A Novel System for Real-Time, in Situ Monitoring of CO2 Sequestration in Photoautotrophic Bioﬁlms

Home , Biosequestration, Primary production

Article A Novel System for Real-Time, In Situ Monitoring of CO2 Sequestration in Photoautotrophic Bioﬁlms

Patrick Ronan 1 , Otini Kroukamp 1, Steven N. Liss 1,2 and Gideon Wolfaardt 1,2,* 1 Department of Chemistry and Biology, Ryerson University, 350 Victoria St., Toronto, ON M5B 2K3, Canada; [email protected] (P.R.); [email protected] (O.K.); [email protected] (S.N.L.) 2 Department of Microbiology, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa * Correspondence: [email protected]

Received: 16 June 2020; Accepted: 24 July 2020; Published: 31 July 2020

Abstract: Climate change brought about by anthropogenic CO2 emissions has created a critical need for effective CO2 management solutions. Microalgae are well suited to contribute to efforts aimed at addressing this challenge, given their ability to rapidly sequester CO2 coupled with the commercial value of their biomass. Recently, microalgal biofilms have garnered significant attention over the more conventional suspended algal growth systems, since they allow for easier and cheaper biomass harvesting, among other key benefits. However, the path to cost-effectiveness and scaling up is hindered by a need for new tools and methodologies which can help evaluate, and in turn optimize, algal biofilm growth. Presented here is a novel system which facilitates the real-time in situ monitoring of algal biofilm CO2 sequestration. Utilizing a CO2-permeable membrane and a tube-within-a-tube design, the CO2 sequestration monitoring system (CSMS) was able to reliably detect slight changes in algal biofilm CO2 uptake brought about by light–dark cycling, light intensity shifts, and varying amounts of phototrophic biomass. This work presents an approach to advance our understanding of carbon flux in algal biofilms, and a base for potentially useful innovations to optimize, and eventually realize, algae biofilm-based CO2 sequestration.

Keywords: bioﬁlms; CO2 sequestration; microalgae; photosynthesis; real-time monitoring

1. Introduction Over the past two centuries, human activity has had a pronounced effect on the natural environment [1]. Anthropogenic CO2 emissions have led to a range of detrimental environmental impacts, including the acidification of ocean waters, rising sea levels, and increasing global temperatures [2,3]. Accordingly, climate change mitigation and CO2 management have become a top priority globally, which is increasingly reflected in our political, social, and economic realms. Engineered bio-sequestration systems which exploit the natural CO2-capturing ability of photosynthetic plants and algae potentially represent a viable and valuable compliment to current CO2 management strategies [4]. Microalgae are particularly well suited for this, given their superior growth and CO2-fixation rates compared to terrestrial plants [5], as well as the commercial value of their biomass. For example, microalgal biomass can be used as animal feed and fertilizer with minimal downstream processing required [6]. It is also a source of numerous valuable compounds such as pigments, cosmetic products, and nutraceuticals like omega-3 fatty acids [5,7,8], and the lipid-rich nature of algal cells makes them an attractive biodiesel feedstock [9–11]. The commercial value of microalgal biomass enables a direct link between CO2 sequestration and the generation of value-added products which can be used to subsidize the overall process [7]. Despite the benefits of such an approach, there is a need for further research focusing on developing new tools, growth systems, and methodologies to make this an economically viable, large-scale

Microorganisms 2020, 8, 1163; doi:10.3390/microorganisms8081163 www.mdpi.com/journal/microorganisms Microorganisms 2020, 8, 1163 2 of 14

option [3,12]. Notably, the ability to track and measure the culture’s CO2 uptake rate and efficiency in real time is crucial in order to allow for process monitoring, assessment, and optimization. To date, studies exploring microalgal CO2 sequestration and biomass production have focused predominantly on suspended cultures [13,14]. In nature, however, microalgae, like most microorganisms, commonly grow as aggregates in suspension, or attached to solid surfaces as biofilms. The cells within these aggregates are encased in a matrix of extracellular polymeric substances (EPSs) and exhibit markedly different physiological and metabolic characteristics compared to single planktonic cells of the same species [15]. From an algae cultivation perspective, biofilm growth systems offer several key advantages over the more conventional suspended growth systems. For example, the time, energy, and financial costs associated with biomass harvesting and dewatering are lessened significantly [14]. Immobilization of the cells in biofilm growth systems creates the potential for more efficient use of light through the positioning of concentrated biomass for optimum light harvesting [16,17]. Conventionally, CO2 uptake is assessed by weighing the algal biomass produced and, using an approximate algal cell carbon content of 40–50%, stoichiometrically calculating the amount of carbon sequestered [18,19]. However, since this is a destructive measurement, it provides little insight about a biofilm’s performance during active growth. Furthermore, the fact that many algal strains are also able to metabolize organic carbon through mixotrophic growth can make such calculations problematic [20]. Another means of quantifying algal CO2 sequestration is by measuring effluent pH. As dissolved CO2 is taken up and fixed, the resulting effluent pH rise can be measured and used as an indirect indicator of changes in CO2 concentration. However, since other factors such as nitrogen assimilation can also influence effluent pH, CO2 quantification on this basis alone may be inaccurate [21]. Lastly, CO2 electrodes exist which can be used to monitor dissolved CO2 concentrations. However, such probes can be prohibitively expensive, become fouled, and are often not adequately sterilizable [21]. Therefore, a system which can accomplish non-destructive monitoring of CO2 sequestration by photoautotrophic biofilms in real time is strongly desirable. Presented here is a novel system for the real-time in situ monitoring of biofilm bio-sequestration. Building upon the CO2 evolution measurement system (CEMS) first described by Kroukamp and Wolfaardt [22], the present CO2 sequestration monitoring system (CSMS) was used to quantify the CO2 uptake rate of photoautotrophic biofilms under various conditions. The data presented demonstrates the ability of the CSMS to detect and monitor CO2 flux during (i). light–dark cycling, (ii). long- and short-duration light intensity shifts, and (iii). changing amounts of photoautotrophic biomass in the system. This work presents an approach to advance our understanding of carbon flux in algal biofilms, carbon exchange between autotrophic and heterotrophic biofilms, as well as a base for potentially useful innovations to optimize, and eventually realize, algae biofilm-based CO2 sequestration.

2. Materials and Methods

2.1. System Configuration The CSMS used in this study comprises two distinct biofilm reactor (BR) modules. The first, referred to as BRprod (CO2 producer), is used to grow a heterotrophic bacterial biofilm, which serves as a biotic source of concentrated CO2. The second, BRcons (CO2 consumer), is used to grow the photoautotrophic biofilm of interest. Each BR consists of a tube-within-a-tube design (Figure1). Biofilms were grown inside a CO2-permeable silicone tube (1.57 mm ID, 2.41 mm OD, 0.41 mm wall thickness; VWR International, Mississauga, ON, Canada) 150 cm in length, with a volume of 2.9 mL and approximately 74 cm2 of colonizable surface area. Housing this tube was a larger diameter CO2-impermeable Tygon™ tube also 150 cm in length (4.76 mm ID, 7.94 mm OD, 1.58 mm wall thickness, E-3603 formulation; VWR International, Mississauga, ON, Canada). The total volume of this tube, accounting for the volume occupied by the inner silicone tube, was 19.85 mL (Table1). Protruding near the influent and effluent end of the outer Tygon™ tube was smaller diameter Tygon™ Microorganisms 2020, 8, 1163 3 of 14 tubing (1.59 mm ID, 3.18 mm OD, 0.79 mm wall thickness, E-3603 formulation; VWR International, Mississauga, ON, Canada). Sealant was used to ensure the connections were airtight. Microorganisms 2020, 8, x FOR PEER REVIEW 3 of 15

Figure 1. SchematicFigure 1. Schematic of the of bioﬁlm the biofilm reactor reactor (BR) (BR) module.module. Liquid Liquid medium medium is pumped is pumped into the silicone into the silicone tube wheretube inoculation where inoculation and subsequentand subsequent bioﬁlm biofilm growthgrowth occur. occur. The Theannular annular space created space by created the by the Tygon™ tube is used for gas channeling (dark arrows). CO2 molecules can readily pass between the Tygon™ tube is used for gas channeling (dark arrows). CO molecules can readily pass between the dry, gaseous environment in the annular space and the aqueous2 environment inside the silicone tube

dry, gaseous(white environment arrows). The inCO the2 analyzer annular downstream space and continuously the aqueous monitors environment the CO2 concentration inside the in the silicone tube (white arrows).gas stream. The CO2 analyzer downstream continuously monitors the CO2 concentration in the gas stream. Biofilms were grown inside a CO2-permeable silicone tube (1.57 mm ID, 2.41 mm OD, 0.41 mm wall thickness; VWR International, Mississauga, ON, Canada) 150 cm in length, with a volume of 2.9 TablemL and 1. COapproximately2 sequestration 74 cm monitoring2 of colonizable system surface (CSMS) area. Housing properties this andtube operatingwas a larger conditions. diameter CO2-impermeable Tygon™ tube also 150 cm in length (4.76 mm ID, 7.94 mm OD, 1.58 mm wall Inner Silicone Tube Outer Tygon™ Tube thickness, E-3603 formulation; VWR International, Mississauga, ON, Canada). The total volume of this tube,Colonizable accounting surface for the area volume occupied74 by cm the2 inner silicone tube, was 19.85n/a mL (Table 1). Protruding nearVolume the influent and effluent end 2.90 ofmL the outer Tygon™ tube 19.85was smaller mL diameter Tygon™ tubingFlow (1.59 rate mm ID, 3.18 mm OD, 0.25 0.79 mL /mmmin wall thickness, E-3603 2.5 mL formulation;/min VWR International,Retention Mississauga, time ON, Canada). Sealant 11.6 wa mins used to ensure the connections 7.9 min were airtight.

