Open Access
Add to BibSonomyAbstract
Most existing photobioreactors do a poor job of distributing light uniformly due to shading effects. One method by which this could be improved is through the use of internal wave-guiding structures incorporating engineered light scattering schemes. By varying the density of these scatterers, one can control the spatial distribution of light inside the reactor enabling better uniformity of illumination. Here, we compare a number of light scattering schemes and evaluate their ability to enhance biomass accumulation. We demonstrate a design for a gradient distribution of surface scatterers with uniform lateral scattering intensity that is superior for algal biomass accumulation, resulting in a 40% increase in the growth rate.
© 2014 Optical Society of America
1. Introduction
One of the biggest challenges of the 21st century will be the increasing demand for energy. Between 85 and 90% of the current global demand for energy is met by consuming fossil fuels [1], which is not only unsustainable but also produces carbon dioxide which has been correlated with global warming [2]. While the daily energy from the sun could meet the global yearly demand for energy [3], the intermittency of solar illumination and the storage and transport of the resulting energy remains a difficult challenge [4]. Fuels produced by solar-chemical reactions, such as biofuels, could replace fossil fuels with little change to current energy infrastructure while mitigating carbon dioxide emissions [5].
Amongst potential biofuels, microalgae represent a promising feedstock for sustainable development [6]. While corn, soybean, canola and palm oil are all used currently as feedstocks [7] they cannot realistically be used to completely replace petroleum-derived fuels in the near term. As described by Chisti [8], meeting 50% of current transportation fuel needs would require more than 100% of the current cropping area in the United States using this current mix of feedstock. By contrast, notwithstanding other resource constraints [9], algae could meet this need requiring between 1 and 3% of current cropping area [8]. In addition, algae grow and replicate at faster rates, can be cultivated on non-arable lands and require less water uptake compared to terrestrial crops [10].
Notwithstanding these advantages, there are many challenges associated with algal cultivation that need to be addressed to achieve cost competitiveness and large-scale commercialization [11–13]. First of all, the energy return on investment has not been convincingly demonstrated in any large scale photo-bioreactor to be above one, which is the minimum level where energy output exceeds energy input [14, 15]. Additionally, fuels produced by current algal cultivation are currently too expensive to be cost-competitive [16, 17]. According to Wijjfels and Barbosa [18], the scale of production needs to increase at least 3 orders of magnitude, with a concomitant decrease in the cost of production by a factor of 10.
A present, the most commonly used methods of algae cultivation rely on raceway ponds and photobioreactors [8]. Raceway ponds have been used to cultivate algae since the 1950s and therefore there is extensive experience relating to their use and construction [8]. Nonetheless, there are also some serious disadvantages ranging from temperature fluctuations, evaporative water loss, contamination, and optical dark zones caused by poor mixing [8, 19, 20]. In contrast, because photobioreactors are generally closed systems, they can better mitigate temperature fluctuations, water loss, and contamination. Reducing optical dark zones in these systems however typically requires active mixing which is more energetically expensive [8, 19–22]. Additionally, the productivity of both systems is strongly linked to the geographical location of the reactors, the algal species used and daily variation in weather conditions [23, 24].
Development of a light delivery methodology that could overcome the limits of optical dark zones without requiring active mixing enabling and sustain high density algal cultures could help to improve the cost competiveness of the process. In highly dense algal cultures, the algae shadow themselves preventing the light from penetrating through the depth [25]. Algae growth is highly dependent on light intensity; overexposure to sunlight creates reactive oxygen species which damage the photosynthetic machinery whereas underexposure is insufficient for growth [26, 27]. Many innovative designs that incorporate larger surface areas [28], air-lift driven raceway reactors [29, 30], surface-plasmon-based light backscattering [31], and evanescent excitation [32, 33] to better distribute light in the photobioreactor have been previously reported.
