Team:MADRID UCM/Proof Of Concept
Proof Of Concept
During the last stage of our project we have wondered many times about how our technology will eventually look like, how we could develop our technology and most important of all how sustainable it will be in a real scenario. To answer all of these questions, we have performed a simulation, where process engineering principles have been applied to shape our technology within the real world. Within this page you will find how we have taken the results of our project and translated them into a clearer image of their real application.
Simulating a Photosynthetic Biomanufacturing Process
After studying how we can implement our technology in a real industrial installation, we realized how product recovery and purification was the most critical step in the process. Then, we decided to outline how the first sketch of our technology will look like.
Within our Implementation proposal we have studied how to develop a viable bioprocess, selecting butanol adsorption as the most promising downstream strategy for efficient product recovery. Based on the available bibliography of n-butanol adsorption, we have decided to design these downstream operations in wider detail in order to prove the feasibility of our technology.
n-Butanol Purification: Fixed-bed adsorption systems
In a fixed bed adsorption operation, a fluid (liquid or gas) is introduced from one of the fixed-bed ends, forcing it to flow through the bed of adsorbent and exiting the bed in the opposite end.. During this process the molecules of the compound of interest are selectively adsorbed within the material, attending to their different physicochemical properties that makes it strongly interact with the adsorbent surface. This adsorption phenomena typically occurs within the micropores of the adsorbent material, which offer a high surface area for adsorption. Once all the available surfaces have been covered” with the molecule of interest, the fixed-bed is “saturated” and the fixed bed needs to be regenerated for product recovery.
The regeneration of the fixed-bed system allows product recovery. This regeneration can be performed via thermal desorption: a sudden increase in bed temperature that liberates the product from the adsorbent. Another common approach consists of the introduction of a fluid whose conditions (temperature or composition) change the affinity of the adsorbent by the product of interest, allows to liberate them again, and recover the product with high purity in the fluid stream exiting the system.
In the case of n-butanol, thermal desorption can be used, after the adsorption stage, the bed can be heated and the adsorbed molecules are released. This way the n-butanol can be transferred to an air stream that can be furtherly cooled down to recover the n-butanol as a liquid with high purity.
Several materials have been evaluated for n-butanol adsorption. While some novel materials like hydrophobic metal-organic frameworks have shown remarkable properties, we have decided to use a cheap and highly available material: silicalite. This material is composed of silicon oxide, showing similar hydrophobic properties to hydrophobic zeolites. Its water-uptake is remarkably low, being possible to consider it almost neglectable, while tha n-butanol adsorption capacities are high,allowing the adsorption of up to 86 g of butanol per each kilogram of material in optimal conditions. However, the interesting part is that at temperatures over 100ºC, the adsorption capacity is lower than 2 g per kilogram, allowing an efficient product recovery.
Fixed-bed systems are usually utilized by pairs, in order to allow a continuous product recovery. One of the beds is utilized for product adsorption. When this first column reaches the saturation point, the second fixed-bed starts to be used for adsorption and the first one enters within the regeneration stage. The main advantage of these systems is their robustness. They are easy to build and the adsorbent can be reused many times.
Fixed bed adsorption systems can be fed with either liquid or gaseous streams, allowing us to recover the product directly from the aeration stream (when it is volatilized at a high rate) or from the liquid supernatant obtained after separating the photobiocatalyst from the culture media. However, when liquid streams are fed into the fixed bed, before the desorption step the adsorbent needs to be dried in order to eliminate the water or other supernatant components that are wetting the bed but not adsorbed in it. This drying step can be achieved first by bed drainage and after it removing the remaining liquid by simultaneous heating and air-drying.
When thermal desorption of the bed is required, a critical aspect of the design is how adsorbent temperature can be increased homogeneously. The easiest way to heat the fixed bed is using external electrical resistances, wrapped around the bed material and insulated from the environment, however, if the fixed bed diameter is too high, heat transfer will be slow, and adsorbent won’t be heated homogeneously. Then a multitubular fixed bed system will be considered for its design.
In the following subsections a more detailed explanation about the design and simulation of the different downstream scenarios is presented. To directly see the results, click here.
