Research Article
Hydrothermal liquefaction of microalgae after different pre-treatments
Energy Exploration & Exploitation 0(0) 1–10 ! The Author(s) 2018 DOI: 10.1177/0144598718777107 journals.sagepub.com/home/eea
Mikhail S Vlaskin1 , Anatoly V Grigorenko1, Nadezhda I Chernova1,2 and Sophia V. Kiseleva1,2
Abstract Hydrothermal liquefaction of different microalgae samples (Arthrospira platensis cultivated by our research group) – fresh (directly after harvesting), dried and frozen – have been performed. In hydrothermal liquefaction process, the samples were heated up to 300 C for 30 min and kept at a constant temperature for 60 min. Then dichloromethane was added to the samples to extract the oil fraction. The products obtained after aqueous and dichloromethane solutions evaporation are referred to as water soluble organics and bio-oil correspondingly. The experiments on hydrothermal liquefaction of microalgae pre-treated in different ways were conducted for three independent harvest samples. The average values of bio-oil yield in the experiments with fresh, dried and frozen microalgae were equal to 44.07%, 39.97% and 39.65%, respectively. The average yields of water soluble organics were equal to 19.34%, 29.00% and 21.43% respectively. In all the experiments, the highest yield of bio-oil was reached for fresh microalgae. From this point of view, direct hydrothermal liquefaction processing of fresh microalgae seems to be more preferable that further enhances the advantage of hydrothermal liquefaction in comparison with other biomass-to-biofuel conversion methods. Keywords biofuel, bio-oil, microalgae, hydrothermal liquefaction, preliminary treatment
1 2
Joint Institute for High Temperatures of Russian Academy of Sciences, Moscow, Russia Renewable Energy Sources Laboratory, Geographical Faculty, Lomonosov Moscow State University, Moscow, Russia
Corresponding author: Mikhail S Vlaskin, Joint Institute for High Temperatures, 13/2 Izhorskaya St, Moscow, 125412, Russia. Email:
[email protected] Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).
2
Energy Exploration & Exploitation 0(0)
Introduction The growth of world energy consumption and the resulting increase in greenhouse gases emissions into the atmosphere stimulate the use of renewable energy sources, including bioenergy. One of the most promising sources of renewable biofuel is microalgae. Fuel derived from microalgae is referred to a separate group of so-called “third-generation biofuels” because this type of biomass is not a food raw material and it is grown for energy purposes in the areas unsuitable for crop production. Microalgae are microscopic organisms efficiently converting the energy of sunlight into biomass. The growth of microalgae is associated with the capturing of carbon dioxide. The productivity of microalgae biomass and oil (lipids) is 10 times higher than that of terrestrial biomass (Chernova et al., 2014). In recent decades, the studies devoted to biofuel production from microalgae were focused mainly on cultivation methods. In this context, the main tasks were first of all the increasing of productivity and lipid content in biomass by strains selection (Hu et al., 2016; Neofotis et al., 2016; Taleb et al., 2016), cultivation conditions optimization (Yee, 2015; Zheng et al., 2012) and new strains creation by genetic engineering (Klanchui et al., 2012). However, little attention was paid to the problem of biomass to biofuels conversion. A rational solution to this problem would improve the energy efficiency of the process of biofuel production from microalgae. The traditional conversion method usually includes drying, solvent extraction of lipids and transesterification with the production of fatty acid methyl esters composing biodiesel fuel (Salam et al., 2016). The obvious disadvantages of the biodiesel production method are high energy costs and the use of dangerous organic solvents (such as methanol). Moreover, in the case of biodiesel production, only the lipid fraction is processed, while the rest of biomass including proteins and carbohydrates is not involved in the biofuel production. At the same time, it is known that lipid-rich strains have relatively low biomass productivity (Lopez Barreiro et al., 2013; Rodolfi et al., 2009). The problem of biofuel production arises primarily from high content of moisture in microalgae biomass after cultivation stage (80–90% by weight). Hydrothermal technologies seem to be more favorable for the processing of wet biomass into biofuels. Hydrothermal liquefaction (HTL) with the production of crude bio-oil (BO) as the end product is of most interest (Tian et al., 2017; Vlaskin et al., 2017). During HTL, the biomass is thermally treated under humid conditions at temperatures up to 370 C and pressures up to 25 MPa. During this treatment, the biomass components undergo hydrolysis and pyrolysis reactions thus forming a number of liquid hydrocarbons (both soluble and insoluble in water) as well as gaseous and solid reaction products. HTL products are BO, aqueous solution with water soluble organics (WSO), solid residue and gas. The target product of HTL is BO, liquid hydrocarbons separated from the solid phase (residue) and extracted from an aqueous solution. One of the advantages of HTL is that the BO produced is composed not only of lipids but also of carbohydrates and proteins that increases the overall yield of the product (Elliott, 2016). The additional advantages of HTL are relatively low temperatures of the process and the possibility of using a single-stage continuous process. It is important that feedstock predrying is unnecessary in this case. Biomass can be fed into a HTL reactor in humid state directly after harvesting. However, most of studies devoted to HTL of microalgae were carried out using dried samples of microalgae. Therefore, a comparison of the efficiencies of fresh and dried microalgae with regard to HTL is interesting. During the drying of microalgae, the change of its chemical composition may take place and the aqueous solution
Vlaskin et al.
