Nazaire, France; telephone: +33-2-40-37-42-20; fax: +33-2-40-37-40-71; e-mail: amuller@ ifremer.fr. Received 8 October 2002; accepted 10 July 2003.
Swirling Flow Implementation in a Photobioreactor for Batch and Continuous Cultures of Porphyridium cruentum Arnaud Muller-Feuga,1 Je ´re ´my Pruvost,2 Rolland Le Gue ´des,2 Loic Le De ´an,1 2 2 Patrick Legentilhomme, Jack Legrand 1 Institut Franc ¸ais pour l’Exploitation de la Mer (IFREMER), BP 21105, 44311 Nantes Cedex 03, France 2 Laboratoire GEPEA – UMR CNRS 6144, Universite ´ de Nantes, CRTT-IUT, SaintNazaire, France; telephone: +33-2-40-37-42-20; fax: +33-2-40-37-40-71; e-mail: amuller@ ifremer.fr
Received 8 October 2002; accepted 10 July 2003 Published online 24 September 2003 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.10818
Abstract: Light is the main limiting factor in photoautotrophic-intensive production of microorganisms, and improvement of its use is an important concern for photobioreactor design and operation. Swirling flows, which are known to improve mass and photon transfers, were applied to annular light chambers of a photobioreactor and studied by simulation and microalgal culture. Two hydrodynamic conditions were compared: axial flow generating poor radial mixing, and tangential flow generating three-dimensional swirling motion. Batch and continuous cultures of the Rhodophyte Porphyridium cruentum were performed in a 100-L, 1.5-m2, fully controlled photobioreactor with eight light chambers. The inlet design of these chambers was modified to create the hydrodynamic conditions for comparison. Various intensities of swirling motion were used, characterized by the velocity factor (VF), defined as the ratio between annular chamber flow and inlet aperture sections. Experiments were performed within the range of photon flux densities (PFD) optimizing the yield of light energy transformation into living substance for the species and the temperature used. Culture kinetics with swirling flows generated by apertures of VF = 2, 4, and 9 were compared with pseudoaxial VF = 2 chosen as reference. Batch cultures with VF = 4 swirling flow showed no significant difference, whereas continuous cultures proved more discriminating. Although no significant difference was obtained for VF = 2, a 7% increase of steady-state productivity and a 26% decrease in time required to reach this steady state were obtained with VF = 4 swirling flow. This beneficial effect of swirling flow could have accounted for increased mixing. Conversely, VF = 9 swirling flow resulted in a 9% decrease of steady-state productivity and a 9% increase in the time required to reach this steady state, a negative effect that could have accounted for increased shear stress. CO2 bioconversion yield at steady state showed a 34% increase for VF = 4. These results suggest that swirling motion makes microalgal cultures more efficient, provided that the resulting adverse effects remain acceptable. Experimental investigation was completed by a theoretical approach in which simulation of continuous cultures of P. cruentum was based on the hydrodynamic conditions achieved in the photobioreactor. Although the
Correspondence to: A. Muller-Feuga
B 2003 Wiley Periodicals, Inc.
results obtained with pseudoaxial flow were correctly predicted, simulations with swirling flow showed a marked enhancement of productivity not observed experimentally. The influence of side effects induced by increased mixing (particularly hydrodynamic shear stress) was considered with respect to modeling assumptions. Comparison of experimental results with theoretical simulation provided a better understanding of the mixing effect, a key factor in improving the efficiency of such bioprocesses. B 2003 Wiley Periodicals, Inc. Biotechnol Bioeng 84: 544 – 551, 2003.