Another method that could be employed to eliminate the optical dark zones is to use controlled light-scattering from internal waveguides [34, 35]. In this type of scheme, light is transported into the depth of the culture using waveguides where it is released into the algal culture through various scattering mechanisms. Similar methodologies have long been employed for a number of industrial applications, most notably in edge-lit LED displays [36]. As a result, there are numerous scattering schemes that have already been developed and characterized including: the use of index-mismatched materials on the surface of the waveguide, embedded nanoparticles or other defects inside the waveguide, and shape distortions in the waveguides themselves [37–39]. Using waveguides to distribute light in bioreactors has also been previously demonstrated [32, 34, 40, 41]. Erickson et al. [42] proposed using of waveguiding structures to deliver light to photosynthetic organisms and microfluidics to introduce reactants and remove products [43]. Previous work has also demonstrated algae growth using evanescent excitation in optofluidic reactors [32] and using large vertically assembled slab-waveguides to scatter light into media [41]. In both of these works, however, algal growth was not uniform across the length of the reactors probably due to the decay of the intensity of the internally transmitting light in the waveguide as shown in Fig. 1(a).
Fig. 1 Scatterers on slabwaveguides for algal cultivation. (a) Algae can also be excited via evanescent waves where growth is confined closer to the surface of the waveguide; (b) Uniform distribution of scatterers results in non-uniform illumination across the length of the reactor; (c) Spatially varying the distribution of pillars results in more uniform illumination along the reactor; (d) SEMs of the pillars at different densities, from left to right is variance down the length of the reactor.
In this paper, we investigate the impact of different scattering schemes on algal biomass accumulation in bioreactors and demonstrate an illumination scheme that achieves uniform lateral illumination in the bioreactor by precisely varying the spatial distribution of light scattering pillars. We show that compared to other schemes, this gradient surface scatterers scheme is superior for algal growth by at least 40%. In the first two sections of the results, we report how to develop this gradient scheme by first investigating the angular scattering profiles of pillars on a surface of a slab waveguide and then optimizing the lateral uniformity of illumination using a light scattering pillar scheme for a thin layer of algae by using shallow dye channels. In the final results section, we test different surface scattering schemes in a bioreactor to validate improvements in growth rates due to uniform illumination. The materials and methods are presented at the end. It is hoped that the demonstrated improvements can lead to more cost-effective high-density bioreactors with lower operational costs and reduced water and energy consumption.
2. Results and discussion
2.1 Controlled light scattering using surface defects
When light strikes a plane interface from a material of lower to higher refractive index, part of the light is transmitted and reflected according to the Fresnel equations. For a thin layer of a higher index material on top of the glass slab-waveguide where the angle of incidence is equal to the angle of reflection (θi = θr) and the polarization does not change, the transmission and reflectance coefficient will not change for the case of a perfectly single-mode, monochromatic wave of incidence. For periodically placed structures of the higher index material like SU-8 shown in Fig. 1 however, the light coupling into the pillars and back into the waveguide may be significant. Nonetheless, as long as the coupling coefficient is low, both the intensity of the internally reflected wave and the transmitting scattered wave will decrease exponentially along the length of the waveguide.
As shown in Fig. 2(a), simulations were performed using a finite difference time domain software package in a 2-D environment with a single SU-8 (NA = 1.59) pillar on a glass surface with the incoming transmitting pane wave at five degrees. Such simulations revealed that there was negligible light that coupled into the pillars which would couple back into the glass. Assuming that the pillar structures on the waveguide are identical, as they appear to be in SEMs taken in Fig. 1, and using a laser input with low divergence, we predicted and verified the presence of interference patterns on a periodic placing of pillars as shown in Fig. 2(b). These interference patterns were then used as a quality control when fabricating the waveguide samples with the scatterers.
Fig. 2 Characterizing angular scattering from surface scatterers. (a) Results from the 2D FEM simulation environment; (b) periodic positions of the pillars scatter the laser light in predictable manner creating interference patterns; (c) Angular scattering profiles vary with respect to the side angle of incidence of the laser; (d) angular scattering profiles also seem to vary depending on the length along the waveguide when seen through pinholes at different locations from front (1cm from front edge) and back (3.5cm from front edge)
Simulations also showed that even with a single pillar under a single-mode, monochromatic wave of incidence, the resulting transmitted wave had an angular profile. This angular profile would matter on the angle of incidence of the internally reflecting wave. By placing a pinhole right next to the waveguide, the angular scattering profile was empirically measured under different input angles and at two different places on the slab-waveguide. As expected, as the input angle increased, the angular scattering profile became more perpendicular to the waveguide as seen in Fig. 2(c). There was also some change in the angular profile along different lengths along the slab waveguide, as seen in Fig. 2(d) which can be explained by the slight change in the distribution of the angle on incidences due to coupling into the pillar and back into glass.