3
coming with fresh microalgae also can exert influence on the reactions occurred during HTL. These effects can change the overall yield and the BO composition. The main objective of the present paper is to study the yield and composition of BO produced by HTL using different microalgae samples: fresh (directly after harvesting), dried, frozen and dried and mixed with olive oil (25 wt. %). The sample with the addition of olive oil was used to evaluate the influence of increased content of lipids in microalgae biomass on the yield of BO and WSO. Olive oil addition is expected to increase the BO yield.
Materials and methods Biomass source HTL experiments were carried out with cyanobacterium Arthrospira platensis (Chernova and Kiseleva, 2017). Arthrospira platensis is a well-known culture that is used in the production of many commercial products including healthy food supplements, food for animals, cosmetics and pharmaceutical products. It is grown usually in an open manner on a large scale. A. platensis is classified as a cyanobacteria by the International Code of Bacterial Nomenclature (ICNB), although according to the International Code of Nomenclature for Algae, Fungi and Plants (Melbourne Codex 2012), it is classified as a blue-green microalgae. When it is used as biomass for biofuel production, it is often considered as a microalgae (Gouveia and Oliveira, 2009; Kostyukevich et al., 2017; Raikova et al., 2016). We used Arthrospira platensis rsemsu 1/02-P strain with straight trichomes formed due to natural morphological variability during prolonged cultivation for 20 years under laboratory conditions. Biomass was obtained by semi-continuous cultivation in an open pond with a volume of 500 l and illumination of 55 5 lE/(m2 s) at steady light conditions and a temperature equal to T ¼ 21 C. The pond is equipped by the means for near-surface mixing. A. platensis was cultivated using Zarrouk’s medium: NaHCO3 – 16.8 g/l, KNO3 – 3.0 g/l, K2432O – 0.66 g/l, K2SO4 – 1.0 g/l, MgSO47H2O – 0.2 g/l, NaCl – 1.0 g/l, CaCl2 – 0.04 g/l, FeSO47H2O – 0.018 g/l, EDTA – 0.08 g/l, Zarrouk’s trace metal solution – 1.0 ml/l, (per 1 l of distilled water) (Zarrouk, 1966). Molecular genetics identification and phylogenetic analysis with a high bootstrap support level based on the nucleotide sequence of the 16S rRNA and ITS genes showed that this culture has a 100% similarity with A. platensis strains from the NCBI strain bank, i.e. it belongs to the A. platensis species. The chemical composition (C, H, N and S content) of A. platensis was determined using a Thermo Scientific Flash 2000 HT analyzer. The content of ash in A. platensis was determined by pyrolysis at 800 C and mass measurements using an analytical balance Sartorius Cubis MSA324S. The results of this analysis (wt. % daf basis) and ash content determination for A. platensis are presented in Table 1. Biochemical composition of A. platensis is presented in Table 2. Table 1. Chemical composition and ash content of Arthrospira platensis, wt. %. Sample
C
H
N
S
(by
A. platensis
45.680.16
6.930.15
10.080.08
0.9700.075
36.340.47
difference)
Ash,% 5.0
4
Energy Exploration & Exploitation 0(0) Table 2. Biochemical composition of Arthrospira platensis, wt. %. Sample
Proteins
Lipids
Carbohydrates
A. platensis
60.7
12.1
7.1
Figure 1. Scheme of laboratory setup. 1, electric heater; 2, sand bath; 3 reactor-autoclave; 4, temperature sensor; 5, thermocouples.