Keywords: photobioreactor; microalgae; swirling flow; Porphyridium cruentum; light bioconversion
INTRODUCTION Photochemical engineering is of considerable interest for modeling and reactor design. Liquid sterilization by ultraviolet radiation, hydrogen production by solar photolysis of water, and catalytic degradation of organic pollutants in water are some examples of industrial photochemical processes (Abu-Ghararah, 1997; Ollis et al., 1991; Ray and Beenackers, 1997; Sczechowski et al., 1995). As photosynthesis also involves light, this study considered the possibility of applying certain photochemical advances (especially light and mass transfers) to photobioreactors to improve their efficiency. Increasing absorption in the reaction medium as a result of optical path-length and concentration makes light the main limiting factor in photoautotrophic production of microorganisms. Thus, any change in hydrodynamic conditions likely to increase the bioconversion yield of this substrate is of particular interest for photobioreactor design and operation. Couette – Taylor vortices, which appear in cylindrical annular space when one of the cylinders is rotated, are known to improve transfers and have been applied to reactor design for chemical and sometimes photochemical reactions. Significant improvement of chemical conversion has been found for fast reactions and strong light absorption (Haim and Pismen, 1994). Photoefficiency was increased nearly threefold for heterogeneous photocatalysis (Sczechowski
et al., 1995). Some attempts have been made to use these efficient hydrodynamic conditions in photosynthesis. Miller et al. (1964) found that photosynthesis increased with rotor speed, but this factor has certain limitations considering the adverse effect of shear forces on microalgae. As the implementation of such vortices complicates reactor design and limits reaction volume, other types of rotating flow patterns have been considered for photobioreactor applications. Dean-type roll-cells generated by centrifugal forces in a twisted duct are also of interest for mixing, in which case irregular trajectories may be obtained by changes in the curvature plane of consecutive bends (Peerhossaini et al., 1993). This technique was used in strongly curved tubular photobioreactors for the culture of Spirulina platensis, but gas injection disturbances occurred, and the beneficial effect was difficult to distinguish from that of turbulence (Carlozzi and Torzillo, 1996). Swirling flows generated by tangential inlets in an annular cavity separating two static cylinders would appear to be of special interest, as they require no particular rotation equipment and create no additional pressure drop or contact surface, contrary to the effects produced by turbulence- and vortex-generating baffles and inserts. Moreover, previous studies of this flow pattern (Farias Neto et al., 1998; Lefe`bvre et al., 1998; Legentilhomme and Legrand, 1991; Legentilhomme et al., 1993; Pruvost et al., 1999, 2000) have provided a solid basis for further investigation. In the present experimental study, swirling decaying flows were investigated as a means of improving the efficiency of light bioconversion by microalgae. The effects of this flow regime were compared with those of classical Poiseuille flow in which velocities are mainly axial. MATERIALS AND METHODS To assess the efficiency of swirling flow as compared with mainly axial flow, batch and continuous cultures of the Rhodophyte Porphyridium cruentum from the Go¨ttingen collection (B 112.79) were performed in a 100-L artificiallight photobioreactor (ALP). The equipment (Fig. 1), a previous version of which has been described (MullerFeuga et al., 1998), consisted of a reaction loop with eight light chambers connected in series and measuring 1500 mm in length, with an internal diameter of 100 mm. A commercial fluorescent tube was placed centrally within another transparent coaxial tube with an external diameter of 40 mm. The annular space of each chamber was 10 L in volume and 30 mm in radial depth. The external light wall of the central glass tubes provided a developed photosynthetic surface area of 1.5 m2. The inlet and outlet of the reaction medium were located at opposite ends of each chamber, in low and high positions, respectively. The eight chambers were placed horizontally and the culture was driven upward, so that the poisoning photosynthetic oxygen was conveyed to the uppermost area of the loop where the gas outlet was located. The culture then circulated through an oblique
Figure 1. General views of the artificial light photobioreactor used for experimentation.
section down to the bottom of a gas-lift, which provided both impulse force and gas – liquid contact. A propeller circulator completed the gas-lift impulse. The separation of gas bubbles from liquid occurred in a glass drum, before the culture was driven to the bottom of the light chamber series. CO2 and air were injected into the gas loop for pH and oxygen regulation, and excess gas was evacuated to keep the pressure constant and equal to a preset value of 1.5 bar. The reactor was equipped with independent control-command loops, which regulated the main parameters of the culture. CO2 injection into the inlet of the lowermost chamber regulated pH, and oxygen was stripped by injection of air in the gas-lift at a constant rate of 7.5 L min1 (4.5 vvh). Culture temperature was regulated by a heating
MULLER-FEUGA ET AL.: SWIRLING FLOW IN CULTURES OF P. CRUENTUM
545
Figure 2. Design of the end of the light chamber, indicating the location of the intermediary disks used to change the hydrodynamic configuration of the photobioreactor.
