistical analysis

Linear regression analysis and Pearson’s correlation coefficient were employed to examine the strength of the linear relationship between the sensor data and measurements from the reference device. These statistical methods were also used to investigate the relationship between light intensity and temperature in cultures harvested daily. The mean absolute error (MAE) was calculated to determine the absolute distance between the sensor measurements and the actual values. A bias score was calculated to indicate whether the sensor reading consistently overestimated or underestimated the true values. The mean absolute percentage error (MAPE) was calculated to determine the relative distance between the sensor measurements and the actual values.

$\:MAE=\frac{1}{n}\sum\:|{y}_{i}-{\stackrel{\prime }{y}}_{i}|$ (Robeson and Willmott 2023) (2)

$\:Bias=\frac{1}{n}\sum\:\:({y}_{i}-{\widehat{y}}_{i}$ ) (Malings et al. 2020) (3)

$\:MAPE=\frac{1}{n}\sum\:\left(\left|\frac{{y}_{i}-{\stackrel{\prime }{y}}_{i}}{{y}_{i}}\right|\right)\times\:100$ (Robeson and Willmott 2023) (4)

where $\:{y}_{i}$ is the actual value of the i^th observation; $\:{\stackrel{\prime }{y}}_{i}$ is the sensor value of the i^th observation; and $\:n$ is the total number of observations.

Evaluating the performance of the sensors under indoor conditions

In this study, we evaluated the performance of turbidity, light, pH, and temperature sensors in three systems (Systems 1, 2, and 3) during the indoor cultivation of L. fusiformis, as shown in Figs. 2, 3, and 4.

A turbidity sensor system was developed to continuously monitor the biomass concentration of L. fusiformis. The relationship between the optical density (OD₇₅₀) and dry weight (DW) with the turbidity sensor measurements (System 1, System 2, and System 3) is shown in Fig. 2a, b. These sensors demonstrated a significant correlation (p < 0.05) with the optical density, with R² of 0.971, 0.943, and 0.986 for Systems 1, 2, and 3, respectively (Fig. 2a). A similar trend was observed for dry weight, with R² of 0.974, 0.954, and 0.968 for systems 1, 2, and 3, respectively (Fig. 2b).

In this study, the light sensor monitored the light intensity of L. fusiformis culture and showed no significant difference from that of the control (p < 0.05). The results displayed that System 1 (300.71 ± 8.8 µmol m^− 2 s^− 1), System 2 (301.85 ± 4.79 µmol m^− 2 s^− 1), and System 3 (304.1 ± 3.71 µmol m^− 2 s^− 1). While the actual light intensity was kept constant (300 µmol m^− 2 s^− 1) (Fig. 3). This indicates that the sensors slightly overestimated the light intensity, with System 1 showing a deviation of 3.44 µmol m^− 2 s^− 1, while System 2 had 1.67 µmol m^− 2 s^− 1, and System 3 deviated by 3.98 µmol m^− 2 s^− 1. The mean absolute error (MAE) for Systems 1, 2, and 3 were 3.79, 3.77, and 4.46 µmol m^− 2 s^− 1, respectively. This implies that, on average, the measurements made by the light sensors in each system were off by these amounts. Furthermore, the mean absolute percentage errors (MAPEs) were 1.88%, 1.33%, and 1.56% for System 1, 2, and 3, respectively (Fig. 3). The measurements obtained by the light sensors in each system were removed using these percentages. The empirical data obtained from the experiment substantiate the assertion that the light sensor is capable of quantifying light intensity with a significant degree of accuracy (p < 0.05).

The reactor pH was monitored using a pH sensor. The results revealed a minor bias of 0.01 in System 1, and a slightly larger bias of 0.05 in System 2, while System 3 demonstrated no bias. The MAE values for Systems 1, 2, and 3 were 0.03, 0.08, and 0.07, respectively. Additionally, the MAPE was calculated to be 0.27%, 0.79%, and 0.69% for Systems 1, 2, and 3, respectively. These results suggest that the pH sensors were highly accurate in measuring pH levels, exhibiting a significant correlation with the actual pH (p < 0.05) (Fig. 4a).