Biomass pretreatment Harvested biomass of A. platensis was used to prepare four samples pre-treated in different ways: fresh (directly after harvesting), dried, frozen and dried and mixed with olive oil (25 wt. %). The sample with the addition of olive oil was used to evaluate the influence of increased content of lipids in microalgae biomass on the yield of BO and WSO. Olive oil was added to the dried microalgae. The contents of C, H and N in the olive oil were 83.64%, 5.88% and 0.18%, respectively. Harvested biomass was passed through a 20 lm mesh sieve. The drying was carried out in the Binder drying oven at 105 C. The moisture of fresh microalgae was determined after the drying. The freezing was carried out in the refrigerator in a closed and sealed tank at 18 C. HTL of fresh microalgae was carried out directly after harvesting, and HTL of frozen microalgae was carried out directly after defreezing. All HTL experiments were conducted during 24 h after harvesting.
Hydrothermal liquefaction HTL experiments were carried out on a laboratory-scale plant, the scheme of which is shown in Figure 1. The reactors represent small autoclaves with a volume of 25 ml. The reactors were placed into the sand bath heated by the electric heater. In the experiments with fresh and frozen biomass, the reactors were loaded with 15 g of aqueous suspension of microalgae. In the experiments with dried microalgae, the reactor also was loaded with 15 g of suspension that contained biomass (1.836 g) and distilled water (13.3 ml). The amount of distilled water corresponding to the moisture content was determined by the drying of fresh microalgae. The same amount of distilled water (13.3 ml) was poured into the rector in the experiment with microalgae dried and mixed with olive oil (25 wt. %).
Vlaskin et al.
5
Figure 2. Test tubes with condensed products of HTL of microalgae and the solvent. Liquids with different density and color are observed (aqueous solution is on top and solvent solution bottom).
During HTL, the reactors were heated up to 300 C. The duration of this heating process was about 30 min; the exposure time at 300 C was 60 min. The pressure inside the reactors was not measured, but as the 25 ml reactors were filled with aqueous suspension with water content of about 13.3 ml, the partial pressure of water reached the saturation pressure – 8.6 MPa (at 300 C). The total pressure inside the reactor should be slightly higher than 8.6 MPa because of biomass gasification. After the experiment, the electric heater was turned off and the reactors were cooled to room temperature. Then, the reactors were opened and about 10 ml of solvent (dichloromethane) was poured into each reactor. After that the reactors were closed and kept in a vertical position with periodical 180 rotation. This solvent extraction procedure facilitating the contact of solvent with the whole internal surface of the reactor lasted 10 h. After that, the reactors were opened and liquid and solid products were removed and placed in plastic test tubes (Figure 2). In these test tubes, the products of HTL (solid residue, aqueous solution with WSO and dichloromethane solution with BO) were separated by density. After sedimentation of solid residue as well as aqueous and dichloromethane solution separation, two layers with different colors were observed in the plastic test tubes (Figure 2). The upper one is aqueous solution; the bottom layer represents dichloromethane solution with solid residue on the bottom of the test tube. Then two liquid samples (each about 6 ml) were taken by a syringe from the layer with dichloromethane solution and from the layer with aqueous solution, and placed into the corresponding Petri dishes. The liquid sample from the layer with dichloromethane solution was taken from its upper part to prevent the
6
Energy Exploration & Exploitation 0(0)
capture of solid residue. The Petri dishes were then placed in the Binder drying oven. The aqueous solution was evaporated at a temperature of 70 C and the solution of dichloromethane – at a temperature of 38 C. The drying in the oven was stopped after the weight of Petri dishes was stabilized. The mass was measured using the analytical balance Sartorius Cubis MSA324S. From the mass of the products obtained in the Petri dishes, the yields of BO (after dichloromethane evaporation) and WSO (after water evaporation) were derived. Solid residues remained in the test tubes were filtered, dried and weighed. From the mass of the dried solid residue, its HTL yield was evaluated. Several series of experiments under the equal (described above) conditions were carried out for each of the microalgae samples pre-treated in different ways. The study of HTL of microalgae was conducted for three independent harvest samples. The chemical composition of BO and WSO was analyzed using a Thermo Scientific Flash 2000 HT analyzer adopted for liquid samples. C, H, N and S contents were determined for liquid samples of BO and WSO (dry ash free basis) with oxygen content determined by calculating the difference. The test procedure was repeated five times for each sample.