element covering the oblique section and a cooling system consisting of an air extractor that circulated ambient air within the gap between the lamp and the central transparent tube of the light chambers. The alternating light sources were a 1500-mm-long Satin 182 commercial fluorescent lamp and a Grolux lamp (Sylvania Co.). These lamps delivered a mean photosynthetically active PFD of 236 and 175 Amol photon m2 s1, respectively, at the contact of the inner glass wall of the light chambers, as measured by a sensor (Li-Cor LI 190 SA). The culture medium was enriched artificial seawater (Hemerick, 1973). Various hydrodynamic conditions were obtained simply by modifying the inlet configurations of the eight light chambers. For this purpose, the chambers assembled by stacking were made watertight by using four drawstrings to compress the rubber rings on intermediary disks. Figure 2 provides further indications about light chamber design and the adaptive inlet configuration. The geometrical characteristics of intermediary disks are a critical point in hydrodynamic optimization with swirling flows, as shown by Aouabed et al. (1995), who used a dot-paint visualization method with tangential inlets. According to Legentilhomme and Legrand (1991), who studied various configurations of tangential inlets of circular type, swirling decaying motion can be classified in terms of inlet diameter, $e, and annular gap width, e: (i) pure swirling flow is achieved if $e = e, (ii) convergent swirling flow if $e e; and (iii) divergent swirling flow if $e< e. Mass flow conservation increases velocity in the inlet for smaller inlet diameters, resulting in more intensive swirl motion in annular space. However, microalgae are known to be
sensitive to hydrodynamic stress (Baldyga and Pohorecki, 1998; Jaouen et al., 1999; Mitsuhachi et al., 1995; Vandanjon et al., 1999), so that mixing needs to be limited because of cell fragility. In the case of convergent swirling flow, inlet diameter is larger than annular gap width, so that the inlet jet impinges on the inner glass tube, reducing initial rotating motion and causing hydrodynamic shear stress. Thus, one means of improving inlet efficiency is to prevent impact on the inner glass tube by changing the initial circular shape of the inlet. The alternative oblong shape proposed here (Fig. 2) reduces jet impact. The geometric characteristics of the oblong inlets investigated were chosen to obtain the same crosssection as for corresponding radial inlets of circular shape. Inlets can be characterized by three parameters: the nature of the inlet (namely radial or tangential); its shape (circular or oblong); and a velocity factor (VF). This last parameter is defined here as the ratio between annular chamber flow and inlet aperture sections, which is equal to the inverse ratio of the velocities in corresponding sections. The intermediary disks used are described in Table I. Swirling flows were obtained in configurations 2 to 4, whereas configuration 1 generated mainly axial flow, a hydrodynamic condition generally achieved in tubular photobioreactors and used here as reference. Axial VFs above 2 were excluded because of their adverse effects on the microalgal population. For the same reason, only pure and convergent swirling flows were investigated to keep the hydrodynamic stress field within an acceptable range. A $e/e ratio of less than unity would cause intensive initial swirl motion, resulting in large recirculation areas and high velocity gradients, as observed by Legentilhomme and Legrand (1991). These flows have been described fully in a previous study (Pruvost et al., 2002a). To make this comparison more discriminating, cultures were performed without any nutrient limitation, and the light source was chosen to obtain the highest light bioconversion yield. This parameter, defined as growth rate divided by PFD, expresses the efficiency of light transformation into living substance. High PFD, such as those obtained with midday sun, were avoided as they result in photoinhibition, and consecutive decrease of this yield. Dermoun et al. (1992) found that the saturating PFD at which photoinhibition appears is 438 Amol photon m2 s1 for Porphyridium cruentum exposed to the same temper-
Table I. Different configurations created by changing the intermediary disks used in experimentation. Intermediary disk no.
a
Velocity factor Position of the inlet Shape of the aperture Type of flow a
546
1 (reference)
2
3
4
2 Axial Round Axial
2 Tangential Oblong Swirling
4 Tangential Oblong Swirling
9 Tangential Round Swirling
The velocity factor (VF) is the ratio between chamber flow and inlet aperture sections.
BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 5, DECEMBER 5, 2003
Figure 3. Light bioconversion yield as a function of photon flux density for Porphyridium cruentum at 25jC. Data were recalculated from Dermoun et al. (1992) and adjusted by the function described by Muller-Feuga (1999).
ature (25jC) as in our experiments. Calculation of light bioconversion yields (Fig. 3) gave the highest values for lowest PFD. Adjustment of a function previously developed to account for this variation (Muller-Feuga, 1999) showed that light bioconversion yield was maximal for 65 Amol photon m2 s1. Thus, the initial PFD provided by light sources (175 and 236 Amol photon m2 s1) were beyond saturation. Because of light absorption in the reaction medium, they then decreased to values optimizing the light bioconversion yield. The photobioreactor was cleaned and steam-sterilized prior to each inoculation. The temperature and pressure attained during this operation were 125jC and 2.5 bar, respectively, over 40 min. Steam was cautiously admitted through gas and liquid sterilizing filters at the inlet and the outlet of the photobioreactor during this sterilizing operation, according to a previously defined precise and microbiologically validated protocol, and the sensors were calibrated. After temperature returned to a nominal value, the photobioreactor was filled with Hemerick medium through sterilizing filters up to the nominal level minus the 5-L inoculum volume. The monoalgal inoculum was prepared at 25jC in an incubator, which required 10 to 15 days to produce a concentration of 500,000 cells mL1 in the reactor. KNF Stepdos pumps were used for medium renewal. The inoculum was introduced into the photobioreactor in aseptic conditions at counterflow through the
Figure 4. Time-course of batch culture biomass concentration (dry weight). Comparison of axial inlet VF = 2 (disk 1) with tangential inlet VF = 4 (disk 3). Error bars represent the standard deviation.
Figure 5. Time-course of continuous culture surface productivity (dry weight) and fitting to the hyperbolic tangent function for different hydrodynamic conditions. (A) VF = 2 radial inlet (disk 1). (B) VF = 2 tangential inlet (disk 2). (C) VF = 4 tangential inlet (disk 3). (D) VF = 9 tangential inlet (disk 4). Error bars represent the standard deviation.
MULLER-FEUGA ET AL.: SWIRLING FLOW IN CULTURES OF P. CRUENTUM
547
harvesting outlet to complete the volume. Dry weight concentration was determined from four biomass samples centrifuged at 1500g for 12 min at 10jC, rinsed twice with isotonic ammonium formiate to eliminate salts, dried at 70jC for 24 h in a ventilated oven, and weighed with a 0.1-mg precision balance. The mean bulk velocity induced by both the marine propeller and the gas-lift was determined beforehand using a conductivity meter with salt as tracer (results not shown). Cultures were thus performed with a constant mean bulk velocity of 5 cm s1 applied in the light chamber. In both axial and tangential flows, the Reynolds number in the light chambers was 3200, indicating a turbulent regime. Circulation was checked daily, as well as culture parameters (temperature, pressure, pH, absorption), all being computer-processed and recorded in run-time. RESULTS Experimental Batch cultures were performed only for configurations 1 and 3, for at least 45 days (Fig. 4). Dry weight concentration was measured every working day. The effects of nutrient deprivation were apparent from day 15, resulting in a lower increase of dry weight concentration. Injection of nitrate and phosphate at day 20 resulted in another concentration increase. The fact that culture growth kinetics was very similar for both investigated configurations indicated that a more accurate method was required for further investigations of the effect of swirling flow. Continuous cultures were chosen for this purpose because of their high sensitivity to ambient factors during dynamic equilibrium between growth and dilution. At least two continuous cultures were performed for each configuration described in Table I. Medium renewal began as soon as the reactor was inoculated to eliminate any discrepancy factor between culture conditions and to facilitate comparisons. Dilution rate was maintained at 0.23 day1 for at least 21 days. As renewal was not strictly steady, results were expressed in terms of surface productivity, P (grams per square meter per day); that is, daily dry weight production divided by the photosynthetic