The temperature sensor revealed that System 1 (30.91 ± 0.87°C) was close to the actual temperature (30.84 ± 0.81°C), indicating a bias of 0.07°C and MAE of 0.66°C. Similarly, System 2 (30.15 ± 0.84°C) aligned with the actual temperature (30.48 ± 0.57°C), showing a bias of -0.34°C and MAE of 0.6°C. Furthermore, system 3 showed a reading of (31.48 ± 1.05°C) which agreed with the actual temperature (30.96 ± 0.88°C), indicating a bias of 0.05°C and an MAE of 1.05°C. Moreover, the MAPE of the sensors were 2.16, 1.99, and 3.40% for Systems 1, 2, and 3, respectively. These findings indicate that all of the three temperature sensors consistently provided readings that were in line with the actual temperature (p < 0.05) (Fig. 4b).

Evaluating the performance of biomass auto-recovery system under outdoor conditions

In this study, a biomass auto-recovery system was designed to enable continuous harvesting and maintain the biomass within a specified threshold range. After day 4, when the biomass reached the maximum threshold value, the auto-recovery system was activated, the biomass was harvested, and fresh medium was added. The system maintained the biomass concentration within the desired range (Fig. 5a, b). More frequent activation of the feeding pump indicates a faster growth rate, requiring the replacement of the harvested medium to maintain sufficient nutrients for the growing culture. On day 12, the volume of the harvested culture decreased, marking a turning point for sampling of the harvested biomass (dry weight and OD) in the harvested culture section. It was observed that most of the samples had OD (0.69–0.85) and dry weight (0.66 g L^− 1– 0.78 g L^− 1) above the maximum threshold (Fig. 5a, b).

In this study, the daily harvested medium ranged from 0.3 L d^− 1 to 2 L d^− 1 (Fig. 5c). This experiment was conducted under outdoor conditions, making the daily harvested volume susceptible to variations in the temperature and light intensity. Light intensity and temperature were high at the beginning of the experiment. This allowed the cells to grow quickly, and about 2 L d^− 1 from the 4.5 L total volume was harvested. However, as the temperature and light decreased, the volume of harvested medium also decreased, reaching up to 0.3 L d^− 1.

Evaluating the performance of the sensors

This study has elucidated the considerable potential of online monitoring sensors as a cost-effective and automated approach for comprehensive monitoring of microalgae cultivation. A significant negative linear correlation (R² = 0.94–0.98, p < 0.05) was observed between transparency with both the OD and dry weight (Fig. 2a, b). The R² value of this study was relatively higher than Thoré et al. (2021) (R² = 0.87–0.93) and comparable to Nguyen and Rittmann (2018) (R² = 0.95) (Table 1). According to Thoré et al. (2021), the exact empirical relationship between transparency and biomass concentration differs among different microalgal species (Table 1). This variation can be attributed to the unique light-scattering properties of each species, which are determined by the size, shape, and other structural features of the algal cells (Kandilian et al. 2016). The three turbidity sensors employed in this study exhibited slight variations in slope and intercept. This is likely because of the dynamic nature of the relationship between turbidity and algal cell biomass within the same culture, which can fluctuate over time (Ferrando et al. 2011). Furthermore, placing the three sensors in separate reactors led to minor variations in the slope and intercept. Therefore, individual calibration of each sensor was deemed essential for obtaining precise measurements.