Results and discussion The results of the experiments carried out with one of the harvest samples (first harvest) are shown in Table 3. Before these HTL experiments, the moisture contents in fresh and frozen suspensions of microalgae were determined. The masses of the residues after the drying of 30.2 and 15.26 g of fresh microalgae were 3.65 and 1.84 g, respectively. The masses of the residues after the drying of 30 and 15.5 g of frozen microalgae were 3.61 and 1.86 g, respectively. The moisture of the suspension with fresh microalgae was about 87.94%, and the moisture of the suspension with frozen microalgae was about 88%. The contents of moisture in the suspensions with dried microalgae and microalgae dried and mixed with olive oil were 87.87% and 87.79%, respectively (suspensions contained 13.3 g of distilled water). The yields of BO obtained in the experiments with fresh, dried, frozen microalgae and microalgae dried and mixed with olive oil (from first harvest) were equal to 43.46%, 40.68%, 42.88% and 43.66%, respectively. The yields of WSO in these experiments were equal to 15.19%, 23.48%, 18.8% and 15.5%, respectively. The relative error of the yields measurements was estimated to be less than 7%. It can be seen that the yield of BO in the case of fresh microalgae exceeds the BO yields for frozen and dried microalgae. At the same time, the yield of WSO in the case of fresh microalgae is the lowest one. The highest yield of WSO was reached in the experiments with dried microalgae. This result could be explained by a partial destruction of proteins during the drying. During drying and freezing, the cells of microalgae undergo dehydration that is likely to increase the stability of organic matter before reactions of hydrolysis and pyrolysis typical for HTL occur. On the contrary, in case of fresh microalgae, the moisture stays inside a cell and it probably increases the contact surface between organic matter and water. In the experiments with the mixture of microalgae and olive oil, an increase in BO and decrease in WSO yields were observed. In case of the mixture of dried microalgae (75 wt. %) and olive oil (25 wt. %), BO yield was the highest. It is about 3% higher than that for 100% dried microalgae. These results were expected and they are explained by the fact that the HTL product of olive oil contributes mostly to oil fraction soluble in dichloromethane.
Vlaskin et al.
7
Table 3. Experimental data for one of the harvest samples: HTL of different microalgae samples: fresh (directly after harvesting), dried, frozen and dried and mixed with olive oil. Dried and mixed with olive oil (25 wt. %)
Biomass sample
Dried
Fresh
Frozen
Masses of components loaded into the reactor, g Moisture content, % Volume of dichloromethane, ml Mass of sample with dichloromethane solution before evaporation, g Mass of product (BO) after dichloromethane evaporation, g BO yield, g or % Mass of sample with aqueous solution before evaporation, g Mass of product (WSO) after water evaporation, g WSO yield, g or % BOþWSO yield, % Solid residue yield, g or %
1.836 (dry biomass) þ 13.3 (distilled water) 87.87 10
15.26 (suspension)
15.5 (suspension)
87.94 10
88 10
1.85 (dry biomass) þ 13.3 (distilled water) 87.79 10
7.04
5.58
6.04
6.88
0.364
0.31
0.33
0.383
0.75 or 40.68 8.92
0.80 or 43.48 8.82
0.80 or 42.88 9.6
0.81 or 43.66 9
0.28
0.18
0.24
0.19
0.43 or 23.48 64.16 0.50 or 27.25
0.28 or 15.19 58.67 0.40 or 22.10
0.35 or 18.80 61.68 0.43 or 23.05
0.29 or 15.50 59.16 0.39 or 21.20
Note. BO: bio-oil; WSO: water soluble organics.
