Pilot production of biodiesel from canola in New England

Canola Production. Canola was chosen as an oilseed crop because it is well adapted to northern Maine. Canola (variety Hyola 401) was sown at a rate of 6 lb per acre in May, 2005 on two farms in Presque Isle, Maine. At both sites, the previous crop had been potatoes. The farms represented two types of management systems including low-input management and conventional management. The low input site was represented by no herbicide or fertilizer applications. At the conventional management site, an herbicide (trifluralin) was preplant incorporated at a rate of 1.5 pint per acre and nitrogen was applied at a rate of 70 lbs per acre as ammonium nitrate. No in-season applications of fungicide or insecticide were made at either site. Both fields were direct combined when seed moisture was less than 11 percent. The seed was kept in one-ton tote bags for four months in an unheated storage (temperatures of -20 to 10 C) until it was crushed. No mold or heating of the seed was observed. Costs of production were estimated by interviews conducted with the farmers after the end of the season. The cost of storage was estimated at $ 8.00 per ton.

Canola variety trials were conducted using a randomized complete block design with four replications at each of three sites (St. Alburg, Vermont; Presque Isle, Maine; Orono, Maine). Plot size was 5 by 20 feet. Twenty six varieties were included in the trials. Plots were harvested at maturity and seed yield and percent moisture were determined.

Oil Extraction: Oil was mechanically extracted from canola seed using an extruder, and expeller (model 1500, and model 2500, respectively; InstaPro, Des Moines, Iowa) which had been set up for crushing soybeans. For canola, the seed was run through the extruder and the expeller, and then the meal was run through both steps a second time to achieve 80 % oil extraction. This work was done in the winter in an unheated structure with temperatures ranging from –15 to –5 C. Under warmer conditions, it is anticipated that a single pass through the extruder, followed by two passes through an expeller would be sufficient. Under the arrangement used here, the system had a crushing capacity of half a ton an hour. If there were two expellers set in a series, and the canola seed was warmed before crushing, then this system would have a capacity of one ton of canola per hour. For economic analysis in this study, the cost of oil extraction was estimated assuming a facility would have two expellers set up in a series and be able to crush one ton per hour.

Biodiesel Production. A small scale batch-system (189 liters per batch) was constructed for processing canola oil into biodiesel. An electric water heater (Kenmore Power Miser 12, Sears Roebuck and Co., Chicago, Illinois) served as the reaction vessel. The system was modeled after one reported by Alovert (2004). Crude canola oil was warmed to 57 C and then methanol and methoxide were introduced as a catalyst. The mixture was agitated using a circulation pump (Clear Water Pump, Northern Tool and Equipment Co., Burnsville, Minnesota). Each batch consisted of 185 liters of crude canola oil, 42 liters of methanol, and sodium hydroxide at a rate of 8 grams per liter of oil. The pump, which circulated the solution, was then engaged for 3 hours and the solution was kept at a temperature of 57 C. After 3 hours the pump and the heating element were turned off and the solution was allowed to sit until the following morning. The glycerin was drained off the bottom, and the methyl ester (biodiesel) was pumped into a storage tank. The crude glycerin was composted. Once the tank was completely drained, then it was refilled and another batch was initiated. The biodiesel was placed in a large tank for washing and storage. Magnesol was added to the biodiesel to help remove impurities. The biodiesel was washed by sprinkling warm water over the surface of the tank in a 6:1 ratio of biodiesel to water. The water with impurities was then pumped out of the bottom of the tank. This washing step was repeated three times. The biodiesel was initially filtered to three microns, and later filtered through a one micron sock filter.

Note that the goal of this project was not to identify the most efficient method of producing biodiesel, nor was it to develop new methods of producing biodiesel. The objective here was to economically evaluate small scale production with commonly used and readily available methods. If this trial were to be repeated, rather than using the reaction vessel as a settling tank, we would suggest having a separate settling tank that could be readily inspected and cleaned as needed, and that could be used for washing each batch of biodiesel as needed.