For the design of an edge-lit LCD display, only the narrower range of scattering angles perpendicular to the waveguide which would be visible to an observer in the far-field need to be considered. However, in photo-bioreactors the algal culture will be in direct contact with the pillars and may have very small penetration depth. Therefore, the design of a bioreactor which incorporates these waveguides must take into account the entire range of scattering angles. This is especially true if the input source is incoherent, multimode, and multi-angled like the LED source that is used in the following sections.
2.2 Illumination uniformity
In order to insure uniformity of illumination (equivalent to integrating over the angular scattering profile), shallow channel dye experiments were conducted in reactors with depths of 300 µm to simulate a thick biofilm, as shown in Fig. 3(a).These experiments were conducted under conditions that were to be mirrored in the final bacteria experiments. The light scattering structures were made from SU-8, a negative photoresist commonly used in photolithography. While each internally transmitting angle, 𝜃, has a different extinction coefficient, k(𝜃), it was found that the total scattering intensity of the broad-angled, and multi-mode LED source used, when integrated over its transmission angles, could be well modeled with an integrated extinction coefficient, kint such that,
where S(x) is the scattering intensity integrated over the various angles along the length of the slab waveguide and A is some constant. This is illustrated in Fig. 3(c).Fig. 3 Characterizing longitudinal scattering illumination in shallow dye channels. (a) Schematic of shallow channel dye experiments; (b) the surface coverage along the length of the reactor of the posts required for uniform scattering; (c) the scattering along the length of the shallow dye channel when sample has uniform surface coverage of posts of 25%; (d) the scattering along the length of the shallow dye channel when sample has gradient surface coverage of pillars as in (b).
In fact, as expected the extinction coefficients were found to vary linearly with the density of scatterers present on the surface for low density surface scatterers:
where sc is the surface coverage of the scatterers, ki is the extinction coefficient for a sample with associated surface coverage sci. It was assumed that this would hold true for higher densities of surface scatterers as well. For a pillar coverage of 25%, the associated kint was found to be equal to −0.028/mm.For a scattering scheme where the extinction coefficients can be carefully controlled, it is easy to see that to achieve uniform scattering intensity across the length of the waveguide, it should vary as:
andwhere L is the length of the waveguide and kmax is the extinction coefficient as the end of the waveguide which is also the maximum extinction coefficient. Using Eq. (2) and (3), one can easily derive the surface coverage of pillars required to achieve uniform scattering as:The highest surface coverage of pillars that was achieved was at 50% coverage. Beyond this coverage, the pillars formed a film on the glass that would peel off the surface due to poor adhesion as expected [44]. Using this as the highest surface coverage, we were able to show relatively uniform scattering intensity across the length of the shallow channel dye reactors of 6 cm, as shown in Fig. 3(d) using samples with gradient coverage described in Eq. (5) and graphed in Fig. 3(b).
2.3 Improved growth rates in bioreactors
After developing the scattering scheme with uniform illumination, its impact was validated on algal growth rates by performing algal growth experiments by integrating a number of different scattering schemes in bioreactors (W: 4 cm × L: 6 cm × H: 6 mm) as detailed in the Materials and Methods. First, a scattering scheme was tested with no scatterers to replicate the conditions of previous work [32] using waveguides to grow algae using only evanescent excitation. Second, there were two distinct schemes with uniform density of scatterers. This included both a waveguide with uniform pillar coverage of 50% and a waveguide with a chemically etched surface. As reported in the materials, for the chemically etched surface, an etching cream was used to etch the glass surface isotropically resulting in a uniform roughness along the waveguide. In principal, one would expect this to behave the same as a waveguide with uniform coverage of SU-8 pillars. Finally, the gradient pillar scheme was also tested.