surface area, which was here the developed external surface of the central glass tubes. For comparative purposes, experimental results were fitted by the minimizing residual variance between the surface productivities measured and calculated with the hyperbolic tangent mathematical function [Eq. (1)], where P(t) is productivity as a function of time t: t ð1Þ PðtÞ ¼ Pmax tanh T This function provided two key criteria retained here: steady-state productivity, Pmax, and the time-constant, T; thus, the abscissa of the intercept of the tangent at origin with the asymptote P(t) = Pmax. This time-constant T is indicative of the duration of transient growth, or of the time required to reach steady state. The time-courses of productivity and their fitting are shown in Figure 5. The function chosen for fitting proved satisfactory, except for the S shape obtained with tangential inlets VF = 2 (Fig. 4B) and VF = 9 (Fig. 4D), which were not properly rendered. The results of fitting experimental data with P(t) function for the different hydrodynamic configurations are summarized in Table II. The residual variance remained at V0.5. The highest Pmax was obtained for VF = 4 tangential inlet, with a 7% significant increase compared with the VF = 2 radial inlet chosen as reference. T was also improved by 26%. VF = 2 radial and tangential inlets exhibited significantly identical values of Pmax and T, whereas VF = 9 tangential inlet gave results lower than reference values. These findings indicate that swirling flow did not differ from pseudoaxial flow for low VF, but was more efficient for intermediary VF and less efficient for high VF. The CO2 bioconversion rate was another key criterion used for comparison. It is understood here as the ratio of the mass of CO2 mobilized in the biomass to the mass of CO2 injected into the photobioreactor per unit of time. Figure 6 shows that this ratio was 34% higher for swirling (6.7% [w/w]) than for axial flow (5.0% [w/w]). Theoretical In addition to the experimental study, a theoretical approach was conducted to consider the influence of cell trajectories on light availability and corresponding photo-
Table II. Results of fitting the productivity time-course of continous cultures by the function P(t) = Pmax tanh (t/T), where Pmax is the surface productivity (dry weight) at steady state and T the constant expressing the time required to reach steady state. VF VF== 2 radial (reference) (disk 1) Mean (n = 2) 2
Pmax (g m T (d) Variance
1
d )
11.6 8.7 0.4
SD 1.1 2.2
VF = VF = 2 tangential (disk 2) Mean (n = 2) 11.3 9.0 0.5
SD
Gain (%)
0.7 1.5
3 3
VF = 4 VF = 4 tangential (disk 3)
VF = 9 (VF = 9 tangential (disk 4)
Mean (n = 2)
Mean (n = 2)
12.5 6.5 0.4
SD
Gain (%)
0.2 2.0
7 26
SD, standard deviation; Gain, relative difference with reference values.
548
BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 5, DECEMBER 5, 2003
10.6 9.5 0.4
SD
Gain (%)
0.5 0.5
9 9
Figure 6. Time-course of the CO2 bioconversion rate. Comparison of axial inlet VF = 2 (disk 1) with tangential inlet VF = 4 (disk 3).
bioreactor productivity. The modeling method for batch culture simulations of P. cruentum has been described fully by Pruvost et al. (2002a). The same algorithm was applied in this study for the same hydrodynamic conditions as those investigated experimentally, except that the continuous culture case was considered. For this purpose, the following equation was substituted in the algorithm for the one relating to batch-type culture: A¼
dC D Cdt
ð2Þ
where A is the specific growth rate, C the culture concentration in dry weight, and D the dilution rate. The main feature of this modeling is the representation of cell displacement by accurate determination of microalgal trajectories (Pruvost et al., 2002b). Consequently, the amount of light received by a cell when flowing in the photobioreactor can be deduced by considering light attenuation in the depth of the culture. Finally, the physical parameters relative to the modeling of light and flow were coupled to biological modeling parameters representing the photosynthetic growth of P. cruentum. Simulations were achieved for the same mean bulk velocity (5 cm s1) as applied experimentally in the photobioreactor. The results of this simulation are compared (Fig. 7) with those of experimentation for pseudoaxial flow with VF = 2 (Fig. 7a) and swirling flow with VF = 4 (Fig. 7b). Simulation of the surface productivity time-course showed marked enhancement with swirling flow. The Pmax and T values obtained were 19.2 g m2 day1 and 10.8 days, respectively, for VF = 4 swirling flow, and 10.8 g m2 day1 and 7.8 days for VF = 2 pseudoaxial flow. The latter values were in good agreement with experimental results, whereas the former largely overpredicted the productivity achieved.