Liquid growthTable medium 1. CO2 sequestration pumped monitoring through system the inner (CSMS) silicone properties tube and operating provided conditions. nutrients to the biofilm. Inner Silicone Tube Outer Tygon™ Tube The silicone membrane has a high CO2 permeability (2013.2 Barrer), which allowed CO2 molecules to Colonizable surface area 74 cm2 n/a readily pass between the aqueousVolume environment inside2.90 mL this tube, and 19.85 the mL dry, gaseous environment in the annular space. BRprodFlow rateand BRcons were 0.25 separately mL/min supplied with2.5 mL/min sterile growth medium to grow heterotrophic bacterial,Retention and time photoautotrophic 11.6 biofilms,min respectively. 7.9 min In contrast, the gas flow in the annular spaceLiquid was growth continuous medium pumped through through both BRs the inner (Figure silicone2), allowing tube provided the heterotrophic nutrients to the biofilm in BRprod to servebiofilm. as The a bioticsilicone source membrane of CO has2 ato high the CO photoautotrophic2 permeability (2013.2 biofilm Barrer), growing which allowed inside CO BR2 cons. moleculesMicroorganisms to 2020 readily, 8, x FOR pass PEER between REVIEW the aqueous environment inside this tube, and the dry, gaseous4 of 15 environment in the annular space. BRprod and BRcons were separately supplied with sterile growth medium to grow heterotrophic bacterial, and photoautotrophic biofilms, respectively. In contrast, the gas flow in the annular space was continuous through both BRs (Figure 2), allowing the heterotrophic biofilm in BRprod to serve as a biotic source of CO2 to the photoautotrophic biofilm growing inside BRcons.

Figure 2. FigureConfiguration 2. Configuration of of the the CSMS. CSMS. CO CO2 produced2 produced by the by heterotrophic the heterotrophic biofilm in BR biofilmprod (CO2 in BRprod producer) enters the annular space where it is carried by the sweeper gas to BRcons (CO2 consumer) via (CO2 producer) enters the annular space where it is carried by the sweeper gas to BRcons (CO2 consumer) Analyzer 1. As the photoautotrophic biofilm inside BRcons develops, its rate of CO2 uptake increases, via Analyzer 1. As the photoautotrophic biofilm inside BRcons develops, its rate of CO uptake increases, causing a subsequent decrease in the amount of gaseous CO2 reaching Analyzer 2. The difference2 in causing aCO subsequent2 concentration decrease logged inby theAnalyzer amount 1 and of 2gaseous provides a CO real-time2 reaching measure Analyzer of the biofilm’s 2. The CO di2ff erence in CO2 concentrationuptake. logged by Analyzer 1 and 2 provides a real-time measure of the biofilm’s CO2 uptake.

2.2. Gas Channeling and Monitoring

A sweeper gas of CO2-free air (TOC grade, 24001980; Linde Canada Limited, Concord, ON, Canada) was channeled into the annular space of the system at a constant flow rate of 2.5 mL/min using a mass flow controller (GFC17 mass flow controller; Aalborg Instruments & Controls, Inc., Orangeburg, NY, USA). This carried CO2 produced by the heterotrophic biofilm in BRprod to BRcons via a non-dispersive infrared CO2 analyzer (Analyzer 1) (LI-820; LI-COR Biosciences, Lincoln, NE, USA). The analyzer has a precision in the range of ~1 ppm and was used to measure and log the CO2 concentration at one-minute intervals. As the photoautotrophic biofilm inside BRcons developed, its rate of CO2 uptake increased, as reflected by the consequent decrease in the amount of gaseous CO2 reaching Analyzer 2, positioned downstream of BRcons (Figure 2). The difference in CO2 concentration logged by Analyzers 1 and 2 (BRcons CO2 influent and effluent, respectively) thus provided a real-time measure of CO2 uptake by the photoautotrophic biofilm. By default, the analyzers record CO2 concentration in parts per million (ppm). However, given a gas flow rate of 2.5 mL/min and the fact that the temperature and pressure inside the CO2 analyzers are also recorded (typically 50 °C and 100 kPa, respectively), the ideal gas law can be applied to convert these raw ppm values to rates of CO2 uptake, provided in units of µmol/h.

2.3. Light Exposure and Intensity

The photoautotrophic biofilm in BRcons was illuminated with a fluorescent plant-growth light (JSV2 Jump Start T5 Grow Light System; Hydrofarm Inc., Petaluma, CA, USA). The bulb provided simulated sunlight with a 6400 K color temperature. This lighting system affords the ability to raise and lower the light as desired. Unless otherwise stated, illumination was provided 5 cm above the biofilm, resulting in a photosynthetic photon flux density (PPFD) of approximately 141.79 µmol/m2/s.

2.4. Growth Media

Microorganisms 2020, 8, 1163 4 of 14

2.2. Gas Channeling and Monitoring

A sweeper gas of CO2-free air (TOC grade, 24001980; Linde Canada Limited, Concord, ON, Canada) was channeled into the annular space of the system at a constant flow rate of 2.5 mL/min using a mass flow controller (GFC17 mass flow controller; Aalborg Instruments & Controls, Inc., Orangeburg, NY, USA). This carried CO2 produced by the heterotrophic biofilm in BRprod to BRcons via a non-dispersive infrared CO2 analyzer (Analyzer 1) (LI-820; LI-COR Biosciences, Lincoln, NE, USA). The analyzer has a precision in the range of ~1 ppm and was used to measure and log the CO2 concentration at one-minute intervals. As the photoautotrophic biofilm inside BRcons developed, its rate of CO2 uptake increased, as reflected by the consequent decrease in the amount of gaseous CO2 reaching Analyzer 2, positioned downstream of BRcons (Figure2). The di fference in CO2 concentration logged by Analyzers 1 and 2 (BRcons CO2 influent and effluent, respectively) thus provided a real-time measure of CO2 uptake by the photoautotrophic biofilm. By default, the analyzers record CO2 concentration in parts per million (ppm). However, given a gas flow rate of 2.5 mL/min and the fact that the temperature and pressure inside the CO2 analyzers are also recorded (typically 50 ◦C and 100 kPa, respectively), the ideal gas law can be applied to convert these raw ppm values to rates of CO2 uptake, provided in units of µmol/h.

2.3. Light Exposure and Intensity

The photoautotrophic biofilm in BRcons was illuminated with a fluorescent plant-growth light (JSV2 Jump Start T5 Grow Light System; Hydrofarm Inc., Petaluma, CA, USA). The bulb provided simulated sunlight with a 6400 K color temperature. This lighting system affords the ability to raise and lower the light as desired. Unless otherwise stated, illumination was provided 5 cm above the biofilm, resulting in a photosynthetic photon flux density (PPFD) of approximately 141.79 µmol/m2/s.

2.4. Growth Media

The heterotrophic biofilm in BRprod was fed a tryptic soy broth at a flow rate of 0.25 mL/min using a Watson Marlow 205S peristaltic pump. This medium was prepared as a 0.6 g/L solution, which is a 2% concentration relative to the manufacturer’s directions for typical batch experiments. The resulting composition was 0.34 g/L casein peptone, 0.06 g/L soya peptone, 0.1 g/L sodium chloride, 0.05 g/L dipotassium hydrogen phosphate, and 0.05 g/L glucose. Media were prepared in distilled water and autoclaved at 121 ◦C for 20 min prior to use. The photoautotrophic biofilm in BRcons was fed a modified Bold’s Basal Medium (M-BBM) (also at 0.25 mL/min) with a composition of 0.22 g/L (NH ) SO , 0.025 g/L NaCl, 0.025 g/L CaCl 2H O, 4 2 4 2 · 2 0.075 g/L MgSO 7H O, 0.175 g/L KH PO , 0.075 g/LK HPO 8.34 mg/L FeSO . It should be noted 4 · 2 2 4 2 4, 4 that no carbon was provided in the liquid medium. All medium components were maintained in filter-sterilized 100 stock solutions, which were combined and diluted in distilled water and autoclave × sterilized at 121 ◦C for 20 min prior to use. The CSMS represents a plug flow reactor, with the BR 1 module having a retention time of 11.6 min and a dilution rate of 5.17 h− . This greatly exceeds the maximum specific growth rate values of microalgal and bacterial strains [23–25], which ensures that non-biofilm-bound cells are readily washed out the system.

