In addition to the in-house research being conducted in the area of strain collection and screening, there was an effort by Dr. Thomas Tornabene and others to characterize various strains via detailed lipid compositional analyses. Dr. Tornabene’s laboratory at SERI (and later at the Georgia Institute of Technology) served as the focal point for the analysis of lipids in algal samples supplied by various researchers in the ASP. This section will describe the results of these analyses, and will provide details about the analytical methods used, as these methods were the most comprehensive used in the program. An early report by Tornabene et al. (1980) described the lipids that were present in the halophilic alga Dunaliella that had been isolated from the Great Salt Lake in Utah. The cells were grown to late logarithmic phase, harvested, and extracted with chloroform/methanol via the method of Bligh and Dyer (1959). Additional extraction by acetate buffer, followed by refluxing with an alkaline methanol/water mixture was then performed, followed by partitioning of lipids into petroleum ether. The extracted lipids were fractionated on the basis of polarity using silicic acid columns via differential elution with hexane, benzene, chloroform, acetone, and methanol. In this procedure, the lipids are eluted as follows: [4]

2. benzene: cyclic hydrocarbons, polyunsaturated acyclic hydrocarbons, sterols, and xanthophylls

3. chloroform: mono-, di — and triacylglycerols, free fatty acids, and phaeophytin a

4. acetone: glycolipids, carotenoids, and chlorophyll a and b; and

5. methanol: phospholipids and chlorophyll c.

The various lipid classes were further analyzed via Si gel thin layer chromatography (both one — and two-dimensional), wherein lipids were detected via the use of iodine vapors (and autoradiography in the case of 14C-labeled lipids). In addition, lipids containing amino groups were detected via the ninhydrin reagent, and phospholipids were detected by the use of molybdate/H2SO4. Fatty acids were analyzed via gas chromatography using either flame ionization or mass spectroscopic detection after being converted to their methyl ester derivatives in the presence of methanolic HCl. The head groups of the polar lipids were identified via gas chromatography after being converted to alditol acetates. These and related methods were described by Tornabene et al. (1982).

These analyses indicated that lipids comprised 45%-55% of the total organic mass of Dunaliella cells. Based on the distribution of 14C after labeling the cells with 14C-bicarbonate, neutral lipids accounted for 58.5% of the lipid mass, whereas phospholipids and galactolipids were 22.9% and 10.9% of the lipid mass, respectively. Isoprenoid hydrocarbons (including p-carotene) and aliphatic hydrocarbons (in which the major components were tentatively identified as straight — chain and methyl-branched C17 and C19 hydrocarbons with various degrees of unsaturation) represented 7.0% and 5.2% of the lipids, respectively. The major fatty acids present were palmitic (20.6%), linolenic (12.5%), linoleic (10.7%) and palmitoleic (7.8%), but no attempt was made to ascertain whether any of these fatty acids predominated a particular lipid class. The high hydrocarbon content of this alga is rather atypical of most of the strains characterized in the ASP. These types of hydrocarbons would probably require catalytic conversion into a usable fuel source, which would perhaps limit their utility as a production organism.

A detailed analysis of the lipids present in the green alga Neochloris oleoabundans was also carried out by Tornabene (who was later to hold a position at the Georgia Institute of Technology), along with G. Holzer (Colorado School of Mines), S. Lien and N. Burris (SERI) (Tornabene et al. 1983). The strain used in this study was obtained from the University of Texas Algal Culture Collection, and reportedly contained substantial quantities of lipid when grown under N-deficient conditions. (However, this is a freshwater strain). Exponentially growing cells were transferred into a low-N medium, and after 5 to 7 days of growth in stirred cultures that were bubbled with 1% CO2 in air, the cells were harvested and the lipids were extracted. Analytical methods were similar to those described earlier, and included the use of pyrrolidine — acetic acid/mass spectrometry to determine the position of double bonds in the fatty acids. These analyses indicated that 35%-54% of the cellular dry weight was in the form of lipids in N — deficient cells. Neutral lipids accounted for more than 80% of the total lipids, and were