volume productivity (Muller-Feuga et al., 2003). This probably accounts for the limited industrial development of photobioreactors, despite the considerable attention that has been paid to their design and modeling (Cornet et al., 1998; Molina et al., 2001; Pulz, 2001; Richmond et al., 1993). Thus, any increase in the efficiency of the photosynthetic reaction should be of great interest. In a concentrated microalgal culture, photonic energy decreases rapidly with the distance to the light source (radial distance in the present study). One part of the culture remains in the dark, whereas another part receives too much energy for proper transformation. Mixing increases photosynthesis by renewing the population relative to the light source, and providing a delay to restore the photosynthetic apparatus and recover from high PFD damages. Both of these beneficial effects of mixing occur for PFD higher than saturation, whereas only the former occurs for PFD under saturation. The first situation is recommended for high photosynthetic efficiency (Janssen et al., 2003), whereas the second one prevailed here. The best increase in productivity obtained was of the same order as that of Carlozzi and Torzillo (1996) for a strongly curved reactor, as compared with a straight tube reactor, in outdoor cultivation of Spirulina. They attributed this benefit to an intermittent illumination pattern resulting from secondary flow motion generated in the bends, an effect similar to that obtained with swirling flow.
DISCUSSION The need to maintain a delicate balance between access to light for growth and mutual cell shading makes production yields much lower for photoautotrophic microorganisms than for heterotrophic ones with fermentation; that is, one order of magnitude for concentrations and two orders for
Figure 7. Time-course of continuous culture surface productivity (dry weight) and simulation by the light availability model for different hydrodynamic conditions. (a) VF = 2 radial inlet (disk 1). (b) VF = 4 tangential inlet (disk 3).
MULLER-FEUGA ET AL.: SWIRLING FLOW IN CULTURES OF P. CRUENTUM
549
Our comparison of experimental differences favored swirling flow generated by intermediate inlet velocity, which gave two improvements: a 7% increase in surface productivity at steady state and, especially, a 27% decrease in the time required to reach steady state. The absence of any significant difference between the results obtained with low inlet velocity and pseudoaxial flow used as reference may be explained by the weakness of the swirling motion obtained, which was not sufficient to influence photobioreactor productivity. Conversely, the negative results obtained with high inlet velocity could have been due to the adverse effects of shear forces, which generate a mechanical stress resulting in lower growth. As no evidence of such stress was observed in the cytomorphology of the cell, this explanation remains largely hypothetical. Although P. cruentum is generally regarded as a resistant species, excessive mechanical stress may result in lower productivity (Gudin and Chaumont, 1990). A mean bulk velocity of 25 cm s1 has been identified as a standard value for photobioreactor design for this species (Chaumont, 1995). As the highest velocity (45 cm s1) was very localized in the inlet constriction, it is likely that the adverse effects of these accelerations were tolerable. Moreover, hydrodynamic shear forces exerted during repeated passages through the constricted inlets may have caused cell damage. Bubble break-up at the gas – culture interface and hydrodynamic shear forces have been identified as possible factors responsible for cell fragility (Garcia Camacho et al., 2000). In the comparison considered here, only the second of these factors was involved, as gas injection conditions remained unchanged in the cases examined. Changes in hydrodynamic conditions mainly affected light availability and shear-stress intensity. Thus, the experimental results obtained provide an overall view of flow influence, integrating several induced effects. Our modeling approach constitutes one way of separating these effects. The discrepancy between experimental and modeling results can be explained by the assumptions on which modeling was based. Our intention was not to represent real culture conditions fully, but to consider the influence of flow on light availability and corresponding photosynthetic growth. However, the good predictions obtained with axial flow showed the validity of this modeling approach for culture conditions in which unconsidered factors such as shear stress are not influential. Thus, the difference between the predicted and experimental culture values observed with the swirling flow induced by high inlet velocity could have been due to the negative effect of this parameter. The increased productivity observed experimentally, although indicative of greater light availability, was lower than predicted, because of the influence of other restrictive factors such as shear stress. However, the much higher productivity (Pmax) theoretically predicted in less time (T) with swirling flow is encouraging. Further investigations are needed to determine the hydrodynamic conditions most suitable for light availability, while reducing adverse effects such as shear stress.