It shows that more bio-oil will be produced by HTL for microalgae with a higher content of lipids. The identical experiments with fresh, dried and frozen microalgae were repeated for three independent harvest samples of A. platensis. The results of these three series of experiments are shown in Table 4. In all the experiments, the highest BO yield was obtained for fresh microalgae while the highest yield of WSO was achieved for dried microalgae. Table 4 shows also the average values of BO and WSO yields. The average value of BO yield in case of fresh microalgae turned out to be 4% higher than that for dried and frozen microalgae. The highest average value of WSO yield is 29% for dried microalgae. The average values of WSO yield for fresh and frozen microalgae are quite low (19.34% and 21.43%, respectively). Table 5 shows chemical composition of BO and WSO samples produced by HTL of fresh, dried and frozen microalgae (obtained in the first series of experiments with first harvest). It can be seen that BO in all cases is enriched in carbon and depleted in oxygen relative to the original A. platensis. The content of carbon in all BO samples is varied in the range of 74.1– 75.1 wt. %. The content of carbon in WSO is varied from 35 to 41 wt. %. The lowest content of carbon is observed in the WSO obtained from fresh microalgae. The content of nitrogen in BO is about 6.3 wt. % and it is usually less than that present in WSO. However,
8
Energy Exploration & Exploitation 0(0)
Table 4. Experimental results for three independent harvest samples: HTL of different microalgae samples: fresh (directly after harvesting), dried and frozen. First harvest
Second harvest
Third harvest
Average values
Biomass samples
BO yield, %
WSO yield, %
BO yield, %
WSO yield, %
BO yield, %
WSO yield, %
BO yield, %
WSO yield, %
Dried Fresh Frozen
40.68 43.48 42.88
23.48 15.19 18.80
38.71 44.55 36.98
32.69 26.83 23.50
40.53 44.17 39.09
30.84 16.00 21.99
39.97 44.07 39.65
29.00 19.34 21.43
Note. BO: bio-oil; WSO: water soluble organics.
Table 5. Chemical composition of BO and WSO produced by HTL of fresh, dried and frozen microalgae (obtained in the first harvest). Content, wt. % Pre-treatment
Sample
C
H
N
S
O
Dried
WSO BO WSO BO WSO BO
38.39 0.06 74.43 0.32 35.27 0.47 74.14 1.74 40.92 0.59 75.14 0.49
6.488 0.087 9.704 0.019 5.757 0.019 9.626 0.072 6.691 0.094 9.747 0.044
7.34 0.03 6.18 0.05 6.31 0.13 6.49 0.49 7.68 0.11 6.43 0.26
1.941 0.282 1.195 0.098 2.353 0.246 1.031 0.049 1.305 0.049 1.604 0.023
45.841 0.459 8.491 0.487 50.31 0.865 8.713 2.351 43.404 0.843 7.079 0.817
Fresh Frozen
Note. BO: bio-oil; WSO: water soluble organics.
these values are very close to each other for the samples obtained from fresh microalgae. The contents of nitrogen in the BO samples obtained from fresh and frozen microalgae are slightly higher than that obtained from dried microalgae. It can be explained by the loss of compounds containing nitrogen during the drying.
Conclusions HTL of different microalgae samples – fresh (directly after harvesting), dried and frozen – have been performed. The experiments with A. platensis cultivated by our research group demonstrated that the highest yield of BO is achieved for fresh microalgae. This same result was obtained three times for three independent harvest samples of A. platensis. The results obtained in the present work have once again demonstrated the convenience of HTL method for microalgae processing. At the same time, the chemical composition of the resulting product in the experiments with fresh, dried and frozen microalgae proved to be very close. The results of the present work show that direct HTL processing of fresh microalgae leads to both the reduction of energy consumption by eliminating the drying stage and increasing of BO yield in comparison with dried and frozen microalgae. Acknowledgements The authors are grateful for Leonid Dombrovsky for useful discussions of the work.
Vlaskin et al.