2.5. Test Cultures and System Inoculation

BRprod and BRcons were inoculated with a heterotrophic and photoautotrophic culture, respectively. The heterotrophic culture used was Pseudomonas aeruginosa (PA01) [26], previously labelled with GFP via a mini-Tn7 transposon system [27]. For each CSMS experiment, PA01 was inoculated as an overnight culture, prepared by transferring one colony from a tryptic soy agar plate into 10 mL of 0.6 g/L tryptic soy broth and incubated overnight on a rotary shaker at 30 ◦C. BRcons was the “test module” in which the photoautotrophic bioﬁlm of interest was grown and studied. The culture used was initially enriched from an aerated wastewater lagoon in Dundalk, Microorganisms 2020, 8, 1163 5 of 14

Ontario, Canada. A 5 mL sample was inoculated into 145 mL of M-BBM and illuminated continuously with the fluorescent plant growth light. Filtered ambient air was bubbled into the culture to provide mixing and dissolved CO2. When the culture was sufficiently dense (qualitatively green and turbid), a 10 mL aliquot was transferred to 90 mL of fresh medium and placed under the same conditions. This transferring procedure was repeated biweekly, ensuring that the photoautotrophic culture inoculated into BRcons was not more than 14 days old. Sterile syringes were used to inoculate reactors by injecting 3 mL of the appropriate culture into the respective BRs through the silicone tubing just upstream of the reactor. A 22-gauge hypodermic syringe needle was used to pierce the tubing, and a dollop of silicone glue was used to seal this puncture as the needle was removed. Liquid medium flow was paused during inoculation and for a period of one hour following this to allow for some cell adherence. The BRs were inoculated concurrently, with the exception of the first experiment. In that case, BRcons was inoculated with the photoautotrophic culture approximately 24 h after the inoculation of PA01 into BRprod, once the CO2 output of the heterotrophic biofilm had plateaued. This allowed for the assessment of CO2 uptake during early stage photoautotrophic biofilm development, which was not possible when the BRs were inoculated concurrently. In subsequent experiments, the CO2 uptake data is presented for mature photoautotrophic biofilms of at least 20 h old.

2.6. Assessing CO2 Uptake during Photoautotrophic Bioﬁlm Development

A photoautotrophic culture was inoculated into BRcons approximately 24 h after BRprod was inoculated with PA01. This ensured that a relatively steady concentration of CO2 was achieved at the time of BRcons inoculation and allowed for the visualization of CO2 uptake during photoautotrophic biofilm development. Taking the difference in CO2 concentration measured by Analyzer 1 and 2 at each time point gave a raw concentration in parts per million, which, using the ideal gas law, was converted to a rate of CO2 uptake and plotted against time. To confirm the presence and photosynthetic activity of the photoautotrophic biofilm, a procedure similar to the light–dark shift method described by Revsbech et al. [28] was performed, in which a steady state biofilm in BRcons was covered with an opaque sheet for approximately four hours and the resulting shift in CO2 flux was recorded. A series of abiotic control experiments were also conducted to confirm that any difference in BRcons influent and effluent CO2 concentration in CSMS experiments was in fact due to the photoautotrophic activity of the biofilm, and not leakage or passive diffusion out of the system. This was achieved by channeling concentrated CO2 (1800 ppm, 24081376, Linde Canada Limited, Concord, ON, Canada) through a sterile, un-inoculated CSMS at two different flow rates (2.5 mL/min and 20 mL/min) for a period of 0.25 h.

2.7. Light Manipulations

2.7.1. CO2 Uptake during Light–Dark Cycling A series of tests were performed to demonstrate the ability of the CSMS to detect real-time changes in CO2 uptake by the photoautotrophic biofilm, brought about by manipulations of light illumination. First, the system was operated on a diurnal cycle (12 h light, 12 h dark), with the photoautotrophic biofilm’s CO2 uptake rate assessed and measured as µmol/h. The system was covered in such a way as to ensure that no light could reach the biofilm during the dark periods, and that only light from the fluorescent growth light could illuminate the biofilm during light periods. The light/dark switching was accomplished using a programmable plug-in timer. This experiment lasted for 120 h, representing a total of five light–dark cycles. Microorganisms 2020, 8, 1163 6 of 14

2.7.2. CO2 Uptake during Changes in Light Intensity

To investigate the CSMS’s ability to detect changes in CO2 uptake brought about by changes in light intensity, the biofilm was first grown under constant illumination with the light source suspended 45 cm above the biofilm. This light intensity resulted in a relatively low PPFD of 25.57 µmol/m2/s. After 2 days of growth, the light source was lowered to 5 cm above the biofilm for a period of 20 h, resulting in an increased PPFD of approximately 141.79 µmol/m2/s. The light source was then returned to its original height for an additional 12 h. A similar experiment was conducted in which the light intensity illuminating an actively growing, mature biofilm was manipulated on a significantly shorter time scale. A mature photoautotrophic biofilm was initially illuminated from a height of 5 cm. The light was then raised to 45 cm, corresponding to a nearly ten-fold decrease in PPFD, for approximately 90 min. Following this, the biofilm was kept in complete darkness for 90 min, before being illuminated again for 90 min from 45 and 5 cm, respectively.

2.8. Variation in Bioﬁlm Biomass

2.8.1. Assessing CO2 Uptake with Supplementary Photoautotrophic Bioﬁlm

To demonstrate the modular nature of the CSMS, a third BR (BRcons2) was added to the system immediately downstream of Analyzer 2. The same photoautotrophic culture that was inoculated into BRcons was also inoculated into BRcons2. BRcons2 received its own feed of fresh M-BBM. The gas stream in the system however was sequential, with the sweeper gas ﬂowing from Analyzer 2, through the annular space of BRcons2 to a third CO2 analyzer (Analyzer 3).

2.8.2. Assessing CO2 Uptake with Increased Photoautotrophic Biofilm Length A similar experiment utilizing a third BR was conducted with sweeper gas flowing directly from BRcons via Analyzer 2 to BRcons2 and finally Analyzer 3. However, this time, the medium flow was also setup in sequence, such that the effluent of BRcons served as growth medium for BRcons2. Six mL of the photoautotrophic culture was injected into the silicone tubing just upstream of BRcons, which was enough to ensure that the inoculum reached the end of BRcons2. In this orientation, BRcons and BRcons2 represented two halves of one biofilm approximately 300 cm in length, with the positioning of Analyzer 2 between BRcons and BRcons2 allowing for the measurement of CO2 flux by each half of the biofilm separately.

3. Results and Discussion

3.1. Development of Algal Bioﬁlms

Algal biofilm growth became evident in BRcons within several hours after inoculation. The inflow and outflow CO2 concentrations over the reactor began to diverge due to increased capture by the developing photosynthetic biofilm (Figure3a). Conversion of concentration data into CO 2 uptake rate resembles a typical microbial growth curve (Figure3b), with a green biofilm visible (Figure4) when the uptake rate levelled off at approximately 2.8 µmol/h (378 µmol/h/m2). In order to confirm that the divergence depicted in Figure3a between Analyzer 1 (BR cons gas influent) and Analyzer 2 (BRcons gas effluent) was attributable to the activity of the photoautotrophic biofilm in BRcons, a procedure similar to the “light–dark shift method” [28] was performed. Briefly, this well-established method provides an assessment of photosynthetic activity by measuring the decrease in dissolved oxygen concentration that occurs when the specimen is placed in the dark after prolonged light exposure [28]. Originally described in the early 1980s, this method was developed to measure primary production in benthic sediments. Here, a similar procedure was used, whereby a photoautotrophic biofilm in BRcons was placed in the dark and the change in CO2 flux within it was observed and recorded. MicroorganismsMicroorganisms 20202020,, 88,, xx FORFOR PEERPEER REVIEWREVIEW 77 ofof 1515

Analyzer 2 between BRcons and BRcons2 allowing for the measurement of CO2 flux by each half of the Analyzer 2 between BRcons and BRcons2 allowing for the measurement of CO2 flux by each half of the biofilmbiofilm separately.separately.