Evaluating technological feasibility: Bioprocess Design and Simulation
The design of a purification process involves the identification and sizing of the required equipment, as well as the definition of the parameters required for its operation. To achieve this design challenge, first of all is to define the parameters of the initial material from which n-butanol needs to be recovered.
We have made the following considerations to define the starting point of the purification process:
-
Steady-state operation. We have considered performing the n-butanol bioproduction under steady state conditions. This way, cells / photocatalysts are cultured until a certain amount of n-butanol is produced. Then a side stream is extracted from the reactor and replaced with fresh media, looking to keep the product concentration stable. This way, a balance between the biosynthesis of n-butanol and the amount of product exiting the system is established.
Productivity. Basing on the Lindblad et al. results, we have considered a n-butanol productivity of 12,5 g*m-3*h-1. This value was obtained in the laboratory with mild illumination and semi-continuous cultivation. This productivity could be enhanced with further engineering and optimization of culture conditions in industrial photobioreactors. Because of that we have later evaluated different productivity levels in order to determine the overall process viability.
-
n- butanol concentration. Considering the conclusions of available literature about n-butanol tolerance of Synechococcus elongatus, we have determined that a value lower than 5 g*L-1 could be tolerated by engineered strains without severely compromising cellular fitness.
We have taken into account that the lower n-butanol concentration, the higher n-butanol productivity could be achieved, since product toxic effects and biosynthesis pathway inhibition by its reversible equilibrium reactions would be less relevant.
Considering all of this, we have simulated the downstream process for multiple productivity - n-butanol concentration pairs. After this, we have identified which are the limits for a viable recovery process.
Details about the mathematical models used for the design of the operations, as well as calculations about energy consumption can be found in the Modelling page.
Bioprocess Simulation. The procedure
For downstream process evaluation we have initially designed a gas-stripping operation followed by n-butanol adsorption in gas-phase. After finding some drawbacks with this process we decided to also perform the design of a liquid-phase adsorption recovery process, where major drawbacks of the former downstream process could be overcomed.
The evaluation procedure has 4 steps...
1. First, a set of realistic n-butanol productivity - n-butanol concentration pais has been defined, taking the productivity values documented by Lindblad et al. as the central reference, and defining the rest of value pairs as a variation of the former in terms of productivity and n-butanol concentration. Different combinations of high/low n-butanol tolerance and high/low n-butanol productivity have also been studied.
2. Second, for each one of these cases, the downstream process is designed, taking the required considerations for the definition of a realistic purification process. During the design of the downstream process the assumption of n-butanol concentration and productivity has been used, to determine the flow of the photobioreactor existing stream. During this design step, those cases where the concentration-productivity pairs lead to unrealistic design parameters have been identified and omitted from the evaluation.
For downstream process evaluation we have initially designed a gas-stripping operation followed by n-butanol adsorption in gas-phase. After finding some drawbacks with this process we decided to also perform the design of a liquid-phase adsorption recovery process, where major drawbacks of the former downstream process could be overcomed.
3. Third, calculation of energy utilization is performed, considering the energy consumption of fluid pressurization (pumps and compressors) as well as heating and cooling requirements.
4. Eventually, the energy requirements of the process has been evaluated, comparing them with the energy stored within the produced n-butanol. The evaluated results will also be used to determine which are the target values of n-butanol concentration in the media and n-butanol productivity that will allow a viable continuous industrial process.
To perform all the calculations, a pilot-plant scale installation with 10 m3 photobioreactor total capacity is considered. Additional parameters like the photobioreactors required area have also been calculated, considering a maximum photobioreactor depth of 15 cm in order to avoid self-shading effects. .
Simulating n-Butanol Recovery Processes
Once identified the most relevant technologies we have designed a n-butanol recovery and purification process. Next step is simulating the the process in order to asses its overall productivity, energy and materials consumption. All of these aspects will eventually determine its sustainability and techno-economic viability, being critical for a succesful implementation of the technology.
