THIRD GENERATION BIODIESEL PRODUCTION FROM MICROALGAE Phormidium autumnale

Abstract

The aim of this work was to evaluate third generation biodiesel production by the microalgae Phormidium autumnale using sucrose as exogenous carbon source. The study focused on optimization of the different C/N ratios and on the analysis of biofuel quality. The results indicate that a C/N ratio of 40 improved the performance of the system, reaching single-cell oil productivities of 18.9 mg/L in steady-state conditions. This oil has a composition predominantly saturated (45.2%) and monounsaturated (34.7%) suitable for biodiesel synthesis (ester content of 99.8%, cetane number of 58.5%, iodine value of 67.2 gI₂/100 g, unsaturation degree of 71.3% and a cold filter plugging point of 6.7 ºC).

Keywords:
Microalgae/cyanobacteria; Heterotrophic cultivation; Sucrose; Biodiesel 3G

INTRODUCTION

The last decade has seen an emergence of biofuels due in part to social and political acknowledgement that fossil fuels are a finite resource. This is evidenced by a reduction in the discovery of new fossil fuel sources and the exploitation of more energy intense reserves such as shale gas and tar sands (Scaife et al., 2015). As a result of this reality, biofuel research and development has progressed through several stages globally and within Brazil.

Currently, the biofuels are classified from first to fourth generation (Harun et al., 2010; Martín and Grossmann, 2012). The third generation biofuels are obtained from microalgae biomass that possess high productivity of lipids, which after extraction are transesterified to obtain biodiesel, turning them into one of the most promising feedstocks for biofuel production (Wijffels and Barbosa, 2010).

There are two possible technological routes to microalgal biomass production: photosynthetic and heterotrophic cultivation. The photosynthetic culture of microalgae, based on CO₂ conversion, is limited by engineering related factors, since design and scale-up methodologies are poorly developed. Factors such as reactor configuration and material construction are considered the main difficulties, when closed photobioreactors are used. On the other hand, open pond technology is limited by biological factors such as organism survival, growth, CO₂ uptake, light utilization, seasonality, harvest and biosafety of transgenics. These questions are related mainly to photosynthetic oil production. An alternative process is based on heterotrophic metabolism of microalgae, in which the organic carbon source is converted in the absence of light. These processes can be conducted in conventional reactor configurations such as stirred tank and bubble column reactors, eliminating some disadvantages of the photosynthetic route (Queiroz et al., 2011).

The microalgae heterotrophic culture, however, is severely limited by organic carbon availability. The organic substrate is estimated at about 80% of the total cost of the process and economically may make it unfeasible. An alternative to the high costs is the replacement of certain sources of organic carbon by low-cost substrates, such as sucrose, which can reduce costs by up to 40% (Xu et al., 2006). According to Li et al. (2007), biodiesel 3G may in turn be produced by microalgae that use sucrose as the substrate, making it a more profitable alternative than ethanol from sugar cane. Additionally, Francisco et al. (2014), in a survey of potential carbon sources for microalgae production, identified sucrose as a suitable substrate to support heterotrophic microalgae cultivation. The heterotrophic microalgae possess structurally specific mechanisms for active transport of sucrose into the cell. This disaccharide is metabolized via the oxidative pentose-phosphate pathway after hydrolysis into monosaccharides (Knowles and Plaxton, 2003).

Thus, the aim of this work was to evaluate the third generation biodiesel production from heterotrophic cultivation of the microalgae Phormidium autumnale employing sucrose as exogenous carbon source. The study focused on optimization of the carbon/nitrogen ratio of the culture media, in the evaluation of different operational modes of the bioreactor and in the analysis of the biofuel quality.

MATERIAL AND METHODS

Microorganisms and Culture Media

Axenic cultures of Phormidium autumnale were originally isolated from the Cuatro Cienegas desert (26º59'N, 102º03'W-Mexico). Stock cultures were propagated and maintained in solidified agar-agar (20 g/L) containing synthetic BG11 medium (Ripka et al., 1979). The incubation conditions used were 25 ºC, a photon flux density of 15 µmol/m²/s and a photoperiod of 12:12 h (light:dark). To obtain the inoculums in liquid form, 1 mL of sterile synthetic medium was transferred to slants, the colonies were scraped and then homogenized with the aid of a mixer tubes. The entire procedure was performed aseptically.