3.3. ResultsResults andand DiscussionDiscussion

3.1.3.1. DevelopmentDevelopment ofof AlgalAlgal BiofilmsBiofilms

Algal biofilm growth became evident in BRcons within several hours after inoculation. The inflow Algal biofilm growth became evident in BRcons within several hours after inoculation. The inflow and outflow CO2 concentrations over the reactor began to diverge due to increased capture by the and outflow CO2 concentrations over the reactor began to diverge due to increased capture by the developing photosynthetic biofilm (Figure 3a). Conversion of concentration data into CO2 uptake rate developing photosynthetic biofilm (Figure 3a). Conversion of concentration data into CO2 uptake rate resemblesresembles aa typicaltypical microbialmicrobial growthgrowth curvecurve (Figure(Figure 3b),3b), withwith aa greengreen biofilmbiofilm visiblevisible (Figure(Figure 4)4) whenwhen Microorganisms 2020, 8, 1163 2 7 of 14 thethe uptakeuptake raterate levelledlevelled offoff atat approximatelyapproximately 2.82.8 µmol/hµmol/h (378(378 µmol/h/mµmol/h/m2).).

FigureFigure 3. BR3. consBRconswas was inoculated inoculated with with the photoautotrophic the photoautotro culturephic culture approximately approximately 24 h after 24 inoculation h after Figure 3. BRcons was inoculated with the photoautotrophic culture approximately 24 h after ofinoculation the heterotrophic of the heterotrophic culture into culture BR1. into (a) DivergenceBR1. (a) Divergence between between BRcons BRinfluentcons influent and and effl uenteffluent CO 2 inoculation of the heterotrophic culture into BR1. (a) Divergence between BRcons influent and effluent concentrationCO2 concentration (Analyzer (Analyzer 1 and 1 and 2, respectively) 2, respectively) was was apparent apparent withinwithin several hours hours of of inoculation. inoculation. CO2 concentration (Analyzer 1 and 2, respectively) was apparent within several hours of inoculation. (b)(( Usingbb)) UsingUsing the the idealthe idealideal gas law,gasgas thelaw,law, di thethefference differencedifference between betweenbetween the influent thethe influentinfluent and effl andanduent effluenteffluent concentration concentrationconcentration was converted waswas toconverted the rate of to CO the uptake,rate of CO which2 uptake, resembled which resembled a typical microbial a typical microbial growth curve. growth curve. converted to the2 rate of CO2 uptake, which resembled a typical microbial growth curve.

Figure 4. Image and close up of BRcons colonized by a mature photoautotrophic biofilm. Green biomass FigureFigure 4. 4.Image Image and and close close up up of of BR BRconscons colonized by by a a mature mature photoautotrophic photoautotrophic biofilm. bioﬁlm. Green Green biomass biomass is apparentisis apparentapparent within withinwithin the thethe inner innerinner silicone siliconesilicone tube. tube.tube.

In order to confirm that the divergence depicted in Figure 3a between Analyzer 1 (BRcons gas AsIn shown order in to Figure confirm5a, that the darkthe divergence response was depicted rapid, in with Figure the CO3a between2 concentration Analyzer in BR1 (BRconsconseffl gasuent influent) and Analyzer 2 (BRcons gas effluent) was attributable to the activity of the photoautotrophic increasinginfluent) sharplyand Analyzer to the 2 same (BRcons level gas aseffluent) the influent, was attributable and ultimately to the activity surpassing of the it inphotoautotrophic less than 10 min. biofilm in BRcons, a procedure similar to the “light–dark shift method” [28] was performed. Briefly, Thisbiofilm higher in CO BR2consconcentration, a procedure was similar likely to duethe “light–dark to some baseline shift method” respiration [28] by was the performed. algae that proceeded Briefly, this well-established method provides an assessment of photosynthetic activity by measuring the in thethis dark.well-established When plotted method as the provides rate of CO an2 assessmentuptake (Figure of photosynthetic5b), the e ffect ofactivity darkness by measuring on the biofilm’s the decrease in dissolved oxygen concentration that occurs when the specimen is placed in the dark after activitydecrease was in clearly dissolved apparent, oxygen demonstrating concentration that a complete occurs when cessation the specimen of photosynthesis. is placed in the This dark provided after evidence that the divergence observed between BRcons influent and effluent CO2 concentration was in fact due to CO2 fixation by the growing photoautotrophic biofilm. When the flow of photons to the biofilm is halted, the light-dependent photosynthetic reaction stops, and the ATP and NADPH needed to power the Calvin–Benson cycle and CO2 fixation is depleted, which happens rapidly as demonstrated by the data. The rapid dark response observed in the CSMS coincides with previous results using the light–dark shift method, where dramatic changes in photosynthetic rate were detected in a matter of seconds after switching to dark conditions [29]. Given that the CSMS is a closed, continuous flow system, one can expect a small lag in measurement time related to the length of tubing linking BRcons to Analyzer 2. This, however, can be mitigated by reducing this distance, such that the time needed for the gas to travel from the biofilm to the CO2 analyzer is minimized. Microorganisms 2020, 8, x FOR PEER REVIEW 8 of 15

prolonged light exposure [28]. Originally described in the early 1980s, this method was developed to measure primary production in benthic sediments. Here, a similar procedure was used, whereby a photoautotrophic biofilm in BRcons was placed in the dark and the change in CO2 flux within it was observed and recorded. As shown in Figure 5a, the dark response was rapid, with the CO2 concentration in BRcons effluent increasing sharply to the same level as the influent, and ultimately surpassing it in less than 10 min. This higher CO2 concentration was likely due to some baseline respiration by the algae that proceeded in the dark. When plotted as the rate of CO2 uptake (Figure 5b), the effect of darkness on the biofilm’s activity was clearly apparent, demonstrating a complete cessation of photosynthesis. This provided evidence that the divergence observed between BRcons influent and effluent CO2 concentration was in fact due to CO2 fixation by the growing photoautotrophic biofilm. When the flow of photons to the biofilm is halted, the light-dependent photosynthetic reaction stops, and the ATP and NADPH Microorganismsneeded to 2020power, 8, 1163 the Calvin–Benson cycle and CO2 fixation is depleted, which happens rapidly8 as of 14 demonstrated by the data.

Figure 5. A steady state photoautotrophic biofilm was placed in the dark by covering BRcons with an Figure 5. A steady state photoautotrophic biofilm was placed in the dark by covering BRcons with an opaqueopaque sheet sheet for for approximately approximately 4 4 h. h. TimeTime zerozero refersrefers to twenty hours hours post post inoculation. inoculation. (a) ( aWithin) Within minutes, the CO concentration in the BR gas effluent (Analyzer 2) increased rapidly and remained minutes, the CO2 2 concentration in the BRconscons gas effluent (Analyzer 2) increased rapidly and remained elevatedelevated for for the the duration duration of of the the dark dark period. period. AfterAfter illuminationillumination resumed, resumed, th thee effluent effluent concentration concentration quicklyquickly returned returned to to its itspre-dark pre-dark level. (b (b) )When When converted converted and and plotted plotted as the as rate the of rate CO of2 uptake, CO2 uptake, the

theplot plot depicts depicts a acomplete complete cessat cessationion of photosynthesis of photosynthesis and andCO2CO fixation2 ﬁxation in the in dark, thedark, demonstrating demonstrating the thepresence presence and and activity activity of ofthe the photoautotrophic photoautotrophic biofilm bioﬁlm in BR incons BR. cons.

ToThe further rapid confirm dark response that the observed divergence in the between CSMS coincides influent andwith e previousffluent CO results2 concentration using the light– is due to thedark presence shift method, of the photoautotrophicwhere dramatic changes biofilm, in a ph setotosynthetic of abiotic control rate were experiments detected werein a matter conducted. of Whenseconds concentrated after switching CO2 was to channeleddark conditions through [29]. the Given uninoculated that the CSMS system is at a twoclosed, flow continuous rates, the influent flow andsystem, effluent one CO can2 concentration expect a small remained lag in measurement virtually unchanged, time related demonstrating to the length thatof tubing without linking assimilation BRcons byto the AnalyzerMicroorganisms photoautotrophic 2. 2020This,, 8 , however,x FOR biofilm, PEER canREVIEW CO be2 loss mitigated was minimal by reducing (Figure this6 distance,). such that the time 9needed of 15 for the gas to travel from the biofilm to the CO2 analyzer is minimized. To further confirm that the divergence between influent and effluent CO2 concentration is due to the presence of the photoautotrophic biofilm, a set of abiotic control experiments were conducted. When concentrated CO2 was channeled through the uninoculated system at two flow rates, the influent and effluent CO2 concentration remained virtually unchanged, demonstrating that without assimilation by the photoautotrophic biofilm, CO2 loss was minimal (Figure 6).

Figure 6. Abiotic control tests were performed in which concentrated CO was channeled through Figure 6. Abiotic control tests were performed in which concentrated CO2 was2 channeled through the the uninoculateduninoculated system for for 0.25 0.25 h hat ata flow a ﬂow rate rate of 20 of mL/min 20 mL/ min(a), and (a), 2.5 and mL/min 2.5 mL (b/min). In both (b). Incases, both the cases,

the influentinfluent and and e ffleffluentuent COCO22 concentration remained remained virtually virtually un unchanged,changed, demonstrating demonstrating that thatwithout without assimilationassimilation by the by photoautotrophicthe photoautotrophic biofilm, biofilm, CO CO22 lossloss was was minimal. minimal.