This study effectively demonstrated the use of a light sensor to monitor the light intensity of L. fusiformis cultures. The empirical data obtained from the experiment provide compelling evidence that the light sensor can quantify the light intensity with a significant degree of accuracy (p < 0.05) (Fig. 3). This suggests that the low-cost sensor is a viable alternative to commercial light meters for future applications. The light sensor used in this study has a wide range of potential applications, such as enabling LCD backlight adjustment in cell phones and PCs, saving energy in large TVs, and various light detection applications. The use of this sensor for monitoring microalgae marks a revolutionary advancement towards cost-effective and accurate light intensity measurements in the field of biological research.

The pH sensor used in this study was highly accurate and exhibited a significant correlation with the actual pH (p < 0.05) (Fig. 4a). In a study conducted by Hermadi et al. (2021) and Sugiharto et al. (2023), a similar sensor was used for algal culture and water quality assessments, and it resulted in a percentage errors of 1.37% and 3.15%, respectively. In contrast, our study demonstrated a more accurate result with a lower MAPE of 0.58%. However, Nguyen et al. (2022) reported a lower average error (0.24%) for a pH sensor that surpassed the accuracy achieved in our study (Table 2). Therefore, owing to their superior accuracy and reliability, pH sensors have emerged as preferred tools for real-time pH monitoring in algal ponds.

This finding underscores that all of the three temperature sensors consistently provided readings that were consistent with the actual temperature (data logger) (p < 0.05) (Fig. 4b). However, it appeared that System 2 might have experienced sensor drift in the middle of the experiment. This could potentially be due to a decrease in power supply from 3.3V to 2.8V from the Raspberry Pi, which had a physical connection problem. This led to the relatively large negative bias (-0.34°C) observed (Fig. 4b). Therefore, ensuring that the sensor receives adequate power from Raspberry Pi is essential for data accuracy. In a study conducted by Sugiharto et al. (2023), the same type of temperature sensor utilized for water quality monitoring reported a lower MAPE (1.46%), which was notably lower than 2.52% observed in this study. This discrepancy could be attributed to the differences in the duration of exposure. When this sensor was used to measure air temperature, the average bias was 0.05°C (Elyounsi and Kalashnikov 2021). However, in this study, the application of this sensor showed an average bias of 0.15°C. This variation could potentially be attributed to the biofouling effect, which is known to be more prevalent in water than in air.

The monitoring sensors used in this study demonstrated remarkable efficiency and provided real-time data. The utility of this system in algal cultivation has been validated, indicating its potential for broader applications. Their performance suggests their strong capacity to improve the accuracy and sustainability of large-scale algal farming practices.

Table 1

Relationship between turbidity sensor and optical density in different microalga species. The three equations under this study showed System 1, 2, and 3, respectively
Microalgae species	Wavelength (λ)	Correlation of turbidity with optical density (R2)	Empirical relation of turbidity with optical density	References
Synechocystis	730	0.95	y = -0.415x + 3.000	Nguyen and Rittmann 2018
Microchloropsis gaditana	435	0.91	y = 0.017x0.668	Thoré et al. 2021
Chloromonas typhlos	435	0.87	y = 0.024x0.696	Thoré et al. 2021
Porphyridium purpureum	435	0.90	y = 0.023x0.657	Thoré et al. 2021
Limnospira fusiformis	750	0.94–0.98	y = -2.920x + 3.383 y = -3.109x + 3.541 y = -3.007x + 3.416	This study

Biomass auto-recovery system

In this study, most samples in the harvested culture section had OD and dry weights above the maximum threshold (Fig. 5a, b). This was because the auto-recovery system was turned off during the night, allowing the cells to grow. When the system was switched back in the morning, the cells were harvested at a high biomass concentration. The daily harvested cultures exhibited variations that were attributed to fluctuations in light intensity and temperature. The daily harvested culture was significantly correlated with the daily average light intensity (R² = 0.88, p < 0.05) (Fig. 6a) and temperature (R² = 0.60, p < 0.05) (Fig. 6b). These findings underscore the substantial effects of light intensity and temperature on biomass production and emphasize the importance of considering these factors when optimizing biomass production.