The Synechocystis S. PCC 6803 2x EFE algal strain was used in these experiments [45]. We envisioned that testing with this ethylene-secreting algae would validate that our method was compatible with such genetically-modified algae strains. During the experiment, the alga tends to settle down in the absence of active mixing. Therefore, the results obtained in the shallow channel dye experiments of the previous section should translate to these bioreactors even though they are significantly thicker.
The evanescently excitation scheme that has been used in previous papers [32] did yield significant growth of algal surfaces as shown in Fig. 4(a)when compared to surface algal presence on the day of inoculation as shown in Fig. 5(a).However, as in these previous works [32], this growth was not even across the length of the reactor. The surface coverage could be modeled well with an exponential decay as shown in Fig. 5(a). It is possible that for this scheme, the algae themselves scatter the light such that there is, in fact, an exponential decrease in the scattering intensity. However, the relatively low index contrast between the water and the algae means that they are relatively weak scatterers that are able to transmit little light within the waveguide.
Fig. 4 The surface coverage of photobioreactors with different scattering schemes over the course of three days. (a) evanescent excitation; (b) uniform density of posts at 50% coverage; (c) chemically etched waveguides; (d) gradient density of pillars
Fig. 5 (a) The total surface coverage as a function of the length after the first day for the different scattering schemes;(b) the total surface coverage for different scattering schemes over the course of the three days; (c) the scattering intensity across the width of a gradient pillar sample in shallow channel dye experiments; (d) a fluorescent image of the bacteria under the uniform density of posts at 50%. Notice that algal growth occurs only in between pillars and seems to be spatially confined.
The uniform coverage of pillars at the maximum surface coverage (50%) also yielded uneven growth across the length of the algal surface, as shown in Fig. 4(b). This can be further seen in Fig. 5(a) where the cumulative surface coverage across the length is graphed after one day. In fact, the scattering intensity should have a steep extinction coefficient, and because algal growth is highly dependent on the light intensity, this is not a surprising result. This uneven surface coverage seems to spread out at longer times. This is likely because of the poor settling of the algae on the surface with such high SU-8 surface coverage. Figure 5(d) illustrates that the algae seem to settle only in between the pillars. This may be due to the hydrophobic nature of SU-8. It is possible that there may be significant portion of the algal culture that was not allowed to settle on the surface violating the assumptions made to relate these reactors to the results derived in the shallow channel experiments. This could make the algal growth in this reactor grow in patches with spatially constrained colonies.
As shown in Fig. 4(c), the scattering scheme with chemically etched surfaces demonstrated strong growth over the course of the three days eventually spreading to cover the entire reactor. However, as shown in Fig. 5(a), the growth on the first day reveals that there is uneven growth characteristics over the length of the reactor. Because the chemically etch treatment was the same across the length of the reactor, the scattering properties should be similar yielding an illumination intensity that decreases exponentially along the length of the reactor. Because algal growth is highly dependent on light intensity, this creates an optimal zone for growth while the intensity outside of this zone is not suitable for growth. However, the growth is quite strong and the algal cultures do spread out to cover the entire reactor over the course of the three day experiment.
The gradient coverage of pillars yielded the strongest growth over the course of the three days eventually covering the entire surface as shown in Fig. 4(d). In these experiments, the density of pillars varied from 19% at the leading edge to 32% at the end. There does seem to be slight indications of uneven growth after the first day of growth but this may be explained by the disinclination of the algae to settle at higher SU-8 densities and possibly by slight variations in the empirical determination of the extinction coefficient in different samples of uniform coverage of pillars in the shallow channel dye experiments. Nonetheless, it is clear that the gradient coverage is uniquely superior to the other scattering schemes, both in relative uniformity of algal surface coverage and the growth of the surface coverage throughout the reactor, as seen in Fig. 5(a) and 5(b).