Gaseous phase adsorption process
The most challenging aspect of this process relies on the low n-butanol volatility at normal cultivation temperatures. Then, our first goal was to evaluate which were the threshold values of n-butanol concentration-productivity pairs that will require realistic gas-stripping flows to remove the produced n-butanol within the culture.
Since in gas-phase adsorption operations, the fixed bed does not require a drying step, and the required air-flows are high, bed desorption can be performed using solely a hot air stream. This way, a wide-diameter single-bed could be used for the operation instead of a small fixed-beds system, simplifying the required equipment costs.
After performing the simulation, we have identified how for these processes, the most critical factor is the n-butanol concentration within the media. The highest n-butanol concentration the lower will be the required flow rate of aeration. At constant n-butanol concentrations, productivity increases the required amount of aeration, since more product has to be removed. Likewise, we have calculated the required amount of adsorbent for product recovery during 12h adsorption time. To do so, we consider that stripping air enters the fixed-bed at culture temperature: 38ºC. Then, n-butanol vapour pressure will be 0,022 atm. Likewise, using the adsorption isotherm data, it corresponds to a capacity of 57,85 g of n-butanol per kg of adsorbent.
Table 1. Design Calculations for the Stripping-Adsorption Operation
It is observable that the required aeration rates are very high. After this simulation,only those cases with an aeration flow rate that will not compromise cells or photobiocatalyst viability has been shown and selected for the design of the adsorption fixed bed (Considering this flow rate threshold as less than 8 Lair*min-1*Lreactor-1).
Because of the high values observed in stripping air-flows, we performed a preliminary calculation of the energy required solely for the stripping-adsorption stage. Only the air-compression energy requirements has been considered. To do so, adsorbent mass has been used to estimate the length of a column with 0,25 m diameter. Then, the hydrostatic pressure loss and Ergun equation was used to estimate the pressure drop across the bed for the required air flow. In the case of culture aeration, 2 m height aeration path is considered. Eventually an isentropic air compression corrected with an efficiency factor of 0.8 is considered for the compression energy estimation.
Table 2. Stripping-Adsorption process viability assessment
In view of the results, clear conclusions arise.
The calculated air stripping flows are excessively high when compared with the required mass of adsorbent to use within the fixed bed. This makes the design unfeasible for most of the proposed scenarios, since the adsorption equipment must be oversized in order to cope with such big stripping flows. This can be observed in the results considering the high gas speeds through the bed, even when a high diameter such as 0.25 m is considered, the gas velocities leading to an unacceptable pressure drop across the fixed bed system.
Taking all of these results together it becomes clear that the required stripping flow is high enough for leading to unrealistic designs. Only the scenario where a high n-butanol concentration in the culture and a low productivity are combined lead to a realistic result, where the energy consumption greatly exceeds that which the product can provide.
Although this energy can be harvested from the environment using renewable sources like solar or wind, it is still an unacceptable amount required to recover a small quantity of the desired product. In addition, the required stripping rates greatly exceed the aeration culture requirements.
Because all of the above, single-stage gas stripping recovery of butanol would not be viable within the continuous production system. Even considering other strategies such as the batch cultivation followed by an increase in photobioreactor temperature, or the utilization of mild vacuum, the required air stripping flow rate is still higher than the allowed for process viability.
The low volatility of n-butanol and the relatively low concentrations that engineered cyanobacteria have demonstrated to tolerate, makes the single-stage stripping-adsorption process not viable for n-butanol recovery. However, for other compounds with higher volatility, like small-chain hydrocarbons or terpenes, could become a really promising alternative, since most of the product could be easily recovered from the culture media, avoiding its loss during culture aeration.
However, this simulation allowed us to evaluated the expected product losses from culture aeration concluding that they can be disregarded during normal operation of photobioreactors.
Liquid phase adsorption process
After the negative results for the gas stripping-adsorption process, we decided to take a different approach: liquid adsorption.
In this process, biomass or photobiocatalyst is separated from the culture media. Depending on the material used within the photobioreactor, centrifugation, filtration or sedimentation operations could be used.
The photobiocatalytic materials we have created can easily sediment in relatively short periods of time, in addition they can also be filtered without the requirement of energy-intensive ultrafiltration operations. Then their utilization could potentially cut down with the energy consumption of the process.