Bioreactor

Measurements were made in a bubble column bioreactor. The system was built of borosilicate glass and had an internal diameter of 15 cm and height of 20 cm, resulting in a height/diameter (h/D) ratio equal to 1.33 and a nominal working volume of 2.0 L. The dispersion system of the reactor consisted of a 2.5 cm diameter air diffuser located inside the bioreactor. The air flow was monitored by a flow meter (KI-Key Instruments^®, Trevose-PA, USA) and the inlet of air and outlet of gases were filtered through filtering units made up of polypropylene membrane with a pore diameter of 0.22 µm and total diameter of 50 mm (Millex FG^®, Billerica-MA, USA). The bioreactor including filtering units was previously sterilized by autoclaving at 121 ºC for 40 min and then for 30 min containing the synthetic medium.

Obtaining Kinetic Data in an Experimental Bioreactor

Initial experiments were performed in a bioreactor operating under a batch regime, fed with 2.0 L of culture medium. The experimental conditions were as follows: initial concentration of inoculum of 100 mg/L, temperature of 30 ºC, pH adjusted to 7.6, aeration of 1 VVM (volume of air per volume of culture per minute) and absence of light. The culture medium consisted of BG11 synthetic medium modified and supplemented with different concentrations of sucrose to obtain carbon/nitrogen ratios (C/N) of 20, 30, 40, 50, 60, 70 and 80. The concentration of sucrose was adjusted stoichiometrically (Francisco et al., 2014).

In the continuous culture, after 120 h of batch culture, feed culture medium was added to the bioreactor at the dilution rate D=0.02 h^-1. At the same time, equal volumes of cell suspension were withdrawn from the bioreactor. The steady-state was considered to have been established after at least three volume charges, with a variation of cell dry weight less than 5%.

The experiments were performed twice, and in duplicate for each operational mode. Therefore, kinetic data refer to the mean value of four repetitions.

Kinetic Parameters

Biomass data were used to calculate the biomass productivity [P_X = (X_i-X_i-1)(t_i-t_i-1)^-1, mg/L/h] and the lipid productivity [P_L = P_X.L_C, mg/L/h], in which X_i is the biomass concentration at the time t_i (mg/L) and X_i-1 is the biomass concentration at the time t_i-1 (mg/L), t is the residence time (h) and L_C is the lipid content of the biomass (%). The concentrations of total organic carbon were used to calculate the substrate consumption rate (r_S=dS/dt, mg/L/h), and the biomass yield coefficient (Y_X/S=dX/dS, mg_cell/mg_substrate), where S₀ is the initial substrate concentration (mg/L), S is the substrate concentration (mg/L) and t is the time (h).

Sampling and Analytical Methods

Samples were collected aseptically in a laminar flow hood. The cell biomass, the pH dynamics and the consumption of organic carbon were monitored every 24 hours during the growth phase of microorganism.

The cell biomass was gravimetrically evaluated by filtering a known volume of culture medium through a 0.45 µm membrane filter (Millex FG^®, Billerica-MA, USA), drying at 60 ºC for 24 h.

The organic carbon concentration was expressed in terms of chemical oxygen demand (COD) and analyzed according to the closed reflux colorimetric method (APHA, 2005).

The total lipid concentration of the biomass was determined gravimetrically by the Bligh and Dyer (1959) method.

The saponification and esterification (methylation reaction) by the modified method of Hartman and Lago (1976) was used with the dried lipid extract to obtain the fatty acid methyl esters (biodiesel). An amount of 250 mg of oil was added to 5.0 mL of 0.50 mol/L NaOH in methanol. The mixture was then heated under reflux for 5 min. After adding 15.0 mL of the esterification reagent (prepared from a mixture of 2.0 g of ammonium chloride, 60.0 mL of methanol, and 3.0 mL of concentrated sulfuric acid for ca. 15min), the mixture was heated under reflux for another 3 min and subsequently transferred to a separation funnel containing 25.0 mL of petroleum ether and 50.0 mL of deionized water. After stirring the mixture and phase separation, the aqueous phase was discarded. Then 25.0 mL of deionized water was added to the organic phase. This mixture was stirred and, after phase separation, the aqueous phase was discarded. This procedure was repeated. The organic phase was collected, the solvent was evaporated in a rotary evaporator and the residue was removed under nitrogen flow. The methyl esters were solubilized in n-heptane before injection in the gas chromatograph. The fatty acid composition was determined using a VARIAN 3400CX gas chromatograph (Varian, Palo Alto-CA, USA). The fatty acid methyl esters were identified by comparison of the retention times with those of the standard (Supelco, St. Louis-MO, USA) and quantified by area normalization.