The surface coverage growth, shown in Fig. 5(b), was fit to a logistic function, in Eq. (6), to extract the characteristic growth rates of each scattering scheme:
where K is 1, and Po is 0.07 or the initial surface coverage of the chip. The equation fits the data very well especially for the chemical and gradient coverage scattering schemes. It also reveals that there are significant differentials in the growth rates for each scattering schemes. The doubling times for the algae in the total chip for the evanescently excited, maximum pillar coverage, chemically etched, and gradient pillar schemes were 4.34 days, 0.95 days, 0.29 days and 0.21 days respectively. The fastest growth rate was that of the scheme incorporating a gradient distribution of light-scattering pillars which had uniform intensity and showed a 40% increase in growth rate as compared to the chemically etched scheme with uniform surface roughness profile but non-uniform scattering intensity.There also seems to be significant variations in the growth of the algae across the width of the reactor. In all the reactors, it seems that growth is stronger in portions near the center. Indeed, this is also an expected result as we did not control for the variations across the width of the reactor. This variation in scattering intensity across the width of the reactor is a result of the fact that the panel of LEDs used to illuminate the reactor was smaller than the width of the waveguides. Of course, this is a problem with an easy solution if the center of the LED panels is aligned with the reactor and the variation in the scattering intensity across the width of the reactor is characterized as in Fig. 5(c) and controlled by varying the surface coverage across the width.
We imagine that our work could be applied in industrial bioreactors in multi-stack reactor designs where the light could be delivered to each waveguide using solar collectors. Of course, the reactor would ideally be more suited towards an engineered algae strand that could produce fuel. Such reactor designs are currently under investigation by our group [46]. Of course, the cost of fabricating the waveguides can be decreased by applying industrial alternatives to the SU-8 surface scatterers used here such as hot embossing on acrylic sheets. While such a reactor would be more costly with regards to capital investment, it would have higher productivity and lower running costs allowing for the potential of competitive photo-bioreactor technology if up-scaled. A comprehensive cost model of such a reactor would emerge after the productivity and operating costs of such reactors is thoroughly evaluated.
3. Materials and methods
3.1 Fabricating waveguide samples
SU-8 pillars were fabricated by spinning a 2.8 micron thick layer of SU-8 2002 (Microchem) on 1 mm borosilicate glass slides and patterned using hard contact exposure on the Suss MA6-BA6 Contact Aligner. The 25% uniform coverage waveguides were made by patterning an array of 5 µm by 5 µm pillars spaced 5 µm apart on the glass (Fig. 1). Gradients of different percent coverage at each mm along the glass were constructed by changing the spacing between the pillars. Chemically etched were fabricated by applying Armour Etch to glass slides for seven hours followed by rigorous rinsing in water.
3.2 Characterizing SU-8 surface scattering
Surface scattering experiments were conducted on a waveguide with a uniform distribution of SU-8 pillars. The surface was concealed except for a pinhole at either 1 cm or 3.5 cm from the leading edge. The leading edge of the slide was illuminated with a laser diode (Newport LPM660-30C) at various input angles with respect to the surface. The angular scattering profile in the altitudinal plane was acquired with a power meter (Thorlabs PM 100D) several inches from the pinhole while varying the capture angle. Simulations were run using FDTD Solutions simulation software. A plane wave internally transmitting at 5 degrees was used in the simulation.
3.3 Shallow channel dye experiments
Shallow channel dye experiments were conducted in waveguide samples bonded with a carbon-black PDMS (1:100) mold to create chambers with a thickness of 300 µm. The leading edge of the waveguides were then sealed with a carbon-black PDMS layer to allow excitation only through the edge slit from an array of LEDs (~630nm) spaced 2mm away. The chambers were filled with Alexa Fluor 680 dye (Life Technologies). The entire length of the chamber length was then imaged every 1mm apart at 700nm.
3.4 Genetically-modified algae
The Synechocystis S. PCC 6803 2x EFE algal strain was used in these experiments 45. This strain of algae is genetically modified to produce ethylene and can be more easily adopted for use inside bioreactors where it would continuously produce biofuel. The algae was maintained in BG 11 Medium under a yellow fluorescent lamp. To start the experiments, the bioreactors were inoculated using a dilution of 1:100 from an OD750 of 0.1. They were illuminated with an array of LEDs at 630 nm several mm away during the course of the experiment.