The development of a simple and low-cost turbidity sensor system controlled by a Raspberry Pi microcomputer is of significant importance for the remote monitoring of algal density and biomass recovery. The utilization of a turbidity sensor operating in the flow-through mode provides an efficient method for the real-time monitoring and control of algal biomass. Real-time data acquisition is crucial for maintaining optimal biomass density. The inherent capability of this sensor to record data autonomously offers a significant advantage over traditional monitoring methods, which are often labor-intensive, time consuming, and costly (Joubert et al. 2010). The implementation of continuous monitoring systems is particularly beneficial for the automation of culturing and harvesting techniques, thereby increasing overall efficiency. The system also enables immediate implementation of corrective measures in response to suboptimal growth rates. This feature ensures that the biomass concentration is maintained within the desired range, which is critical for achieving optimal growth rates, and consequently, high productivity. Furthermore, a biomass auto-recovery system was designed to enable continuous harvesting and maintain biomass within a specified threshold range. Continuous harvesting, coupled with the timely addition of nutrients, ensures that sufficient nutrients are available for culturing. Therefore, this system represents a significant advancement in microalgae cultivation. Its implementation can lead to substantial improvements in the sustainability of biomass production processes.

Table 2

Comparison of percentage errors of light, pH, and temperature sensors with the previous study
Sensor	Percentage error (%)	References
Light	0.50	This study
pH	1.37	Hermadi et al. 2021
	0.25	Nguyen et al. 2022
	3.15	Sugiharto et al. 2023
	0.58	This study
Temperature	1.46	Sugiharto et al. 2023
Temperature	2.52	This study

Social implementation and feasibility of the system

The development of a comprehensive real-time monitoring and biomass auto-recovery system for microalgae cultivation requires detailed financial planning and cost analyses to ensure feasibility and sustainability. The primary component of this system was a monitoring apparatus. The procurement of the necessary items for the present monitoring and auto-recovery systems costs approximately $330, excluding the dual-chamber pump, which highlights the cost-effectiveness of the system setup. This cost analysis was based on the costs in November 2023. It’s important to note that this cost analysis was conducted considering the expenses associated with setting up the system for a single reactor or pond. Notably, the financial implications of scaling up the system for large-scale applications would remain largely unchanged, with the primary additional expense being the high-flow-rate pump required for biomass harvesting. This analysis underscores the viability of employing the system at both small- and large-scales, offering a scalable solution that can enhance the productivity and sustainability of microalgal cultivation with minimal variation in cost.

Competing interests

The authors declare no competing interests.

Funding

This work was supported by the Science Research Promotion Fund from Promotion and Mutual Aid Corporation for Private Schools of Japan; and the Japan Science and Technology Agency (JST)/Japan International Cooperation Agency (JICA), Science and Technology Research Partnership for Sustainable Development (SATREPS), through the project Eco Engineering for Agricultural Revitalization toward Improvement of Human Nutrition (EARTH): [grant number JPMJSA2005].

Author Contribution

Workie Desalegn: conceptualization, data acquisition, formal analysis, investigation, methodology, visualization, writing-original draft, review and editing.Anupreet Kaur Chowdhary and Mutsumi Sekine: conceptualization, formal analysis, review and editing. Washburn Larry and Masatoshi Kishi: conceptualization and review. Woldie Ayirkm: review and editing. Tatsuki Toda: conceptualization, formal analysis, review and editing, supervision, funding acquisition.

Acknowledgement

The authors gratefully acknowledge Prof. Ken Furuya and Dr. Masahiro Ohtake for their valuable feedback that greatly improved our manuscript.

Data availability

All data generated or analyzed during this study are included in this published article.