9
Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported through a grant of the Russian Science Foundation (project no. 17–19-01617).
ORCID iD Mikhail S Vlaskin
http://orcid.org/0000-0001-6549-9939
References Chernova NI and Kiseleva SV (2017) Microalgae biofuels: Induction of lipid synthesis for biodiesel production and biomass residues into hydrogen conversion. International Journal of Hydrogen Energy 42(5): 2861–2867. Chernova NI, Kiseleva SV and Popel’ OS (2014) Efficiency of the biodiesel production from microalgae. Thermal Engineering 61(6): 399–405. Elliott DC (2016) Review of recent reports on process technology for thermochemical conversion of whole algae to liquid fuels. Algal Research 13: 255–263. Gouveia L and Oliveira AC (2009) Microalgae as a raw material for biofuels production. Journal of Industrial Microbiology & Biotechnology 36(2): 269–274. Hu X, Zhou J, Liu G, et al. (2016) Selection of microalgae for high CO2 fixation efficiency and lipid accumulation from ten Chlorella strains using municipal wastewater. Journal of Environmental Sciences 46: 83–91. Klanchui A, Vorapreeda T, Vongsangnak W, et al. (2012) Systems biology and metabolic engineering of Arthrospira cell factories. Computational and Structural Biotechnology Journal 3(4): e201210015. Kostyukevich Y, Vlaskin M, Vladimirov G, et al. (2017) The investigation of the bio-oil produced by hydrothermal liquefaction of Spirulina platensis using ultrahigh resolution Fourier transform ion cyclotron resonance mass spectrometry. European Journal of Mass Spectrometry (Chichester, England) 23(2): 83–88. Lopez Barreiro D, Zamalloa C, Boon N, et al. (2013) Influence of strain-specific parameters on hydrothermal liquefaction of microalgae. Bioresource Technology 146: 463–471. Neofotis P, Huang A, Sury K, et al. (2016) Characterization and classification of highly productive microalgae strains discovered for biofuel and bioproduct generation. Algal Research 15: 164–178. Raikova S, Smith-Baedorf H, Bransgrove R, et al. (2016) Assessing hydrothermal liquefaction for the production of bio-oil and enhanced metal recovery from microalgae cultivated on acid mine drainage. Fuel Processing Technology 142: 219–227. Rodolfi L, Chini Zittelli G, Bassi N, et al. (2009) Microalgae for oil: Strain selection, induction of lipid synthesis and outdoor mass cultivation in a low-cost photobioreactor. Biotechnology and Bioengineering 102(1): 100–112. Salam KA, Velasquez-Orta SB and Harvey AP (2016) A sustainable integrated in situ transesterification of microalgae for biodiesel production and associated co-product - A review. Renewable and Sustainable Energy Reviews 65: 1179–1198. Taleb A, Kandilian R, Touchard R, et al. (2016) Screening of freshwater and seawater microalgae strains in fully controlled photobioreactors for biodiesel production. Bioresource Technology 218: 480–490.
10
Energy Exploration & Exploitation 0(0)
Tian C, Liu Z and Zhang Y (2017) Hydrothermal liquefaction (HTL): A promising pathway for biorefinery of algae. In: Gupta SK, Malik A and Bux F (eds) Algal biofuels: Recent advances and future prospects. Cham: Springer International Publishing, pp.361–391. Vlaskin MS, Chernova NI, Kiseleva SV, et al. (2017) Hydrothermal liquefaction of microalgae to produce biofuels: State of the art and future prospects. Thermal Engineering 64(9): 627–636. Yee W (2015) Feasibility of various carbon sources and plant materials in enhancing the growth and biomass productivity of the freshwater microalgae Monoraphidium griffithii NS16. Bioresource Technology 196: 1–8. Zarrouk C (1966) Contribution a L’etude D’ Une Cyanophycee. Influence de Diverse Facteursphysiques et Chimiques Sur La Croissance et la Photosynthese de Spirulina Maxima (Setch et Gardner) Geitler. University of Paris, France. Zheng Y, Chi Z, Lucker B, et al. (2012) Two-stage heterotrophic and phototrophic culture strategy for algal biomass and lipid production. Bioresource Technology 103(1): 484–488.