3.5 Bioreactors, imaging, and analysis
The SU-8 waveguides were bonded to a carbon-black PDMS mold to create chambers of sinle-stack algal bioreactors (W: 4 cm × L: 6 cm × H: 6 mm). The bottom of the waveguide was attached with a grid of 1mm2 boxes for the purposed of imaging. The bioreactor was sealed from all sides to block the entry of light except through the leading edge during operation. The experiments were run for three days. Fluorescent images were taken over the surface of the reactors daily over an area of 25mm by 40mm and the surface coverage information of the algae inside the waveguide was extracted using a method reported by Kalontarov et.al [47].
4. Conclusion
In this work, we designed and fabricated light scattering waveguides that uniformly illuminated photo-bioreactors to improve growth of algal cultures. This gradient scattering scheme was tested along with other scattering schemes and was shown to be superior both in terms of the uniformity of growth and total coverage resulting in improvements of growth rates of 40%. Our reactors can be easily stacked to enable us to distribute light efficiently when stacked in close proximity and are compatible with genetically modified strains that could produce biofuels directly. Ultimately, we hope this will reduce the maintenance and harvesting costs biofuels from photobioreactor.
Acknowledgment
This work was supported by the Advanced Research Project Agency – Energy (DE-AR0000312). This work was performed in part at the Cornell NanoScale Facility, a member of the National Nanotechnology Infrastructure Network, which is supported by the National Science Foundation (Grant ECCS-0335765). B. Pereyra’s research was supported through the 2013 NNIN REU Program.
References and links
1. D. L. Klass, Biomass for Renewable Energy, Fuels, and Chemicals (Elsevier Science, 1998).
2. P. M. Cox, R. A. Betts, C. D. Jones, S. A. Spall, and I. J. Totterdell, “Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model,” Nature 408(6809), 184–187 (2000). [CrossRef] [PubMed]
3. O. Morton, “Solar energy: A new day dawning? Silicon Valley sunrise,” Nature 443(7107), 19–22 (2006). [CrossRef] [PubMed]
4. N. S. Lewis and D. G. Nocera, “Powering the planet: chemical challenges in solar energy utilization,” Proc. Natl. Acad. Sci. U.S.A. 103(43), 15729–15735 (2006). [CrossRef] [PubMed]
5. A. Demirbas, “Biodiesel from oilgae, biofixation of carbon dioxide by microalgae: A solution to pollution problems,” Appl. Energy 88(10), 3541–3547 (2011). [CrossRef]
6. M. F. Demirbas, “Biofuels from algae for sustainable development,” Appl. Energy 88(10), 3473–3480 (2011). [CrossRef]
7. D. Pimentel and T. W. Patzek, “Ethanol production using corn, switchgrass, and wood; biodiesel production using soybean and sunflower,” Nat. Resour. Res. 14(1), 65–76 (2005). [CrossRef]
8. Y. Chisti, “Biodiesel from microalgae,” Biotechnol. Adv. 25(3), 294–306 (2007). [CrossRef] [PubMed]
9. R. Pate, G. Klise, and B. Wu, “Resource demand implications for US algae biofuels production scale-up,” Appl. Energy 88(10), 3377–3388 (2011). [CrossRef]
10. Y. Chisti and J. Y. Yan, “Energy from algae: Current status and future trends Algal biofuels - A status report,” Appl. Energy 88(10), 3277–3279 (2011). [CrossRef]
11. M. K. Lam and K. T. Lee, “Microalgae biofuels: A critical review of issues, problems and the way forward,” Biotechnol. Adv. 30(3), 673–690 (2012). [CrossRef] [PubMed]