Anderson SL, Mcintosh L (1991) Light-activated heterotrophic growth of the cyanobacterium Synechocystis sp. strain PCC 6803: a blue-light-requiring process. J Bacteriol 173(9):2761-2767.
Berberoglu H, Pilon L (2007) Experimental measurements of the radiation characteristics of Anabaena variabilis ATCC 29413-U and Rhodobacter sphaeroides ATCC 49419. Int J Hydrog Energy 32:4772–4785. https://doi.org/10.1016/j.ijhydene.2007.08.018
Bluma A, Höpfner T, Lindner P, Rehbock C, Beutel S, Riechers D, Hitzmann B, Scheper T (2010) In-situ imaging sensors for bioprocess monitoring: state of the art. Anal Bioanal Chem 398:2429–2438.
Defa RP, Ramdhani M, Priramadhi RA, Aprillia BS (2019) Automatic controlling system and IoT based monitoring for pH rate on the aquaponics system. J Phys: Conference Series 1367 (2019) 012072.
Elyounsi A, Kalashnikov AN (2021) Evaluating suitability of a DS18B20 temperature sensor for use in an accurate air temperature distribution measurement network. Engi Proceedings 10:1-7. https://doi.org/10.3390/ecsa-8-11277
Esposito S, Cafiero A, Giannino F, Mazzoleni S, Diano M (2017) A Monitoring, modeling and decision support system (DSS) for a microalgae production plant based on internet of things structure. Procedia Comput Sci 113:519–524. https://doi.org/10.1016/J.PROCS.2017.08.316
Ferrando NS, Benítez HH, Gabellone NA, Claps MC, Altamirano PR (2011) A quick and effective estimation of algal density by turbidimetry developed with Chlorella vulgaris cultures. Limnetica 29:397–406.
Ganuza E, Anderson AJ, Ratledge C (2008) High-cell-density cultivation of Schizochytrium sp. in an ammonium/pH-auxostat fed-batch system. Biotechnol Lett 30:1559–1564. https://doi.org/10.1007/s10529-008-9723-4
Hermadi I, Setiadianto IR, Al Zahran DFI, Simbolon MN, Saefurahman G, Wibawa DS, Arkeman Y (2021) Development of smart algae pond system for microalgae biomass production. In: IOP Conference Series: Earth and Envi Sci 8:1-9.
Joubert A, Calmes B, Berruyer R, Pihet M, Bouchara JP, Simoneau P, Guillemette T (2010) Laser nephelometry applied in an automated microplate system to study filamentous fungus growth. Biotechni 48:399–404. https://doi.org/10.2144/000113399
Joy SR, Anju T.R (2023) Microalgal biomass: introduction and production methods. Springer, Singapore.
Kandilian R, Pruvost J, Artu A, Lemasson C, Legrand J, Pilon L (2016) Comparison of experimentally and theoretically determined radiation characteristics of photosynthetic microorganisms. J Quant Spectrosc. Radiat Transfer 175:30–45. https://doi.org/10.1016/j.jqsrt.2016.01.031ï
Khavari F, Saidijam M, Taheri M, Nouri F (2021) Microalgae: therapeutic potentials and applications. Mol Biol Rep 48:4757–4765.
Kishi M, Toda T (2018) Carbon fixation properties of three alkalihalophilic microalgal strains under high alkalinity. J Appl Phycol 30:401–410. https://doi.org/10.1007/s10811-017-1226-z
Malings C, Tanzer R, Hauryliuk A, Saha PK, Robinson AL, Presto AA, Subramanian R (2020). Fine particle mass monitoring with low-cost sensors: corrections and long-term performance evaluation. Aerosol Sci and Technol 54: 160–174.
Mutanda T, Naidoo D, Bwapwa JK, Anandraj A (2020) Biotechnological applications of microalgal oleaginous compounds: current trends on microalgal bioprocessing of products. Front Energy Res 8:1-21.
Nguyen BT, Rittmann BE (2018) Low-cost optical sensor to automatically monitor and control biomass concentration in microalgal cultivation. Algal Res 32:101–106. https://doi.org/10.1016/j.algal.2018.03.013