The systemThe system described described here here may may allow allow more more eefficientﬃcient capture capture of of CO CO2 than2 than conventional conventional bioreactor bioreactor conﬁgurations,configurations, since since its bulkits bulk supply supply is is in in the the gasgas phasephase and and therefore therefore not not limited limited to its to solubility its solubility in in water.water. Gas transportGas transport across across the the silicone silicone wall wall was was described described in in detail detail inin [[22]22] when the the bulk bulk solutions solutions on the oppositeon the opposite sides are sides water are andwater air, and respectively, air, respective whichly, which was was followed followed by by experiments experiments that that showed showed high high linearity2 (R2 = 0.999) between CO2 measured in the gas phase and various dissolved CO2 linearity (R = 0.999) between CO2 measured in the gas phase and various dissolved CO2 concentrations. concentrations. In the present study, CO2 transport in the opposite direction was measured. Gas In the present study, CO transport in the opposite direction was measured. Gas permeability of silicon permeability of silicon2 has found applications such as gas exchange in blood [30] as well as membrane has foundgills for applications submarines suchand underwater as gas exchange stations in [31]. blood Here, [30 ]we as wellexplored as membrane the potential gills utility for submarinesof such permeability in microbial systems with biotechnological relevance.

3.2. Light Manipulation

Figure 7 depicts the CO2 uptake rate of the photoautotrophic biofilm grown under alternating 12 h light and dark periods, demonstrating the biofilm’s photosynthesis on–off toggling during light– dark cycling. The data shows a net increase in CO2 consumption during the light phases with increasing biofilm age. At the end of each dark period, the biofilm’s CO2 uptake rate rapidly rebounded to approximately the rate reached at the end of the previous light period, ultimately achieving a rate of approximately 5.5 µmol/h (743 µmol/h/m2) by the last light period. There was also a small, decreasing rate of CO2 uptake in each sequential dark phase. Although we do not expect CO2 consumption by direct biologically related reactions during periods of darkness, this may be the result of other biotic or abiotic processes (e.g., changes in pH) which would have affected the CO2- bicarbonate equilibrium.

Microorganisms 2020, 8, 1163 9 of 14 and underwater stations [31]. Here, we explored the potential utility of such permeability in microbial systems with biotechnological relevance.

3.2. Light Manipulation

Figure7 depicts the CO 2 uptake rate of the photoautotrophic biofilm grown under alternating 12 h light and dark periods, demonstrating the biofilm’s photosynthesis on–off toggling during light–dark cycling. The data shows a net increase in CO2 consumption during the light phases with increasing biofilm age. At the end of each dark period, the biofilm’s CO2 uptake rate rapidly rebounded to approximately the rate reached at the end of the previous light period, ultimately achieving a rate of approximately 5.5 µmol/h (743 µmol/h/m2) by the last light period. There was also a small, decreasing rate of CO2 uptake in each sequential dark phase. Although we do not expect CO2 consumption by direct biologically related reactions during periods of darkness, this may be the result of other biotic or abiotic processes (e.g., changes in pH) which would have affected the CO2-bicarbonateMicroorganisms 2020 equilibrium., 8, x FOR PEER REVIEW 10 of 15

Figure 7. The photoautotrophic biofilm in BRcons was illuminated on a cycle of alternating 12 h light Figure 7. The photoautotrophic biofilm in BRcons was illuminated on a cycle of alternating 12 h light and dark periods. On–off toggling of biofilm photosynthesis and CO2 uptake is readily apparent. and dark periods. On–off toggling of biofilm photosynthesis and CO2 uptake is readily apparent. Following each dark period, CO uptake rebounded to approximately the same level reached at the Following each dark period, CO2 2 uptake rebounded to approximately the same level reached at the end ofend the of previousthe previous light light period. period. Time Time zero zero refersrefers to twenty twenty hours hours post post inoculation. inoculation.

SinceSince the the overarching overarching industrial industrial goal goal ofof bio-sequestrationbio-sequestration is isto tomaximize maximize the theamount amount of CO of2 CO2 captured,captured, it may it may be temptingbe tempting to to provide provide thethe sequesteringsequestering culture culture with with continuous continuous 24 h 24illumination. h illumination. This wouldThis would prevent prevent the cessationthe cessation of CO of2 COfixation2 fixation which which occurs occurs in the in darkthe dark and and promote promote higher higher biomass accumulationbiomass accumulation rates. However, rates. dark However, periods dark are periods important are inimportant allowing in for allowing the repair for ofthe photosynthetic repair of machineryphotosynthetic and proteins, machinery which and may proteins, benefit which the culture’s may benefit long-term the culture’s performance long-term [ 5performance]. A key operational [5]. A key operational decision in bio-sequestration systems, therefore, is how to achieve maximum CO2 decision in bio-sequestration systems, therefore, is how to achieve maximum CO2 capture while capture while maintaining the long-term health of the sequestering culture. Biofilm accumulation is maintaining the long-term health of the sequestering culture. Biofilm accumulation is not an infinite not an infinite process, and ultimately slows down to a quasi-steady state as dictated by factors such process, and ultimately slows down to a quasi-steady state as dictated by factors such as flow regimes as flow regimes and gas exchange/transport through the biofilm matrix, which will lead to a and gas exchange/transport through the biofilm matrix, which will lead to a simultaneous lowering of simultaneous lowering of CO2 uptake. By enabling real-time in situ monitoring of the biofilm’s CO2 CO2 uptakeuptake. rates, By enabling the CSMS real-time may offer in an situ approach monitoring to manage of the biofilm’sreactor performance, CO2 uptake such rates, as thetiming CSMS of may offerharvesting, an approach while to managereal-time reactorCO2 uptake performance, rates may be such useful as timingin setting of an harvesting, optimal light–dark while real-time cycling CO2 uptakeregimen. rates may be useful in setting an optimal light–dark cycling regimen. PhotosynthesisPhotosynthesis is driven is driven by photon by photon energy energy that excites that electronsexcites electrons in the chloroplasts’ in the chloroplasts’ photosynthetic pigments.photosynthetic Through pigments. a cascade Through of redox a reactions,cascade of thisredox electron reactions, energy this electron is transiently energy stored is transiently as ATP and stored as ATP and NADPH, which in turn are used in the Calvin–Benson cycle to fix CO2 and NADPH, which in turn are used in the Calvin–Benson cycle to fix CO2 and generate glucose. Generally, generate glucose. Generally, a higher density of photons irradiating the cell will lead to a higher rate a higher density of photons irradiating the cell will lead to a higher rate of CO2 fixation. That is, of CO2 fixation. That is, however, before an upper light intensity threshold is reached and however,photoinhibition before an upperoccurs. lightThis intensitycritical threshold threshold is a is product reached of and nume photoinhibitionrous abiotic and occurs. biotic Thisfactors critical thresholdincluding is a productlight wavelength(s) of numerous and abiotic color temperatur and biotice, factors growth including system configuration light wavelength(s) and material and color temperature,properties, growth as well system as culture configuration cell density [5]. and As material such, it properties,is extremely asdifficult well as to cultureaccurately cell predict density [5]. the onset of photoinhibition of a culture in a given system without some degree of real-time monitoring of culture performance. The next series of tests was conducted to assess the ability to detect changes in the biofilm’s CO2 uptake rate brought about by fluctuations in light intensity. This was done on both long (Figure 8a) and short (Figure 8b) timescales. Figure 8a shows the CO2 uptake rate of a photoautotrophic biofilm growing in the system with the fluorescent growth light initially placed at a height of 45 cm above the biofilm, corresponding to a relatively low PPFD of 25.57 µmol/m2/s. During this period, the CO2 uptake rate gradually increased from approximately 0.8 to 2 µmol/h (108 to 270 µmol/h/m2). After 24 h, the light was lowered to 5 cm above the biofilm, resulting in an increased light intensity and PPFD of 141.79 µmol/m2/s. Within minutes, the biofilm’s CO2 uptake rate increased sharply by approximately 40%, before continuing to gradually increase at roughly the same trajectory as before the light intensity shift. After 20 h, the growth light was returned to a height of 45 cm, and the biofilm once again exhibited a near immediate response with its CO2 uptake rate rapidly falling by