In order to simulate the liquid adsorption we decided to base our adsorption simulation in an already studied setup for n-butanol adsorption-desorption, maintaining those parameters that allow to scale-up the design. Within adsorption operations, keeping the geometrical design of the adsorption equipment allows the scale up of the system.
Considering these and the formerly mentioned aspects, a modular system composed of multiple adsorption columns of fixed dimensions will be utilized. Each column will have 0.15 m length and 0.0105m inner diameter. This design allows having moderate pressure drops across the column while column heating for adsorbent drying can be easily achieved. Each one of these fixed-bed modules has capacity for 3.88 g of adsorbent.
The amount of required adsorbent will determine the system dimensions in terms of the number of parallel modules to utilize. Each one of these modules will be equipped with an electric heating element for the thermal desorption stage. The amount of required adsorbent will determine the system dimensions in terms of the number of parallel modules to utilize.
To perform the simulation of the process, firstly we have calculated the flow rate of the liquid stream continuously extracted from the cultivation system. To do so a mass-balance under stationary conditions is applied, considering the n-butanol concentration and productivity values of each scenario. This flow-rate will be used for the design of the liquid adsorption stage.
Available literature has shown that the optimal drying conditions of the fixed bed are 120 min at 50ºC with an air flow of 12 L*h-1. The optimal thermal desorption conditions maintain the air flow of 12 L*h-1 while the temperature is increased to 150ºC for 34 minutes. With these operation conditions the product can be efficiently absorbed. Then a total desorption time of 180 minutes is considered.
Once total desorption time, product concentration in the feed stream, adsorbent load and flow are known, the required number of parallel modules can be easily calculated. For that, adsorption isotherm is used for calculation of capacity during adsorption conditions (38ºC, CBOH of the culture).
Table 3. Liquid Adsorption Design Calculations
Attending the results shown by the simulation it is observable how the size of the adsorption unit is only dependent on the productivity with the system. However, higher the butanol concentration kept constant in the culture media, the smaller the flow will be required to process for the adsorption operation.
After defining the adsorption unit, only the product condensation step needs to be calculated. To do so, a condenser operating at 10 ºC will be considered. This condenser could operate in two different modes.
First, only acting when the drying process has ended and fast n-butanol desorption has begun. In this case part of the adsorbed n-butanol will be lost, since during column drying a small amount of n-butanol is also desorbed. This product loss has been observed to be 4.7% of the initially fed to the column. In this mode, the energy consumption of the condenser can be calculated using the vaporization enthalpy of n-butanol and the total amount of recoverable product.
Second, acting continuously during drying and desorption steps. During drying, a water n-butanol liquid mixture is obtained that could be recirculated to the fixed-bed system operating in the adsorption stage at that time. Then, once desorption has started a purified n-butanol stream will be obtained. In this mode, the energy consumption of the condenser has to consider either the n-butanol vaporization enthalpy, and the partial condensation of n-butanol - water mixture that is recovered during column drying.
Since the operation of the condenser for long periods of time could increase the energy consumption, we select the first possibility for the sake of design simplicity. However it is important to notice that if the condenser energy requirements can be supplied with residual heat derived from other neighbouring processes it will be desirable to implement the second strategy.
After the design of the process for the different proposed scenarios, the energy consumption assessment is performed. To do so, energy requirements for culture aeration, liquid pumping, centrifugation, gas compression (for drying and desorption steps), column heating and condenser energy consumptions is calculated.
Culture aeration has been considered to be as 0.03 vvm, it is that for a 10 m3 total reactor volume, an aeration flow of 18 m3*h-1 would be required. Energy consumption is considered the aeration consumption per every adsorption cycle, since product will be recovered in a continuous fashion. Aeration consumption can be calculated considering the pressure loss of gas across the liquid column of the PBR and corrected by an excess factor for taking into account pressure drop in pipes.