The fuel properties of biodiesel (ester content, EC; degree of unsaturation, DU; cetane number, CN; iodine value, IV and cold filter plugging point, CFPP) were determined according to the methodology proposed by Francisco et al. (2010).

The cetane number of the mixture was estimated by empirical equations. The cetane number, saponification value and iodine value were calculated in accordance with Eqs. (1)-(3)

(1)

(2)

(3)

where CN is the cetane number, SV is the saponification value, IV is the iodine value, D is the number of double bonds, M is the molecular mass and N is the percentage of each fatty acid component.

The degree of unsaturation was calculated from empirical Eq. (4), taking into account the amount of monounsaturated and polyunsaturated methyl ester (wt%) present in the microalgae oil:

(4)

where DU is the unsaturation degree (%), MUFA is the weight percentage of the monounsaturated fatty acids (wt%).

The long-chain saturated factor was obtained from empirical Eq. (5), taking into account the composition of fatty acids and assigning more weight to the composition of fatty acids with a long chain. This parameter was correlated with the cold filter plugging point, using Eq. (6):

(5)

(6)

where LCSF is the long-chain saturated factor; C16, C18, C20, C22, and C24 are the weight percentage of each of the fatty acids (wt%) and CFPP is the cold filter plugging point.

RESULTS AND DISCUSSION

The assessment of suitable concentration of the carbon sources for the production microalgal biomass and bioproducts is a fundamental step in the consolidation of the process. The kinetic parameters using different C/N ratios for heterotrophic culture of Phormidium autumnale are shown in Table 1. The best results were evidenced in the range of C/N ratio between 30-50, with maximum kinetic performance at the C/N ratio of 40. In this condition, a maximum specific growth rate of 0.02 h^-1, generation times of 32.3 h, maximum cell density of 6170 mg/L, average biomass productivity of 40.7 mg/L/h, average rate of sucrose consumption of 42.3 mg/L.h and a biomass yield coefficient of 0.44 mg_cell/mg_sucrose were obtained. This biomass had a lipid content of 20.7%, resulting in an oil productivity of 8.46 mg/L/h.

Table 1
Kinetic parameters for different C/N ratios using sucrose as substrate in batch cultures.

C/N	µmax(1/h)	tg(h)	Xmax(mg/L)	PX(mg/L/h)	rS(mg/L)	YX/S(mgcell/mgsucrose)	LC(%)	PL(mg/L/h)
20	0.013±0.00	51.3±0.5	2775±56	14.7±0.4	30.9±0.8	0.47±0.00	8.9±0.2	1.32±0.03
30	0.020±0.00	34.1±0.3	5260±30	34.4±0.4	36.1±0.7	0.43±0.00	18.6±0.2	6.41±0.08
40	0.021±0.00	32.3±0.4	6170±28	40.7±0.4	42.3±0.4	0.44±0.00	20.7±0.3	8.46±0.15
50	0.024±0.00	28.8±0.1	5330±53	34.9±0.3	21.2±0.3	0.64±0.00	9.9±0.1	3.50±0.09
60	0.016±0.00	42.0±0.3	2925±18	18.2±0.2	25.0±0.2	0.72±0.00	8.1±0.1	1.40±0.05
70	0.014±0.00	48.8±0.3	1938±16	12.9±0.1	30.4±0.6	0.42±0.00	6.1±0.2	0.80±0.01
80	0.015±0.00	44.7±0.2	1683±22	10.3±0.1	53.1±0.7	0.18±0.00	5.1±0.1	0.53±0.01

According to Fay (1983), in general cyanobacterial culture requires a minimum C/N ratio of 20. Specifically for sucrose, the results obtained indicate a bell-shaped curve pattern in the range between 20 to 80. Comparatively, the results obtained are higher than those reported by Markou and Georgakakis (2011) that indicate maximum cell densities of 1180 mg/L for heterotrophic culture of Phormidium sp. using sucrose as exogenous carbon source.