Nguyen DK, Nguyen HQ, Dang HT, Nguyen VQ, Nguyen (2022) A low-cost system for monitoring pH, dissolved oxygen and algal density in continuous culture of microalgae. HardwareX 12:1-18. https://doi.org/10.1016/j.ohx.2022.e00353
Ning H, Li R, Zhou T (2022) Machine learning for microalgae detection and utilization. Front Mari Sci 9: 73-94.
Nowicka-Krawczyk P, Mühlsteinová R, Hauer T (2019) Detailed characterization of the Arthrospira type species separating commercially grown taxa into the new genus Limnospira (Cyanobacteria). Sci Rep 9:1-11. https://doi.org/10.1038/s41598-018-36831-0
Research and Markets (2024) Microalgae Global Market Report. https://www.researchandmarkets.com/reports/5790809/microalgae-global-market-report
Robeson SM, Willmott CJ (2023) Decomposition of the mean absolute error (MAE) into systematic and unsystematic components. PloS one, 18: e0279774. https://doi.org/ 10.1371/journal.pone.027977
Salgueiro JL, Pérez L, Sanchez Á, Cancela Á, Míguez C (2022) Microalgal biomass quantification from the non-invasive technique of image processing through red–green–blue (RGB) analysis. J Appl Phycol 34:871–881. https://doi.org/10.1007/s10811-021-02634-6
Sarrafzadeh MH, La HJ, Lee JY, Cho DH, Shin SY, Kim WJ, Oh HM (2015) Microalgae biomass quantification by digital image processing and RGB color analysis. J Appl Phycol 27:205-209. https://doi.org/10.1007/s10811-014-0285-7/Published
Schagerl M, Siedler R, Konopáčová E, Ali SS (2022) Estimating biomass and vitality of microalgae for monitoring cultures: a roadmap for reliable measurements. Cells 11:24-55. https://doi.org/10.3390/cells11152455
Sugiharto WH, Susanto H, Prasetijo AB (2023) Real-time water quality assessment via IoT: monitoring pH, TDS, temperature, and turbidity. Ing Syst Inf 28: 823–831.
Thoré ES, Schoeters F, Spit J, Van Miert S (2021) Real-time monitoring of microalgal biomass in pilot-scale photobioreactors using nephelometry. Processes 9:1-9. https://doi.org/10.3390/pr9091530
Uyar B (2013) A novel non-invasive digital imaging method for continuous biomass monitoring and cell distribution mapping in photobioreactors. J Chem Technol Biotechnol 88:1144–1149. https://doi.org/10.1002/jctb.3954
Woldie A, Chowdhary AK, Sekine M, Kish M, Kurosawa N, Zegeye M, Toda T (2024) Growth characteristics and molecular identification of indigenous Limnospira strains from Ethiopian soda lakes as a protein source. J Biocatal Agric Biotechnol, Manuscript Number: BAB-D-24-00109 (Under review)
Yu H, Kim J, Rhee C, Shin J, Shin SG, Lee C (2022) Effects of different pH control strategies on microalgae cultivation and nutrient removal from anaerobic digestion effluent. Microorganisms 10:1-15. https://doi.org/10.3390/microorganisms10020357

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Development of a low-cost real-time monitoring system for biomass concentration and environmental factors in microalgae Limnospira fusiformis cultivation

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Microalgae strain and pre-culture

Experimental design of a microalgae monitoring system under indoor conditions

Development of biomass auto-recovery system under outdoor conditions

Analytical parameters

Statistical analysis

Results

Evaluating the performance of the sensors under indoor conditions

Evaluating the performance of biomass auto-recovery system under outdoor conditions

Discussion

Evaluating the performance of the sensors

Biomass auto-recovery system

Social implementation and feasibility of the system

Declarations

Funding

Author Contribution

Acknowledgement

Data availability

References

Additional Declarations

Status:

Version 1