Microorganisms 2020, 8, 1163 10 of 14

As such, it is extremely difficult to accurately predict the onset of photoinhibition of a culture in a given system without some degree of real-time monitoring of culture performance. The next series of tests was conducted to assess the ability to detect changes in the biofilm’s CO2 uptake rate brought about by fluctuations in light intensity. This was done on both long (Figure8a) and short (Figure8b) timescales. Figure8a shows the CO 2 uptake rate of a photoautotrophic biofilm growing in the system with the fluorescent growth light initially placed at a height of 45 cm above 2 the biofilm, corresponding to a relatively low PPFD of 25.57 µmol/m /s. During this period, the CO2 uptake rate gradually increased from approximately 0.8 to 2 µmol/h (108 to 270 µmol/h/m2). After 24 h, the light was lowered to 5 cm above the biofilm, resulting in an increased light intensity and PPFD of 2 141.79 µmol/m /s. Within minutes, the biofilm’s CO2 uptake rate increased sharply by approximately 40%, before continuing to gradually increase at roughly the same trajectory as before the light intensity shift. After 20 h, the growth light was returned to a height of 45 cm, and the biofilm once again Microorganisms 2020, 8, x FOR PEER REVIEW 11 of 15 exhibited a near immediate response with its CO2 uptake rate rapidly falling by approximately 40% 2 to 2.2 approximatelyµmol/h (297 µ 40%mol to/h /2.2m µmol/h). Similarly, (297 µmol/h/m when the2). Similarly, light intensity when wasthe light adjusted intensity on was a much adjusted shorter time scaleon a (Figuremuch shorter8b), the time same scale rapid (Figure responses 8b), the same were rapid observed. responses After were 100 observed. min in the After dark, 100 it min took in only 20 minthe for dark, the it biofilm’s took only CO 202 minuptake for the rate biofilm’s to return CO2 to uptake its pre-dark rate to return level. to its pre-dark level.

FigureFigure 8. To 8. assess To assess whether whether the the CSMS CSMS can can detect detect changes changes inin the biofilm’s bioﬁlm’s CO CO2 uptake2 uptake rate rate brought brought

aboutabout by small by ﬂuctuationssmall fluctuations in light in intensity,light intensit the heighty, the height of light of source light source illuminating illuminating BRcons BRwascons adjustedwas on bothadjusted long (aon) andboth short long ((ba) timescales.and short (b Heights) timescales. of 5 Heights and 45 cmof 5 referand to45 PPFDscm refer of to approximately PPFDs of µmol/mapproximately2/s and 25.57 µmol/mµmol/2m/s2 and/s, respectively.25.57 µmol/m2/s, respectively.

PhototrophicPhototrophic biofilms biofilms typically typically consist consist of aof photosynthetically a photosynthetically active active zone, zone, whichwhich is closest to to the light-source,the light-source, and a photosynthetically and a photosynthetically inactive inactive zone wherezone where cells cells are shadedare shaded by theby the cells cells of theof the active layer.active With increasinglayer. With lightincreasing intensities, light intensities, photon flux photon penetrates flux penetrates deeper intodeeper the into biofilm the biofilm and the and thickness the thickness of the photosynthetically active layer increases [14]. Therefore, at higher light intensities, a of the photosynthetically active layer increases [14]. Therefore, at higher light intensities, a larger larger proportion of the biofilm is photosynthetically active, and its gross CO2 uptake increases, as proportion of the biofilm is photosynthetically active, and its gross CO uptake increases, as depicted depicted in Figure 8. Collectively, these results coincide with the notion 2that increasing light intensity in Figure8. Collectively, these results coincide with the notion that increasing light intensity leads to a leads to a higher rate of CO2 fixation, and demonstrates the ability of the CSMS to detect real-time higherchanges rate of in CO CO2 2fixation, uptake caused and demonstrates by small fluctuations the ability in light of theintensity. CSMS This to detectsystem real-timetherefore also changes has in CO2 uptakeutility in caused its ability by to small monitor fluctuations and alert when in light light intensity. intensities This are above system the therefore critical threshold, also has which utility in its abilitywould to otherwise monitor andcause alert photoinhibition when light an intensitiesd consequent are diminished above the productivity. critical threshold, which would otherwise cause photoinhibition and consequent diminished productivity. 3.3. Variation in Biofilm Biomass 3.3. Variation in Biofilm Biomass Amongst the main objectives of bio-sequestration is to maximize the amount of CO2 capture and to valorize the biomass produced. This study utilized the CSMS at bench-scale to develop an Amongst the main objectives of bio-sequestration is to maximize the amount of CO2 capture and toexperimental valorize the approach biomass that produced. may be combined This study with bioprocess utilized the modelin CSMSg to at optimize bench-scale conversion to develop rates an of the algal biomass, and ultimately techno-economic analysis. First, an additional biofilm reactor experimental approach that may be combined with bioprocess modeling to optimize conversion (BRcons2) inoculated with the same photoautotrophic culture was added to the system. In this rates of the algal biomass, and ultimately techno-economic analysis. First, an additional biofilm configuration the biofilms in BRcons and BRcons2 each received fresh growth medium, while the gas reactorflowed (BRcons2 sequentially) inoculated through with theBRcons same and photoautotrophic Analyzer 2, to BR culturecons2 and was Analyzer added 3. to The the system.addition Inof this configurationAnalyzer the3 allows biofilms for a indirect BRcons comparisonand BR cons2of theeach CO2 receiveduptake rates fresh in BR growthcons and medium, BRcons2, respectively. while the gas As expected, both biofilms consumed CO2 at approximately the same rate for the duration of the experiment. The addition of a second photoautotrophic biofilm therefore appeared to double the system’s overall CO2-sequestering capacity (Figure 9a). In scenarios with a finite source of CO2, as is the case in this experiment, the CSMS provides a means to determine optimal reactor length or number of modular reactors. This may be relevant to applications aimed at CO2 capture downstream from industrial processes where space and cost savings become a factor. In a subsequent experiment, the system was configured such that both medium and gas passed sequentially from BRcons to BRcons2. That is, the medium effluent from BRcons was the medium influent of BRcons2. As such, BRcons and BRcons2 each represented half of one long biofilm which was approximately 300 cm in length. In this case, CO2 and nutrients are finite. Interestingly, when comparing the first half of the biofilm (BRcons) to the second half of the biofilm (BRcons2), the latter

Microorganisms 2020, 8, 1163 11 of 14

flowed sequentially through BRcons and Analyzer 2, to BRcons2 and Analyzer 3. The addition of AnalyzerMicroorganisms 3 allows 2020 for, 8, x a FOR direct PEER comparison REVIEW of the CO2 uptake rates in BRcons and BRcons2, respectively. 12 of 15 As expected, both biofilms consumed CO2 at approximately the same rate for the duration of the exhibited a diminished rate of CO2 uptake (Figure 9b). In this case, more than half of the observed experiment. The addition of a second photoautotrophic biofilm therefore appeared to double the CO2 uptake was occurring in the first half of the biofilm. It is possible that some key nutrient(s) had system’s overall CO -sequestering capacity (Figure9a). In scenarios with a finite source of CO as is been depleted by2 the time the growth medium reached the second half of the biofilm, resulting in an2, the caseoverall in thislower experiment, rate of CO2 the uptake. CSMS There provides is also a the means potential to determine that the waste optimal effluent reactor from length BRcons or is numberin of modularequilibrium reactors. with the This CO may2 in the be relevantgas phase, to causing applications a subsequent aimed reduction at CO2 capture in CO2 flux downstream across the from industrialsilicone processes membrane where in BRcons2 space. and cost savings become a factor.

FigureFigure 9. The 9. The CSMS CSMS was was amended amended with with an an additionaladditional biofilm bioﬁlm reactor reactor (BR (BRcons2cons2) downstream) downstream of BR ofcons BR. cons. The gasThe streamgas stream in thein the system system was was sequential, sequential, with the the sweeper sweeper gas gas flowing ﬂowing from from Analyzer Analyzer 2, through 2, through the annularthe annular space space of BR of cons2BRcons2to to a a third third CO CO22 analyzeranalyzer (Analyzer 3). 3). (a ()a Initially,) Initially, BR BRconscons andand BRcons2 BR werecons2 were eacheach inoculated inoculated with with the photoautotrophicthe photoautotrophic culture cultur separatelye separately and and received received its its own own fresh fresh feed feed of of growth