550
Nonetheless, it is difficult to discriminate experimentally between adverse and favorable effects of mixing. The inclusion of only a limited number of parameters in the modeling approach allows certain aspects of photobioreactor operation to be better understood, and the parallel approaches of modeling and experimental investigation used in the present study facilitate analysis of the influence and optimization of flow. Another important influence of flow observed in this study was the gas – liquid mass transfer effect. As the consumption of inorganic nutrients by microalgae tends to reduce water or gas loads, CO2 biofixation is one of the possible applications of photobioreactors. Thus, the efficiency of CO2 bioconversion was of particular interest in our study. In our experiments, the general photobioreactor design included gas-lift systems for both culture circulation impulse and oxygen stripping, in combination with pHregulated CO2 injection. This set-up generally led to poor CO2 bioconversion rates, paradoxically resulting in CO2 release into the atmosphere. Using solar photobioreactors, Mazzuca Sobczuk et al. (2000) showed that CO2 losses are closely related to the energy hitting the culture, ranging from over 100% at night, due to respiration, down to 20% for high PFD (2000 Amol photon m2 s1). Their values for low PFD (200 Amol photon m2 s1) are similar to those obtained here. The large improvement in CO2 bioconversion rate with intermediary swirling flow could have been related to an increase of mass transfer due to liquid renewal at the gas – liquid interface of the upper part of the chambers, where undissolved CO2 accumulated. It may be concluded that implementation of swirling flow in photobioreactors has a beneficial effect. Productivity is higher at a steady state reached in less time, and the CO2 bioconversion rate is improved, at least for P. cruentum, provided that the hydrodynamic conditions induced do not generate too many adverse effects (e.g., mechanical shear stress) for microalgae. The results obtained are promising for future investigations intended to design photobioreactors with improved mixing conditions adapted to biological constraints.
References Abu-Ghararah Z. 1997. A numerical model for estimating average light intensity in annular ultraviolet disinfection reactors. Environ Technol 18:929 – 936. Aouabed H, Legentilhomme P, Legrand J. 1995. Wall visualization of swirling decaying flow using a dot-paint method. Exper Fluids 19:43 – 50. Baldyga J, Pohorecki R. 1998. Influence of turbulent mechanical stresses on microorganisms. ASME Appl Mech Rev 51:121 – 140. Carlozzi P, Torzillo G. 1996. Productivity of Spirulina in a strongly curved outdoor tubular photobioreactor. Appl Microbiol Biotechnol 45:18 – 23. Chaumont D. 1995. Bases physiologiques et technologiques de la conception de photobiore´acteurs pour la culture controˆle´e de microalgues. Doctorate thesis Aix-Marseille, University of Marseilles, Marseilles, France. 246 p. Cornet JF, Dussap CG, Gros JB. 1998. Kinetics and energetics of photosynthetic microorganims in photobioreactors. Biotechnology 59: 153 – 224.
BIOTECHNOLOGY AND BIOENGINEERING, VOL. 84, NO. 5, DECEMBER 5, 2003
Dermoun D, Chaumont D, Thebault JM, Dauta A. 1992. Modelling of growth of Porphyridium cruentum in connection with two interdependent factors: Light and temperature. Biores Technol 42:113 – 117. Farias Neto SR, Legentilhomme P, Legrand J. 1998. Finite-element simulation of laminar swirling decaying flow induced by means of a tangential inlet in an annulus. Comput Meth Appl Mech Eng 165: 189 – 213. Garcia Camacho F, Contreras Gomez T, Mazzuca Sobczuk E, Molina Grima E. 2000. Effects of mechanical and hydrodynamic stress in agitated, sparged cultures of Porphyridium cruentum. Proc Biochem 35:1045 – 1050. Gudin C, Chaumont D. 1990. Cell fragility—the key problem of microalgae mass production in closed photobioreactors. Biores Technol 38:145 – 151. Haim D, Pismen LM. 1994. Performance of a photochemical reactor in the regime of Taylor-Go¨rtler cortical flow. Chem Eng Sci 49:1119 – 1129. Hemerick G. 1973. Culture methods and growth measurements. In: Stein JR, editor. Handbook of physiological methods. Cambridge, UK: Cambridge University Press. p 259 – 260. Janssen M, Tramper J, Mur LR, Wijffels RH. 2003. Enclosed outdoor photobioreactors: Light regime, photosynthetic efficiency, scale-up, and future prospects. Biotechnol Bioeng 81:193 – 210. Jaouen P, Vandanjon L, Que´meneur F. 1999. The shear stress of microalgal cell suspensions (Tetraselmis suecica) in tangential flow filtration systems, the role of pumps. Biores Technol 68:149 – 154. Lefe`bvre G, Farias Neto SR, Aouabed H, Legentilhomme P, Legrand J. 1998. Transfert de matie`re et chute de pression lors d’un e´coulement tourbillonnaire non-entretenu induit par une entre´e tangentielle. Can J Chem Eng 49:1119 – 1129. Legentilhomme P, Legrand J. 1991. The effects of inlet conditions on mass transfer in annular swirling decaying flow. Int J Heat Mass Transfer 34:1281 – 1291. Legentilhomme P, Aouabed H, Legrand J. 1993. Developing mass transfer in swirling decaying flow induced by tangential inlet. Chem Eng J 52:137 – 147. Mazzuca Sobczuk T, Garcia Camacho F, Camacho Rubio F, Acien Fernandez FG, Molina Grima E. 2000. Carbon dioxide uptake efficiency by outdoor microalgal cultures in tubular airlift photobioreactors. Biotechnol Bioeng 67:465 – 475. Miller RL, Fredrickson AG, Brown AH, Tsuchiya HM. 1964. Hydromechanical method to increase efficiency of algal photosynthesis. Indust Eng Chem Proc Design Devel 3:134 – 143.