Pumping energy considers the energy consumption of pumps required to extract culture media and refresh photobioreactors with fresh growth media. Required pumping pressure is calculated considering a +2m level difference between the photobioreactor inlet/outlet, the required fluid velocity for the design flow (Bernouilli’s principle). An excess of 10% to account for friction loss in the pipes has been also considered.. Likewise, during liquid adsorption, the pressure drop of the bed is calculated using Ergun’s equation.
Gas compression requirements consider the energy required to drive the gas streams through the fixed-bed columns during drying and desorption stages. To do so, pressure drop through the column has been estimated with Ergun’s equation. Pressure drop in the condenser has been considered as a standard value of 30 kPa, while the necessary pressure to drive the air at the desired flow is calculated considering the required air kinetic energy.
Heating requirements have been calculated considering the specific heat of each column module (adsorbent and casing materials) as well as the heat removed by the air stream passing through (considering perfect heat transport from the bed to the air in both phases, drying and desorption).
Cooling energy in the condenser has been estimated considering the n-butanol vaporization enthalpy and a condenser energy efficiency of 0,33 in accordance with available literature.
In the case of centrifugation a reference value of 1.2 kWh per m3 of processed fluid has been used. Despite current industrial high speed centrifuges can reach efficiencies of up to 0,6 kWh per m3 of liquid to clarify, we have decided to take a conservative approach during energy calculations.
More details about the calculation of each of these parameters can be found in the Modelling page.
Eventually we have compared in each case the Energy Return on Investment (EROI) for the production. This parameter measures the amount of energy available in the product with respect to the energy required for its manufacturing. Then processes with an EROI higher than 1 are energetically sustainable. Results are summarized in the table below.
Table 4. Liquid adsorption process. Energy Assessment
Observing the design results, it is evident that for all the proposed scenarios the process will be viable, since energy consumption is several times lower than that stored in the manufactured product.
A trend can also be observed. In contrast with how engineered cells performed, the higher n-butanol concentration is held in the culture, the more efficient the downstream process becomes. This is logical since smaller culture volumes need to be processed and then less energy is required to purify the same amount of the product. It is also important to remark that this energy calculation considers only the main energy inputs of a well defined process. Additional energy requirements during the preparation of culture growth media or the energy requirement for plant start-up after a technical shutdown. However the overall numbers allow us to consider the process viable even accounting for an extra energy consumption.
Likewise, energy consumption for cell separation does not account for a big part of the overall process requirements. This way, main advantages derived from the utilization of photobiocatalytic materials will rely on the observed enhancement in productivity as well as reducing the required photobioreactor sizes, making the total plant investment costs lower.
Then... the process has been outlined, defining all of its required operation parameters and defining the main features for further equipment sizing. Table 5 summarizes all the relevant process information considering an installation with 10 m3 of Photobioreactors volume. For the sake of simplicity, only the scenario corresponding with the productivity of an already existing n-butanol producing strain has been selected.
Table 5. Downstream processing summary.
Design & Simulation Results
After evauating two potential downstream schemes for n-butanol recovery and designing the processes in deeper detail, we have calculated the threshold value of required productivity for an EROI equal to 1. EROI = 1 is the frontier for a energetically sustainable process. To do asses the treshold scenarios of a energetically viable process, we have considered four different product concentrations in the culture media.
Table 6. EROI Sensitivity analysis in terms of concentration and strain productivity.
The targets for biological engineering
After evaluating the productivity threshold values, we can observe that they remain almost constant for any possible concentration in the culture. Since adsorption is a highly selective process, a high purity product can be obtained without high relevance in the feed concentration, and these results are a confirmation of that.
Then, for biological engineering purposes it is notable that the importance of the tolerated product concentration is secondary for industrially viable organisms. On the contrary an engineered strain should produce at the highest possible rate and product toxicity can be controlled by the dilution rate of a continuous cultivation process. It is important to note that product tolerance will define the final size required for an installation, being a critical factor in the investment cost of the technology.
The minimum viable product generation should be of 1.38 g*h-1*m2. A value which is currently under the performance of some already existing engineered strains!
However, the satisfactory implementation of an engineered strain will also require some additional features. An engineered cyanobacteria that can be used in the core of a biomanufacturing process should also be adapted to the harsh industrial environment. The strain must be easy to cultivate, feature a reduced contamination risk, and withstand a wide range of cultivation temperatures without severely compromising its productivity.