predominantly in the form of TAGs. Small amounts of straight-chain hydrocarbons and sterols were also found (one sterol was identified as a Д7 sterol, but low quantities of material made identification of the sterols difficult). A number of polar lipids were also quantified, but all polar lipids combined accounted for less than 10% of the lipid mass. The fatty acids that comprised the TAGs were present in the following proportions: 36% oleic (18:1 Д9), 15% palmitic (16:0), 11% stearic (18:0), 8.4% iso-17:0 (an unusual fatty acid for microalgae), and 7.4% linoleic (18:2 Д9,12). Other saturated and monounsaturated fatty acids were present in TAGs, but represented less than 5% each of the total fatty acids present. The high proportion of saturated and monounsaturated fatty acids in this alga is considered optimal from a fuel quality standpoint, in that fuel polymerization during combustion would be substantially less than what would occur with polyunsaturated fatty acid-derived fuel (Harrington, 1986).

Additional research carried out in Tornabene’s laboratory (Ben-Amotz et al. 1985) examined the lipid composition of 7 algal species. Some were from existing culture collections and others were isolated by ASP researchers. The lipid contents of these strains were determined under conditions of N sufficiency, after 10 days of N deficiency, and under different salinity levels.

Botryocooccus braunii has received considerable interest as a fuel production organism in other laboratories because of its high lipid content. This study confirmed the high lipid levels (55% of the organic mass for N-deficient cells). Most of this lipid was in the form of hydrocarbons, including C29 to C34 aliphatic hydrocarbons and a variety of branched and unsaturated isoprenoids. Glycerolipids were less abundant than the hydrocarbons, and were composed primarily of 16:0 and various CJ8 fatty acids. These data, coupled with the fact that this species grows very slowly (one doubling per 72 hours), indicated that Botryococcus would not function well as a feedstock for lipid-based fuel production.

The other species examined in this study were the chlorophytes Ankistrodesmus, Dunaliella, and Nannochloris, the diatom Nitzschia, and the chrysophyte Isochrysis. N deficiency led to an increase in the lipid content of Ankistrodesmus (from 24.5% to 40.3%), Isochrysis (from 7.1% to 26.0%), and Nannochloris (from 20.8% to 35.5%), but resulted in a decrease in the lipid content of Dunaliella (from 25.3% to 9.2%). Elevating the NaCl concentration of the medium had little effect on the lipid content of Botryococcus cells, but caused a slight decrease in the lipid content of Dunaliella salina (from 25.3% to 18.5% with an increase in [NaCl] from 0.5 to 2 M). Conversely, the lipid content of Isochrysis increased from 7.1% to 15.3% as the NaCl increased from 0.5 to 1 M. These results once again highlight the impact of culture conditions on the quantities of lipids present. However, as stated before, the most important characteristic of a lipid production strain is the overall lipid productivity for a given amount of time, which was not examined in this study.

The polar lipid composition of the strains examined in this study were typical of photosynthetic microalgae, and included phosphatidylcholine, phosphatidylinositol, phosphatidylethanolamine, phosphatidylglycerol, monogalactosyldiacylglycerol, and digalactosyldiacylglycerol.

Table II. A.5 indicates the major fatty acids (those at levels exceeding 5% of the total) present in these strains, both under N-sufficient and N-deficient growth conditions.

In conclusion, the work carried out by Tornabene’s laboratory provided a detailed characterization of the lipids present in a variety of microalgae. No general conclusions could be made from the work except that the lipid composition of various microalgal strains can differ quite substantially. Because the nature of the lipids can have a large impact on the quality of the fuel product, characterizing the potential production strains is important to ensure that deleterious lipids (e. g., highly polyunsaturated fatty acids in the case of biodiesel fuel) are not present at high levels.

Table II. A.5. Major fatty acids of various microalgae. (Fatty acids in bold are present at levels of 15% or higher)