12. C. Y. Chen, K. L. Yeh, R. Aisyah, D. J. Lee, and J. S. Chang, “Cultivation, photobioreactor design and harvesting of microalgae for biodiesel production: a critical review,” Bioresour. Technol. 102(1), 71–81 (2011). [CrossRef] [PubMed]
13. H. M. Amaro, A. C. Guedes, and F. X. Malcata, “Advances and perspectives in using microalgae to produce biodiesel,” Appl. Energy 88(10), 3402–3410 (2011). [CrossRef]
14. C. M. Beal, R. E. Hebner, M. E. Webber, R. S. Ruoff, and A. F. Seibert, “The Energy Return on Investment for Algal Biocrude: Results for a Research Production Facility,” Bioenerg Res 5(2), 341–362 (2012). [CrossRef]
15. L. F. Razon and R. R. Tan, “Net energy analysis of the production of biodiesel and biogas from the microalgae: Haematococcus pluvialis and Nannochloropsis,” Appl. Energy 88(10), 3507–3514 (2011). [CrossRef]
16. R. Davis, A. Aden, and P. T. Pienkos, “Techno-economic analysis of autotrophic microalgae for fuel production,” Appl. Energy 88(10), 3524–3531 (2011). [CrossRef]
17. A. Singh and S. I. Olsen, “A critical review of biochemical conversion, sustainability and life cycle assessment of algal biofuels,” Appl. Energy 88(10), 3548–3555 (2011). [CrossRef]
18. R. H. Wijffels and M. J. Barbosa, “An Outlook on Microalgal Biofuels,” Science 329(5993), 796–799 (2010). [CrossRef] [PubMed]
19. M. Janssen, J. Tramper, L. R. Mur, and R. H. Wijffels, “Enclosed outdoor photobioreactors: Light regime, photosynthetic efficiency, scale-up, and future prospects,” Biotechnol. Bioeng. 81(2), 193–210 (2003). [CrossRef] [PubMed]
20. E. M. Grima, F. G. A. Fernandez, F. G. Camacho, and Y. Chisti, “Photobioreactors: light regime, mass transfer, and scaleup,” J. Biotechnol. 70(1-3), 231–247 (1999). [CrossRef]
21. C. Posten, “Design principles of photo-bioreactors for cultivation of microalgae,” Eng. Life Sci. 9(3), 165–177 (2009). [CrossRef]
22. F. Lehr and C. Posten, “Closed photo-bioreactors as tools for biofuel production,” Curr. Opin. Biotechnol. 20(3), 280–285 (2009). [CrossRef] [PubMed]
23. P. M. Slegers, P. J. M. van Beveren, R. H. Wijffels, G. van Straten, and A. J. B. van Boxtel, “Scenario analysis of large scale algae production in tubular photobioreactors,” Appl. Energy 105, 395–406 (2013). [CrossRef]
24. P. M. Slegers, R. H. Wijffels, G. van Straten, and A. J. B. van Boxtel, “Design scenarios for flat panel photobioreactors,” Appl. Energy 88(10), 3342–3353 (2011). [CrossRef]
25. C.-G. Lee, “Calculation of light penetration depth in photobioreactors,” Biotechnology and Bioprocess Engineering 4(1), 78–81 (1999). [CrossRef]
26. R. Bosma, E. van Zessen, J. H. Reith, J. Tramper, and R. H. Wijffels, “Prediction of volumetric productivity of an outdoor photobioreactor,” Biotechnol. Bioeng. 97(5), 1108–1120 (2007). [CrossRef] [PubMed]
27. E. M. Grima, J. M. F. Sevilla, J. A. S. Perez, and F. G. Camacho, “A study on simultaneous photolimitation and photoinhibition in dense microalgal cultures taking into account incident and averaged irradiances,” J. Biotechnol. 45(1), 59–69 (1996). [CrossRef]
28. C. U. Ugwu, H. Aoyagi, and H. Uchiyama, “Photobioreactors for mass cultivation of algae,” Bioresour. Technol. 99(10), 4021–4028 (2008). [CrossRef] [PubMed]
29. W. B. Zimmerman, M. Zandi, V. Tesar, D. J. Gilmour, and K. Z. Ying, “Design of an airlift loop bioreactor and pilot scales studies with fluidic oscillator induced microbubbles for growth of a microalgae Dunaliella salina,” Appl. Energy 88(10), 3357–3369 (2011). [CrossRef]