To sum up!
We have successfully designed a bioprocess following the photobiocatalytic approach for engineered cyanobacteria utilization. The results have demonstrated its technological feasibility and the potential of phototrophic microorganisms for industrial biomanufacturing.
In addition, we have identified that for development of n-butanol overproducing strains, critical features are its productivity and robustness for industrial application. Tolerance to the intrinsic product toxicity is also important in order to scale-up the process at the industrial scale, the higher tolerant strains we have, the smaller size for an installation will be required, since smaller cultures could provide a higher amount of product. We have surprisingly discovered that this value is under the current results for engineered cyanobacteria.
We have made some numbers and defined with detail the bioprocess functioning and operating conditions. However, to settle some example words, we can take the 10 m3 photobioreactors production plant we have been evaluating.
Within a production plant with 10 m3 culture capacity, 10 m2 of surface will be used for light capture. Every day up to 4 L of butanol could be recovered. This butanol will provide 7 times the energy required for its production.
Of course, there is still a long road to pave, especially in terms of engineering better producing strains, capable of increasing the amount of product recovered per volume of photobioreactors. However the overall process is viable and while better strains and new smart engineering solutions for industrial cultivation of phototrophs appear, the lower the investment requirements for its large-scale industrial implementation.
Further Developments: Solid Photobiocatalyst
During our project we have successfully synthesized several biohybrid nanomaterials, among all of them, yolk-shell self-assembled nano-structures and silica gel encapsulation of cyanobacterial cells has arisen as the most promising materials for encapsulated cell photobiocatalysts development.
The next step towards a large-scale implementation of photosynthetic biomanufacturing is the development of cell containing bio-hybrid materials, capable to act as industrial catalysts. The creation of this materials will overcome the main limitations of free cells cultures, reduce production costs and expand the technological possibilities for photosynthetic biomanufacturing applications.
Despite being unable to successfully generate a n-butanol producing cyanobacteria in the time span of an iGEM project (mainly because of time and cloning issues), we have tested the performance of the designed photobiocatalytic materials employing an engineered cyanobacteria capable of sacarose-secretion. Considering a sucrose secreting engineered cyanobacteria a good model for our direct-biosynthesis photobiocatalyst approach, we have performed real-case production experiments. To know more about the importance of photobiocatalysis visit our Implementation and Encapsulation pages.
Seader, J. D., Henley, E. J., & Roper, D. Keith. (2011). Separation Process Principles with Applications using Process Simulators.
Kuan, D., Duff, S., Posarac, D., & Bi, X. (2015). Growth optimization of Synechococcus elongatus PCC7942 in lab flasks and a 2-D photobioreactor. The Canadian Journal of Chemical Engineering, 93(4), 640–647. https://doi.org/10.1002/CJCE.22154
Myers, J. A., Curtis, B. S., & Curtis, W. R. (2013). Improving accuracy of cell and chromophore concentration measurements using optical density. BMC Biophysics 2013 6:1, 6(1), 1–16. https://doi.org/10.1186/2046-1682-6-4
Cousin Saint Remi, J., Baron, G., Denayer, J., 2012. Adsorptive separations for the recovery and purification of biobutanol. Adsorpt. 2012 185 18, 367–373. https://doi.org/10.1007/S10450-012-9415-1
Incropera, F.P., DeWitt, D.P., Bergman, T.L., Lavine, A.S., 2017. Incropera’s principles of heat and mass transfer. Wiley 1000.
García Rodríguez, Á., García Rodríguez, Á., 2017. Recuperación de biobutanol para la producción de combustibles mediante ciclos de adsorción-desorción.
Magdaong, J.B., Ubando, A.T., Culaba, A.B., Chang, J.S., Chen, W.H., 2019. Effect of aeration rate and light cycle on the growth characteristics of Chlorella sorokiniana in a photobioreactor. IOP Conf. Ser. Earth Environ. Sci. 268, 012112. https://doi.org/10.1088/1755-1315/268/1/012112
García Rodríguez, Á., García Rodríguez, Á., 2017. Recuperación de biobutanol para la producción de combustibles mediante ciclos de adsorción-desorción.