medium.growth In thismedium. configuration, In this configuration, the CO2 uptake the rateCO2 within uptake BR ratecons andwithin BR cons2BRconswere and approximatelyBRcons2 were the same,approximately meaning that the the same, presence meaning of athat second the presence photoautotrophic of a second biofilm photoautotrophic roughly doubled biofilm roughly the system’s doubled the system’s overall CO2-sequestering capacity. (b) Subsequently, the system was configured overall CO2-sequestering capacity. (b) Subsequently, the system was configured such that both medium such that both medium and gas passed sequentially from BRcons to BRcons2, producing a biofilm the and gas passed sequentially from BRcons to BRcons2, producing a biofilm the length of two BRs. In this length of two BRs. In this case, the CO2 uptake rate in the first half of the biofilm (BRcons) is higher than case, the CO2 uptake rate in the first half of the biofilm (BRcons) is higher than the second half of the the second half of the biofilm (BRcons2), likely due to the depletion of some key nutrient(s) by the time biofilm (BRcons2), likely due to the depletion of some key nutrient(s) by the time the growth medium the growth medium reached BRcons2. reached BRcons2. 3.4. System Utility and Future Considerations In a subsequent experiment, the system was configured such that both medium and gas passed The system described here utilizes a heterotrophic biofilm as a source of concentrated CO2. Once sequentially from BRcons to BRcons2. That is, the medium effluent from BRcons was the medium this biofilm reaches steady state, its CO2 output remains relatively consistent and provides an influent of BR . As such, BR and BR each represented half of one long biofilm which was inexpensivecons2 and renewable conssource of cons2carbon with which to study bio-sequestration by approximately 300 cm in length. In this case, CO and nutrients are finite. Interestingly, when comparing photoautotrophic biofilms. In this sense, the heterotrophic2 biofilm in BRprod can be viewed as a stand the firstin for half any of real-world, the biofilm CO2 (BR-emittingcons) tobiological the second process, half such of the as wastewater biofilm (BR treatmentcons2), the or various latter exhibited types a diminishedof fermentation. rate of COWhen2 uptake algal syst (Figureems are9b). installed In this to case, capture more CO than2 from half flue of gas the emitted observed from CO large2 uptake was occurringindustrial point in the sources, first half theof overarching the biofilm. goal It would is possible be to thatmaximize some the key ratio nutrient(s) of CO2 captured had been to CO depleted2 by thereleased time theto the growth atmosphere. medium While reached it is possible the second to increase half of this the ratio biofilm, by optimizing resulting inlight an exposure overall lower (Figure 8a,b) and increasing the biofilm surface area (Figure 9a,b), there is also the potential to re- rate of CO2 uptake. There is also the potential that the waste effluent from BRcons is in equilibrium with circulate the gas in a closed loop to achieve near 100% CO2 removal. the CO2 in the gas phase, causing a subsequent reduction in CO2 flux across the silicone membrane In algal biodiesel production, lipid accumulation is advantageous and can be induced by in BRcons2. starving the algal culture of nitrogen as well as other key nutrients [32]. Cells respond to this stress 3.4. Systemby directing Utility a higher and Future proportion Considerations of their energy and carbon toward neutral lipids, the feedstock for biodiesel production. However, this stress response is also accompanied by a decrease in growth rate, photosynthesis,The system described and CO2 fixation here utilizes [33–35]. aHence, heterotrophic a delicate bala biofilmnce must as abe source struck between of concentrated inducing CO2. Oncelipid this accumulation biofilm reaches through steady starvation state, its and CO maintain2 outputing remains a robust relatively biofilm with consistent an adequate and growth provides an inexpensiverate. By depicting and renewable a culture’s source photosynthetic of carbon with activity which in real to study time, a bio-sequestration system like the CSMS by photoautotrophic would have significant utility in algal biodiesel production by helping to optimize the balance between starvation, biofilms. In this sense, the heterotrophic biofilm in BRprod can be viewed as a stand in for any real-world, desired product yield through photosynthesis, and collapse of the culture. CO2-emitting biological process, such as wastewater treatment or various types of fermentation. In conventional algae culturing, CO2 is typically delivered via sparging into the liquid growth When algal systems are installed to capture CO2 from flue gas emitted from large industrial point medium. This, however, causes a significant amount of CO2 to be lost to the atmosphere, a pervasive sources, the overarching goal would be to maximize the ratio of CO2 captured to CO2 released to the

Microorganisms 2020, 8, 1163 12 of 14 atmosphere. While it is possible to increase this ratio by optimizing light exposure (Figure8a,b) and increasing the biofilm surface area (Figure9a,b), there is also the potential to re-circulate the gas in a closed loop to achieve near 100% CO2 removal. In algal biodiesel production, lipid accumulation is advantageous and can be induced by starving the algal culture of nitrogen as well as other key nutrients [32]. Cells respond to this stress by directing a higher proportion of their energy and carbon toward neutral lipids, the feedstock for biodiesel production. However, this stress response is also accompanied by a decrease in growth rate, photosynthesis, and CO2 fixation [33–35]. Hence, a delicate balance must be struck between inducing lipid accumulation through starvation and maintaining a robust biofilm with an adequate growth rate. By depicting a culture’s photosynthetic activity in real time, a system like the CSMS would have significant utility in algal biodiesel production by helping to optimize the balance between starvation, desired product yield through photosynthesis, and collapse of the culture. In conventional algae culturing, CO2 is typically delivered via sparging into the liquid growth medium. This, however, causes a significant amount of CO2 to be lost to the atmosphere, a pervasive issue with negative economic and environmental implications [36,37]. This problem could potentially be addressed by mimicking the CSMS’s configuration, where concentrated CO2 remains in a closed environment and is delivered to the culture across a permeable silicone membrane. While previous studies have discussed the use of membranes for this purpose [36,38,39], none to our knowledge have applied this specifically in the context of photoautotrophic biofilm CO2 sequestration monitoring and optimization. From the data presented, it can be concluded that the system described here is capable of the reliable, real-time detection and quantification of CO2 uptake by phototrophic biofilms. This information is critical in understanding how these industrially relevant biofilms respond to changes in their environments (e.g., light intensity), information which is necessary in order to optimize growth and maximize energy- and cost-efficiency. It may also have significant utility in examining the interplay between inorganic and organic carbon utilization in mixotrophic or mixed species phototrophic biofilms, which would be relevant in biofilm systems using a wastewater medium with variable composition. The fact that the CSMS provides a non-destructive, in situ quantification of CO2 uptake makes it a more attractive option than conventional stoichiometric analysis, in which at least a portion of the biofilm must be sacrificed for dry weight measurements [18,19]. The CSMS also allows for a more direct CO2 flux analysis than can be obtained from simply monitoring effluent pH, since this is influenced by numerous other factors (e.g., nitrogen assimilation) and in buffered systems pH changes may not adequately reflect CO2 dynamics [21]. Some studies have utilized microelectrodes integrated inside the biofilm itself to monitor gas flux and photosynthetic activity [29]. While this approach provides a spatial resolution and resolving power not possible with the CSMS, these electrodes often require frequent calibration and are vulnerable to interference from fouling and gas bubbles [29]. As the effects of climate change continue to increase in severity, there is a critical need for solutions which can slow or prevent the increase in CO2 in our atmosphere. Bio-sequestration by photoautotrophic biofilms represents one promising solution which may be a valuable complement to existing CO2 management strategies. The biomass generated has inherent commercial value, which can be used to subsidize the process and improve its economic viability [7]. However, process optimization under different environmental and operating conditions remains a major challenge [12]. Technologies like the CO2 sequestration monitoring system (CSMS) introduced here can significantly aid in this endeavor and will become increasingly important if we are to effectively mitigate the effects of CO2 on climate change.

Author Contributions: Project conceptualizing and planning was by P.R., O.K., S.N.L. and G.W. The project was executed was by P.R., who also wrote and assembled the manuscript. G.W. also wrote several passages. O.K., G.W. and S.N.L. each provided supervision, contributed to the analysis of collected data, and conducted numerous revisions of the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) in the form of a Postgraduate Doctoral Scholarship. Microorganisms 2020, 8, 1163 13 of 14

Acknowledgments: The authors would like to acknowledge Jamie Nguyen for her help with lab work that contributed to this article. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Steﬀen, W.; Broadgate, W.; Deutsch, L.; Gaﬀney, O.; Ludwig, C. The trajectory of the Anthropocene: The Great Acceleration. Anthr. Rev. 2015, 2, 81–98. [CrossRef] 2. Solomon, K.; Plattner, G.-K.; Knutti, R.; Friedlingstein, P. Irreversible climate change due to carbon dioxide emissions. Proc. Natl. Acad. Sci. USA 2009, 106, 1704–1709. [CrossRef]

3. Hosseini, N.S.; Shang, H.; Scott, J.A. Biosequestration of industrial oﬀ-gas CO2 for enhanced lipid productivity in open microalgae cultivation systems. Renew. Sustain. Energy Rev. 2018, 92, 458–469. [CrossRef] 4. Zhou, W.; Wang, J.; Chen, P.; Ji, C.; Kang, Q.; Lu, B.; Li, K.; Liu, J.; Ruan, R. Bio-mitigation of carbon dioxide using microalgal systems: Advances and perspectives. Renew. Sustain. Energy Rev. 2017, 76, 1163–1175. [CrossRef]

5. Bhola, V.; Swalaha, F.M.; Kumar, R.R.; Singh, M.; Bux, F. Overview of the potential of microalgae for CO2 sequestration. Int. J. Environ. Sci. Technol. 2014, 11, 2103–2118. [CrossRef] 6. Gouveia, L.; Batista, A.P.; Sousa, I.; Raymundo, A.; Bandarra, N.M. Microalgae in novel food products. In Food Chemistry Research Developments; Papadopoulos, K.N., Ed.; Nova Science Publishers, Inc.: New York, NY, USA, 2008; Chapter 2, pp. 75–111.