Sucrose is metabolized by microalgae through the pentose-phosphate pathway (Smith, 1982). These microorganisms possess structurally specific mechanisms for the active transport of sucrose into the cell membrane. The heterotrophic microalgae have an inducible active carbohydrate symport system responsible for uptake of these molecules from the culture medium. The induction of this transport is achieved by some specific sugars. In general, in cultivations with suitable concentration and type of sugars, the symport system is induced to promote the alkalinization of the culture media by a net movement of protons accompanied by sugar uptake (Figure 1). The rate of the increase in pH fundamentally depends on the concentration and type of sugar used (Hong and Lee, 2007). Additionally, the disaccharides such as sucrose are only used in the pentose phosphate pathway, after a previous hydrolysis that transforms this sugar in monosaccharides, particularly fructose and glucose. Several enzymes, specifically invertase, are involved in these reactions and have been identified in microalgae cultures (Fuchs et al., 1994; Gupta et al., 2011).

Figure 1
Variation of the pH (closed circle) and cellular concentration (open circle) vs. time at a C/N ratio of 40.

To further improve bulk oil production, continuous cultivations were performed (Table 2). Maintaining cells in the steady-state resulted in an oil productivity of 18.9 mg/L/h, an increase of 2.2-fold over batch cultivations. A faster product formation rate implies a higher productivity and corresponding reductions in plant operating time and operating cost, for an existing plant. On the other hand, for a new plant to be built, the increased rate implies, in addition to improved productivity, a smaller reactor and therefore a lower capital investment cost. Likewise, improved productivities imply a lower raw material cost and a lower capital investment for existing and new plants (Francisco et al., 2015).

Table 2
Kinetic parameters of the steady-state process.

Parameter	Value
Xmax(mg/L)	4300±698
PX(mg/L/h)	92.1±1.5
rS(mg/L/h)	56.3±1.2
LC(%)	20.5±0.3
PL(mg/L/h)	18.9±0.4

Scale-up projections for these values in the present scenario are limited by the large-scale application of heterotrophic microalgal bioreactors. However, in comparison with the productivity that can be achieved with soybeans, the Brazilian feedstock commonly used industrially for biodiesel, a scale-down projection analysis indicates that, for the Brazilian harvest of 2014, the average lipid productivity of soybeans was 0.46g_oil/m².day, considering a production cycle of 120 days (CONAB, 2015). Each hectare of arable Brazilian soil produces an average of 2,700 kg of soybean, containing up to 20% oil. The data obtained for Phormidium autumnale indicate that a continuous bioreactor with 1.0 L/m² of working volume, operating on a cycle of 120 days/year, would yield the same amount of lipids as that produced by soybean. This comparison indicates that the oil productivity of Phormidium autumnale can be increased by several fold by associating bioreactor optimal design with an operating cycle of 330 days/year.

Finally, in terms of oil composition and fuel properties of biodiesel (Table 3), the lipid fraction of biomass indicated eight different compounds, with oleic acid (26.2%) being the major. The single-cell oil showed a profile predominantly saturated (45.2%) and monounsaturated (34.7%), which determine the fuel properties of microalgal biodiesel (Francisco et al., 2010). The biodiesel produced from microalgal oil had the following fuel properties: ester content of 99.8%, cetane number of 58.5, iodine value of 67.2 gl₂/100 g, degree of unsaturation of 71.3%, and cold filter plugging point of 6.7 ºC. All these parameters comply with the limits established by the US, European, and Brazilian standards (ASTM, 2002, UNE-EN, 2003, ANP, 2003), besides being comparable with soybean biodiesel (Knothe, 2005). These results indicate the potential for the exploitation of this feedstock for biofuel production.

Table 3
Fatty acid profile and fuel properties of biodiesel 3G.