Consider that the packed column is originally at a solute concentration cA (in the interparticular gas phase) [WWW Document], n.d. URL https://web.fe.up.pt/~lepae/simsorb/SMT.html (accessed 10.14.21).
Kommareddy, A.R., Anderson, G.A., Gent, S.P., Bari, G.S., 2013. The Impact of Air Flow Rate on Photobioreactor Sparger/Diffuser Bubble Size(s) and Distribution. Am. Soc. Agric. Biol. Eng. Annu. Int. Meet. 2013, ASABE 2013 6, 1-. https://doi.org/10.13031/AIM.20131620764
Seader, J.D., Henley, E.J., Roper, D.K., 2011. Separation Process Principles with Applications using Process Simulators. 849.
Myers, J.A., Curtis, B.S., Curtis, W.R., 2013. Improving accuracy of cell and chromophore concentration measurements using optical density. BMC Biophys. 2013 61 6, 1–16. https://doi.org/10.1186/2046-1682-6-4
Kuan, D., Duff, S., Posarac, D., Bi, X., 2015. Growth optimization of Synechococcus elongatus PCC7942 in lab flasks and a 2-D photobioreactor. Can. J. Chem. Eng. 93, 640–647. https://doi.org/10.1002/CJCE.22154
Szepessy, S., Thorwid, P., 2018. Low Energy Consumption of High-Speed Centrifuges. Chem. Eng. Technol. 41, 2375–2384. https://doi.org/10.1002/CEAT.201800292
S, X., Q, Z., X, W., C, Y., W, C., 2013. A novel photobioreactor structure using optical fibers as inner light source to fulfill flashing light effects of microalgae. Bioresour. Technol. 138, 141–147. https://doi.org/10.1016/J.BIORTECH.2013.03.156
Do Gasoline Based Cars Really Use More Electricity than Electric Vehicles Do? | Council on Foreign Relations [WWW Document], n.d. URL https://www.cfr.org/blog/do-gasoline-based-cars-really-use-more-electricity-electric-vehicles-do (accessed 10.9.21).
Vapor Pressure Calculation by Antoine Equation (1-Butanol) [WWW Document], n.d. URL http://ddbonline.ddbst.de/AntoineCalculation/AntoineCalculationCGI.exe (accessed 10.9.21).
(44) So: Exactly How Much Electricity Does it Take To Produce A Gallon of Gasoline? | LinkedIn [WWW Document], n.d. URL https://www.linkedin.com/pulse/so-exactly-how-much-electricity-does-take-produce-gallon-paul-martin/ (accessed 10.9.21).
Cousin Saint Remi, J., Baron, G., Denayer, J., 2012. Adsorptive separations for the recovery and purification of biobutanol. Adsorpt. 2012 185 18, 367–373. https://doi.org/10.1007/S10450-012-9415-1
Martin-Calvo, A., Perre, S. Van der, Claessens, B., Calero, S., Denayer, J.F.M., 2018. Unravelling the influence of carbon dioxide on the adsorptive recovery of butanol from fermentation broth using ITQ-29 and ZIF-8. Phys. Chem. Chem. Phys. 20, 9957–9964. https://doi.org/10.1039/C8CP01034J
Cousin Saint Remi, J., Rémy, T., Van Hunskerken, V., van de Perre, S., Duerinck, T., Maes, M., De Vos, D., Gobechiya, E., Kirschhock, C.E.A., Baron, G. V., Denayer, J.F.M., 2011. Biobutanol Separation with the Metal–Organic Framework ZIF-8. ChemSusChem 4, 1074–1077. https://doi.org/10.1002/CSSC.201100261
Rafa, N., Ahmed, S.F., Badruddin, I.A., Mofijur, M., Kamangar, S., 2021. Strategies to Produce Cost-Effective Third-Generation Biofuel From Microalgae. Front. Energy Res. 9. https://doi.org/10.3389/FENRG.2021.749968/FULL