30. B. Ketheesan and N. Nirmalakhandan, “Development of a new airlift-driven raceway reactor for algal cultivation,” Appl. Energy 88(10), 3370–3376 (2011). [CrossRef]
31. S. Torkamani, S. N. Wani, Y. J. Tang, and R. Sureshkumar, “Plasmon-enhanced microalgal growth in miniphotobioreactors,” Appl. Phys. Lett. 97(4), 043703 (2010). [CrossRef]
32. E. E. Jung, M. Kalontarov, D. F. R. Doud, M. D. Ooms, L. T. Angenent, D. Sinton, and D. Erickson, “Slab waveguide photobioreactors for microalgae based biofuel production,” Lab Chip 12(19), 3740–3745 (2012). [CrossRef] [PubMed]
33. M. D. Ooms, V. J. Sieben, S. C. Pierobon, E. E. Jung, M. Kalontarov, D. Erickson, and D. Sinton, “Evanescent photosynthesis: exciting cyanobacteria in a surface-confined light field,” Phys. Chem. Chem. Phys. 14(14), 4817–4823 (2012). [CrossRef] [PubMed]
34. K. Mori, H. Ohya, K. Matsumoto, H. Furuune, K. Isozaki, and P. Siekmeier, “Design for a bioreactor with sunlight supply and operations systems for use in the space environment,” Adv. Space Res. 9(8), 161–168 (1989). [CrossRef] [PubMed]
35. C. Y. Chen, G. D. Saratale, C. M. Lee, P. C. Chen, and J. S. Chang, “Phototrophic hydrogen production in photobioreactors coupled with solar-energy-excited optical fibers,” Int. J. Hydrogen Energy 33(23), 6886–6895 (2008). [CrossRef]
36. G. Harbers, W. Timmers, and W. Sillevis-Smitt, “LED backlighting for LCD HDTV,” J. Soc. Inf. Disp. 10(4), 347–350 (2002). [CrossRef]
37. J. G. Chang and Y. B. Fang, “Dot-pattern design of a light guide in an edge-lit backlight using a regional partition approach,” Opt. Eng. 46(4), 043002 (2007). [CrossRef]
38. C. G. Son, J. H. Yi, J. S. Gwag, J. H. Kwon, and G. Park, “Improvement of Color and Luminance Uniformity of the Edge-Lit Backlight Using the RGB LEDs,” J Opt Soc Korea 15(3), 272–277 (2011). [CrossRef]
39. J. F. V. Derlofske, “Computer modeling of LED light pipe systems for uniform display illumination,” SPIE Proceedings 4445, 119–129 (2001).
40. J. C. Ogbonna, T. Soejima, and H. Tanaka, “An integrated solar and artificial light system for internal illumination of photobioreactors,” J. Biotechnol. 70(1-3), 289–297 (1999). [CrossRef] [PubMed]
41. J. W. F. Zijffers, M. Janssen, J. Tramper, and R. H. Wijffels, “Design process of an area-efficient photobioreactor,” Mar. Biotechnol. (NY) 10(4), 404–415 (2008). [CrossRef] [PubMed]
42. D. Erickson, D. Sinton, and D. Psaltis, “Optofluidics for energy applications,” Nat. Photonics 5(10), 583–590 (2011). [CrossRef]
43. S. S. Ahsan, A. Gumus, and D. Erickson, “Redox mediated photocatalytic water-splitting in optofluidic microreactors,” Lab Chip 13(3), 409–414 (2013). [CrossRef] [PubMed]
44. A. del Campo and C. Greiner, “SU-8: a photoresist for high-aspect-ratio and 3D submicron lithography,” J. Micromech. Microeng. 17(6), R81–R95 (2007). [CrossRef]
45. J. Ungerer, L. Tao, M. Davis, M. Ghirardi, P. C. Maness, and J. P. Yu, “Sustained photosynthetic conversion of CO2 to ethylene in recombinant cyanobacterium Synechocystis 6803,” Energ Environ Sci 5(10), 8998–9006 (2012). [CrossRef]
46. E. E. Jung, A. Jain, N. Voulis, D. D. R. Doud, L. T. Angenent, and D. Erickson, “Stacked optical waveguide photobioreactor for high density algal cultures,” Bioresource Technol. (2014).
47. M. Kalontarov, D. Doud, E. E. Jung, L. Angenent, and D. Erickson, “In situ hollow fiber membrane facilitated CO2 delivery to a cyanobacterium for enhanced productivity,” RSC Adv. (2013).