Fatty acid profile
Methyl esters (g/100g)	Lauric (12:0)	Myristic (C14:0)		Palmitic (C16:0)		Palmitoleic (C16:1)	Stearic (C18:0)	Oleic (C18:1n9c)		Linoleic (C18:2n6c)		γ- Linolenic (C18:3n6)
	4.9±0.1	7.3±0.1		22.5±0.5		8.5±0.1	10.5±0.2	26.2±0.8		17.8±0.5		2.10±0.03
Fuel properties
Properties	EC (wt %)		CN		DU (wt %)		IV (gI2/100g)		LCFS (wt %)		CFPP (ºC)
	99.8±0.1		58.5±0.9		71.3±1.8		67.2±0.9		7.4±0.2		6.70±0.06

CONCLUSION

The results obtained indicate that sucrose is an exogenous carbon source with the potential to produce bulk oil and biodiesel by Phormidium autumnale, enabling oil productivities of 18.9 mg/Lh. This oil had a composition of predominantly saturated (545.2%) and monounsaturated (34.7%) fatty acids, suitable for biodiesel synthesis (ester content of 99.8%, cetane number of 58.5, iodine value of 67.2 gl₂/100g, degree of unsaturation of 71.3%, and cold filter plugging point of 6.7 ºC.

NOMENCLATURE

Acronyms C/N Carbon/nitrogen ratio CFPP Cold filter plugging point (ºC) CN Cetane number COD Chemical oxygen demand (mg/L) D Number of double bonds DU Degree of unsaturation (%) EC Ester content (%) IV Iodine value (gl₂/100g) Lc Lipid content of the biomass (%) LCSF The long-chain saturated factor M Molar mass (g/mol) MUFA Weight percentage of the monounsaturated fatty acids (%) N Percentage of each fatty acid component (%) PUFA Weight percentage of the polyunsaturated fatty acids (%) SV Saponification value (%) VVM Volume of air per volume of culture per minute Symbols µmax Maximum specific growth rate (1/h) PL Lipid productivity (mg/L/h) PX Average cellular productivity (mg/L) rS Substrate consumption rate (mg/L/h) S Substrate concentration (mg/L) S0 Initial substrate concentration (mg/L) T Residence time (mg/L) tg Generation time (h) Xmáx maximum cell biomass (mg/L) Yx/s Biomass yield coefficient (mg/mg)

Acronyms
C/N	Carbon/nitrogen ratio
CFPP	Cold filter plugging point (ºC)
CN	Cetane number
COD	Chemical oxygen demand (mg/L)
D	Number of double bonds
DU	Degree of unsaturation (%)
EC	Ester content (%)
IV	Iodine value (gl₂/100g)
Lc	Lipid content of the biomass (%)
LCSF	The long-chain saturated factor
M	Molar mass (g/mol)
MUFA	Weight percentage of the monounsaturated fatty acids (%)
N	Percentage of each fatty acid component (%)
PUFA	Weight percentage of the polyunsaturated fatty acids (%)
SV	Saponification value (%)
VVM	Volume of air per volume of culture per minute
Symbols
µmax	Maximum specific growth rate (1/h)
PL	Lipid productivity (mg/L/h)
PX	Average cellular productivity (mg/L)
rS	Substrate consumption rate (mg/L/h)
S	Substrate concentration (mg/L)
S0	Initial substrate concentration (mg/L)
T	Residence time (mg/L)
tg	Generation time (h)
Xmáx	maximum cell biomass (mg/L)
Yx/s	Biomass yield coefficient (mg/mg)

• This is an extended version of the work presented at the 20th Brazilian Congress of Chemical Engineering, COBEQ-2014, Florianópolis, Brazil.

ACKNOWLEDGMENTS

Funding for this research was provided by the National Council of Technological and Scientific Development (CNPq, Brazil).