Mitsuhachi S, Hosaka K, Tomonaga E, Muramatsu H, Tanishita K. 1995. Effects of shear flow on photosynthesis in a dilute suspension of microalgae. Appl Microbiol Biotechnol 42:744 – 749. Molina E, Ferna´ndez J, Acie´n FG, Chisti Y. 2001. Tubular photobioreactor design for algal cultures. J Biotechnol 92:113 – 131. Muller-Feuga A. 1999. Growth as a function of rationing, a model applicable to fish and microalgae. J Exp Mar Biol Ecol 236:1 – 13. Muller-Feuga A, Kaas R, Moal J. 2003. The microalgae of aquaculture. In: Stottrup J, McEvoy L, editors. Live feeds in marine aquaculture. London: Blackwell Science. p 207 – 252. Muller-Feuga A, Le Gue´des R, Herve´ A, Durand P. 1998. Comparison of artificial light photobioreactors and other production systems using Porphyridium cruentum. J Appl Phycol 10:83 – 90. Ollis DF, Pelizetti E, Serpone N. 1991. Destruction of water contaminants. Environ Sci Technol 25:1523 – 1529. Peerhossaini H, Castelain C, Leguer Y. 1993. Heat exchanger design based on chaotic advection. Exp Therm Fluid Sci 7:333 – 344. Pruvost J, Legrand J, Legentilhomme P. 1999. Transfert photonique dans un photore´acteur a` e´coulement tourbillonnaire. Can J Chem Eng 77: 869 – 876. Pruvost J, Legrand J, Legentilhomme P, Doubliez L. 2000. Particle image velocimetry investigation of the flow-field of a 3D turbulent annular swirling decaying flow induced by means of a tangential inlet. Exp Fluids 29:291 – 301. Pruvost J, Legrand J, Legentilhomme P, Muller-Feuga A. 2002a. Simulation of microalgae growth in limiting light conditions—flow effect. AIChE J 48:1109 – 1120. Pruvost J, Legrand J, Legentilhomme P, Muller-Feuga A. 2002b. Trajectory Lagrangian model for turbulent swirling flow in an annular cell. Comparison with RTD measurements. Chem Eng Sci 57:1205 – 1215. Pulz O. 2001. Photobioreactors: Production systems for phototrophic microorganisms. Appl Microbiol Biotechnol 57:287 – 293. Ray A, Beenackers A. 1997. Novel swirl-flow reactor for kinetic studies of semiconductor photocatalysis. AIChE J 43:2571 – 2578. Richmond A, Boussiba S, Vonshak A, Kopel R. 1993. A new tubular reactor for mass production of microalgae outdoors. J Appl Phycol 5:327 – 332. Sczechowski JG, Koval CA, Noble RD. 1995. A Taylor vortex for heterogeneous photocatalysis. Chem Eng Sci 50:3163 – 3173. Vandanjon L, Rossignol N, Jaouen P, Robert JM, Que´meneur F. 1999. Effects of shear on two microalgae species (Skeletonema costatum and Haslea ostrearia). Contribution of pumps and valves in tangential flow filtration systems. Biotechnol Bioeng 63:1 – 9.
MULLER-FEUGA ET AL.: SWIRLING FLOW IN CULTURES OF P. CRUENTUM
551