7. Judd, S.; Broeke, L.J.V.D.; Shurair, M.; Kuti, Y.; Znad, H. Algal remediation of CO2 and nutrient discharges: A review. Water Res. 2015, 87, 356–366. [CrossRef] 8. Jais, N.M.B.; Mohamed, R.; Al-Gheethi, A.; Hashim, M.K.A. The dual roles of phycoremediation of wet market wastewater for nutrients and heavy metals removal and microalgae biomass production. Clean Technol. Environ. Policy 2016, 19, 37–52. [CrossRef] 9. Deshmukh, S.; Kumar, R.; Bala, K. Microalgae biodiesel: A review on oil extraction, fatty acid composition, properties and effect on engine performance and emissions. Fuel Process. Technol. 2019, 191, 232–247. [CrossRef] 10. Goh, B.H.H.; Ong, H.C.; Cheah, M.Y.; Chen, W.-H.; Yu, K.L.; Mahlia, T.M.I. Sustainability of direct biodiesel synthesis from microalgae biomass: A critical review. Renew. Sustain. Energy Rev. 2019, 107, 59–74. [CrossRef] 11. Nagappan, S.; Devendran, S.; Tsai, P.-C.; Dahms, H.-U.; Ponnusamy, V.K. Potential of two-stage cultivation in microalgae biofuel production. Fuel 2019, 252, 339–349. [CrossRef] 12. Kesaano, M.; Sims, R.C. Algal biofilm based technology for wastewater treatment. Algal Res. 2014, 5, 231–240. [CrossRef] 13. Hoffmann, J.P. Wastewater Treatment with Suspended and Nonsuspended Algae. J. Phycol. 1998, 34, 757–763. [CrossRef] 14. Schnurr, P.J.; Allen, D.G. Factors affecting algae biofilm growth and lipid production: A review. Renew. Sustain. Energy Rev. 2015, 52, 418–429. [CrossRef] 15. Donlan, R.M. Biofilms: Microbial Life on Surfaces. Emerg. Infect. Dis. 2002, 8, 881–890. [CrossRef] 16. Choudhary, P.; Prajapati, S.K.; Kumar, P.; Malik, A.; Pant, K.K. Development and performance evaluation of an algal biofilm reactor for treatment of multiple wastewaters and characterization of biomass for diverse applications. Bioresour. Technol. 2017, 224, 276–284. [CrossRef] 17. Katarzyna, L.; Sai, G.; Singh, O.A. Non-enclosure methods for non-suspended microalgae cultivation: Literature review and research needs. Renew. Sustain. Energy Rev. 2015, 42, 1418–1427. [CrossRef] 18. Salih, F.M. Microalgae Tolerance to High Concentrations of Carbon Dioxide: A Review. J. Environ. Prot. 2011, 2, 648–654. [CrossRef] 19. Chisti, Y. Constraints to commercialization of algal fuels. J. Biotechnol. 2013, 167, 201–214. [CrossRef] 20. El-Sheekh, M.; Bedaiwy, M.Y.; E Osman, M.; Ismail, M.M. Mixotrophic and heterotrophic growth of some microalgae using extract of fungal-treated wheat bran. Int. J. Recycl. Org. Waste Agric. 2012, 1, 12. [CrossRef]

21. Padmanabhan, P. Technical insight on the requirements for CO2-saturated growth of microalgae in photobioreactors. 3 Biotech 2017, 7, 119. [CrossRef]

22. Kroukamp, O.; Wolfaardt, G.M. CO2 Production as an Indicator of Bioﬁlm Metabolism. Appl. Environ. Microbiol. 2009, 75, 4391–4397. [CrossRef] Microorganisms 2020, 8, 1163 14 of 14

23. Krzemińska,I.; Pawlik-Skowrońska,B.; Trzcińska,M.; Tys, J. Influence of photoperiods on the growth rate and biomass productivity of green microalgae. Bioprocess Biosyst. Eng. 2013, 37, 735–741. [CrossRef] [PubMed] 24. Yu, J.; Liberton, M.; Cliften, P.F.; Head, R.D.; Jacobs, J.M.; Smith, R.D.; Koppenaal, D.W.; Brand, J.J.; Pakrasi, H.B. Synechococcus elongatus UTEX 2973, a fast growing cyanobacterial chassis for biosynthesis using

light and CO2. Sci. Rep. 2015, 5, 8132. [CrossRef] 25. Chang, H.-X.; Huang, Y.; Fu, Q.; Liao, Q.; Zhu, X. Kinetic characteristics and modeling of microalgae Chlorella

vulgaris growth and CO2 biofixation considering the coupled effects of light intensity and dissolved inorganic carbon. Bioresour. Technol. 2016, 206, 231–238. [CrossRef] 26. Ghadakpour, M.; Bester, E.; Liss, S.N.; Gardam, M.; Droppo, I.; Hota, S.; Wolfaardt, G.M. Integration and Proliferation of Pseudomonas aeruginosa PA01 in Multispecies Biofilms. Microb. Ecol. 2014, 68, 121–131. [CrossRef] 27. Lambertsen, L.; Sternberg, C.; Molin, S. Mini-Tn7 transposons for site-specific tagging of bacteria with fluorescent proteins. Environ. Microbiol. 2004, 6, 726–732. [CrossRef] 28. Revsbech, N.P.; Jørgensen, B.B.; Brix, O. Primary production of microalgae in sediments measured by 14 - oxygen microprofile, H CO3 fixation, and oxygen exchange methods. Limnol. Oceanogr. 1981, 26, 717–730. [CrossRef] 29. Revsbech, N.P.; Jørgensen, B.B. Photosynthesis of benthic microflora measured with high spatial resolution by the oxygen microprofile method: Capabilities and limitations of the method. Limnol. Oceanogr. 1983, 28, 749–756. [CrossRef] 30. Spaeth, E.E.; Friedlander, S.K. The Diffusion of Oxygen, Carbon Dioxide, and Inert Gas in Flowing Blood. Biophys. J. 1967, 7, 827–851. [CrossRef] 31. Robb, W.L. Thin Silicone Membranes-Their Permeation Properties and some Applications. Ann. N. Y. Acad. Sci. 1968, 146, 119–137. [CrossRef] 32. Hu, Q.; Sommerfield, M.; Jarvis, E.; Ghirardi, M.; Posewitz, M.; Seibert, M.; Darzins, A. Microalgal triaceylglycerols as feedstocks for biofuel production: Perspectives and advances. Plant J. 2008, 54, 621–639. [CrossRef][PubMed] 33. Zhao, L.-S.; Li, K.; Wang, Q.-M.; Song, X.-Y.; Su, H.-N.; Xie, B.-B.; Zhang, X.-Y.; Huang, F.; Chen, X.-L.; Zhou, B.-C.; et al. Nitrogen Starvation Impacts the Photosynthetic Performance of Porphyridium cruentum as Revealed by Chlorophyll a Fluorescence. Sci. Rep. 2017, 7, 1–11. [CrossRef] 34. Liefer, J.; Garg, A.; Campbell, D.A.; Irwin, A.J.; Finkel, Z.V.Nitrogen starvation induces distinct photosynthetic responses and recovery dynamics in diatoms and prasinophytes. PLoS ONE 2018, 13, e0195705. [CrossRef] [PubMed] 35. Yang, L.; Chen, J.; Qin, S.; Zeng, M.; Jiang, Y.; Hu, L.; Xiao, P.; Hao, W.; Hu, Z.; Lei, A.; et al. Growth and lipid accumulation by different nutrients in the microalga Chlamydomonas reinhardtii. Biotechnol. Biofuels 2018, 11, 40. [CrossRef] 36. Carvalho, A.; Malcata, F. Transfer of Carbon Dioxide within Cultures of Microalgae: Plain Bubbling versus Hollow-Fiber Modules. Biotechnol. Prog. 2001, 17, 265–272. [CrossRef] 37. Blanken, W.; Schaap, S.; Theobald, S.; Rinzema, A.; Wijffels, R.H.; Janssen, M. Optimizing carbon dioxide utilization for microalgae biofilm cultivation. Biotechnol. Bioeng. 2016, 114, 769–776. [CrossRef]

38. Lee, Y.-K.; Hing, H.-K. Supplying CO2 to photosynthetic algal cultures by diffusion through gas-permeable membranes. Appl. Microbiol. Biotechnol. 1989, 31, 298–301. [CrossRef] 39. Xu, Z.; Wang, Y.; Chen, Y.; Spalding, M.H.; Dong, L. Microfluidic chip for automated screening of carbon dioxide conditions for microbial cell growth. Biomicrofluidics 2017, 11, 1–9. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).