REFERENCES

• ANP 255, Provisional Brazilian Biodiesel Standard ANP (Agência Nacional do Petróleo). (2003).
• APHA, AWWA, WEF, (American Public Health Association, American Water Works Association, Water Environmental Federation). Standard Methods for the Examination of Water and Wastewater, Baltimore, Maryland (2005).
• ASTM 6751, Standard Specification for Biodiesel Fuel (B100). Blend Stock for Distillate Fuels (2002).
• Bligh, E. G. and Dyer, J. W., A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology, 37(8), 911-917 (1959).
• CONAB, Brazilian Supplying Company. Available: http://www.conab.gov.br (2015).
  » http://www.conab.gov.br
• Fay, P., The Blue-Greens (Cyanophyta-cyanobacteria). Edward Arnold Ltd, London (1983).
• Francisco, E. C., Franco, T. T., Wagner, R. and Jacob-Lopes, E., Assessment of different carbohydrates as exogenous carbon source in cultivation of cyanobacteria. Bioprocess Biosystems Engineering, 37(8), 1497-1505 (2014).
• Francisco, E. C., Franco, T. T., Zepka, L. Q. and Jacob-Lopes, E., From waste-to-energy: The process integration and intensification for bulk oil and biodiesel production by microalgae. Journal of Environmental Chemical Engineering, 3(1), 482-487 (2015).
• Francisco, E. C., Neves, D. B., Jacob-Lopes, E. and Franco, T. T., Microalgae as feedstock for biodiesel production: Carbon dioxide sequestration, lipid production and biofuel quality. Journal of Chemical Technology and Biotechnology, 85(3), 395-403 (2010).
• Fuchs, B., Suttner P., Sterner, S., Wastlhuber, R. and Loos, E., Disproportionating transglycosylase (D-enzyme) in green algae and cyanobacteria. Partial purification and characterization. Zeitschrift für Naturforschung, 49, 163-170 (1994).
• Gupta, V., Natarajan, C., Kumar, K. and Prasanna, R., Identification and characterization of endoglucanases for fungicidal activity in Anabaena laxa (Cyanobacteria). Journal of Applied Phycology, 23(1), 73-81 (2011).
• Hartman, L. and Lago, R. C. A., A rapid determination of fatty acid methyl esters from lipids. Lab Pratice, 22, 475-476 (1976).
• Harun, R., Danquah, M. K. and Forde, G. M., Microalgae biomass as a fermentation feedstock for bioethanol production. Journal of Chemical Technology and Biotechnology, 85(2), 199-203 (2010).
• Hong, S. J. and Lee, C. G., Evaluation of central metabolism based on a genomic database of Synechocystis PCC6803. Biotechnology and Bioprocess Engineering, 12(2), 165-173 (2007).
• Knothe, G. H., The Biodiesel Handbook. AOAC Press, Champaign, USA (2005).
• Knowles, V. L. and Plaxton, W. C., From genome to enzyme: Analysis of key glycolytic and oxidative pentose-phosphate pathway enzymes in the cyanobacterium Synechocystis sp. PCC 6803. Plant Cell Physiology, 44(7), 758-763 (2003).
• Li, X., Xu, H. and Wu, Q., Large-scale biodiesel production from microalga Chlorella protothecoides through heterotrophic cultivation in bioreactors. Biotechnology and Bioengineering, 98(4), 764-771 (2007).
• Markou, G. and Georgakakis, D., Cultivation of filamentous cyanobacteria (blue-green algae) in agro-industrial wastes and wastewaters: A review. Applied Energy, 88(10), 3389-3401 (2011).
• Martín, M. and Grossmann, I. E., On the systematic synthesis of sustainable biorefineries. Industrial & Engineering Chemistry Research, 52(9), 3044-3064 (2012).
• Queiroz, M. I., Hornes, Silva-Manetti, A. G. and Jacob-Lopes, E., Single-cell oil production by cyanobacterium Aphanothece microscopica Nägeli cultivated heterotrophically in fish processing wastewater. Applied Energy, 88(10), 3438-3443 (2011).
• Rippka, R., Derueles. J., Waterbury, J. B., Herdman, M. and Stanier, R. Y., Generic assignments strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology, 111(1), 1-61 (1979).
• Scaife, M. A., Merkx-Jacques, A., Woodhall, D. L. and Armenta, R. E., Algal biofuels in Canada: Status and potential. Renewable and Sustainable Energy Reviews, 44, 620-642 (2015).
• Smith, A. J., Modes of Cyanobacterial Carbon Metabolism. The Biology of Cyanobacteria. University of California Press, Berkeley (1982).
• UNE-EN 14214, Automotive Fuels, Fatty Acid Methyl Esters (FAME) for Diesel Engines. Requirements and Test Methods (2003).
• Wijffels, R. H. and Barbosa, M. J., An outlook on microalgal biofuels. Science, 329(5993), 796-799 (2010).
• Xu, H., Miao, X. and Wu, Q., High quality biodiesel production from a microalgae Chlorella protothecoides by heterotrophic growth in fermenters. Journal of Biotechnology, 126(4), 499-507 (2006).

Publication Dates

• Publication in this collection
  Jul-Sep 2016

History

• Received
  28 Feb 2015
• Reviewed
  28 Apr 2015
• Accepted
  12 May 2015