Grantee Research Project Results
2009 Progress Report: Novel Reactor Design for Biodiesel Production
EPA Grant Number: SU834015Title: Novel Reactor Design for Biodiesel Production
Investigators: Cairncross, Richard A. , Cernansky, Nicholas P.
Institution: Drexel University
EPA Project Officer: Page, Angela
Phase: II
Project Period: August 15, 2008 through August 14, 2010
Project Period Covered by this Report: August 15, 2008 through August 14,2009
Project Amount: $74,960
RFA: P3 Awards: A National Student Design Competition for Sustainability Focusing on People, Prosperity and the Planet - Phase 2 (2008) Recipients Lists
Research Category: Pollution Prevention/Sustainable Development , P3 Challenge Area - Air Quality , P3 Challenge Area - Chemical Safety , P3 Awards , Sustainable and Healthy Communities
Objective:
The goal of this project is to demonstrate the feasibility of scaling up a novel bubble-column reactor for producing biodiesel from alternative feedstocks. Biodiesel is an alternative fuel that can be produced from a wide variety of plant oils, animal oils and waste oils from food processing. The conventional feedstocks for biodiesel production are refined vegetable oils (triglycerides) produced by intensively managed high-value food crops, such as soybeans. The reactor being designed in this project is especially suited for low-value waste oils, such as yellow grease and trap grease. Many waste oils and alternative plant oils contain a significant amount of free fatty acids (FFA), which lead to excessive soap production and low conversion to biodiesel using the reaction processes. The novel reactor in this project uses acid catalysts that do not produce soaps, and it runs at temperatures above the boiling point of methanol—so methanol bubbles rise through a column of oil and react with the oil to produce biodiesel. The bubbling methanol provides agitation and removes the byproduct water, enabling higher overall conversion to biodiesel. Our results show that this reactor is robust for converting oils to biodiesel over the full range of FFA composition (from 100% triglycerides to 100% FFA).
In Phase II of this U.S. Environmental Protection Agency People, Prosperity and the Planet (EPA P3) project, the bubble column reactor design is being optimized for scale-up to pilot or production scale. Prototype reactors are being used to measure reaction rate constants and mass transfer coefficients for water and methanol. A mathematical model of the reactions and mass transfer within the bubble reactor has been developed. There is a trade-off between faster intrinsic kinetics and higher methanol volatility at higher temperatures. The bubble reactor model will be used to determine appropriate scaling parameters for a bubble column reactor. Results from the experiments and model will be used for a detailed economic feasibility study of a large-scale production in the United States and small-scale production in rural developing communities. Because this reactor is more robust for various oil compositions and for contamination of the feed with water, this reactor may be appropriate for a wide range of applications in both industrialized and developing nations.
Progress Summary:
More than 20 undergraduate students from the Chemical & Biological Engineering Department and the Mechanical Engineering & Mechanics Department have been involved in this EPA P3 project. Two Mechanical Engineering & Mechanics senior design teams designed, constructed and tested continuous bubble column reactors. Two Chemical & Biological Engineering senior design teams analyzed the economic feasibility of full-scale biodiesel production using bubble column reactors. There have also been seven researchers involved in developing the analytical procedures for measuring conversion, conducting exploratory reactor experiments and developing a mathematical model.
In the laboratory, there are now several batch and continuous reactors that have all been used successfully to produce biodiesel from oleic acid; oleic acid is a commercially available FFA that is used as a model of purified trap grease because dewatered and filtered trap grease is nearly 100 percent FFA. Several of the reactors are laboratory scale ranging from about 70–600 mL; these reactors are used for exploratory experiments. Two continuous reactors have been built with volumes of 1 and 1.5 liters; these reactors can be run in batch or continuous mode. In continuous mode, production of more than 24 liters per day is possible but has not yet been demonstrated. The effect of the internal structure of the continuous reactors was investigated, and baffles nearly double the rate of reaction. Baffles were as effective at increasing the rate of conversion as agitation; so in scaling up to large columns, appropriate design of baffles to disperse the bubbles should be sufficient to maximize conversion to biodiesel.
All of the reactors produce biodiesel with similar yield. For most reactions, sulfuric acid is used as a catalyst. In many experiments, the biodiesel produced is very dark in color, and often a tar-like residue settles on the bottom of the reaction vessel. Several hypotheses have been evaluated for the side-reactions causing this color change. Based on prior literature, formation of estolides is a possible side reaction; estolides are polymers of fatty acids formed by reaction of a carboxylic acid with a carbon-carbon double bond on the fatty acid chain. Estolides also improve lubricity of biodiesel and reduce the gelation temperature so they may be beneficial for biodiesel. Reactions were conducted to enhance formation of estolides, and estolide peaks were identified via H1-NMR. However, the estolide NMR peaks have not been observed in any of the biodiesel samples produced under typical reaction conditions. Although the cause of the color change has not been determined, we have found experimental procedures to minimize the color change: dissolving the sulfuric acid in methanol prior to adding it to the reactor, pre-heating the oil in the reactor and starting the flow of methanol vapor into the reactor prior to adding the catalyst (this ensures that the catalyst mixes quickly into the oil).
Our results show that bubble-column reactors are very robust for feed alcohols that contain water. Typically, water leads to low conversions and soap formation in biodiesel reactors. Soap formation is avoided by using an acid catalyst. With water in the feed alcohol, high conversions are achieved in a bubble column reaction because the vapor bubbles strip byproduct water out of the oil and drive the reaction towards higher conversion to biodiesel. Reactions using methanol containing up to 20 percent water by volume achieved greater than 97 percent conversion to biodiesel, although the rate of reaction decreases somewhat. Likewise, reactions with ethanol-water mixtures achieved greater than 97 percent conversion with up to 10 percent water in the feed; thus, it is not necessary to break the ethanol/water azeotrope prior to using ethanol to produce biodiesel.
A two-phase reactor model was derived and solved numerically to predict performance of the bubble column reactor. The model is based on mass balances of four reacting species (FFA, biodiesel, methanol and water) within both the liquid and vapor phases. Reactions occur within the liquid phase and are first order in the concentration of each component. Methanol and water can transfer between the vapor and liquid phases and the rate of transport is modeled using a mass transfer coefficient. The model is appropriate for semi-batch reactions or continuous reactions. There are several important parameters that arise from the model. These parameters were estimated from the literature and from semi-batch laboratory experiments; the model agrees nicely with the experimental data. Using parameters from the pure methanol feed experiments, the model predicted the reduction in reaction rate for experiments in which the methanol feed contained up to 20 percent water. This model will be used to guide scale-up of experiments and to guide economic feasibility studies.
The bubble column biodiesel reactor is robust for converting low-value and impure feedstocks into biodiesel. Oil feeds containing 0 percent and 100 percent free fatty acids react with methanol and are converted completely to biodiesel within about 90 minutes at a reactor temperature of 120°C. Reaction with ethanol exhibits similar conversion kinetics to reaction with pure methanol. For both ethanol and methanol as the alcohol feed, high conversions are achieved when the alcohol contains up to 20 percent water. A reaction model has been developed that predicts the reaction kinetics, equilibrium limitations and mass transfer resistances within the bubble column reactor. Using model parameters based on results with pure methanol, the model accurately predicts changes in conversion when the feed contains up to 20 percent water. With baffles, similar conversion kinetics are observed in reactors ranging from 70 mL to over 1 liter; so the reactor is scalable to higher production capacity. This EPA P3 project has been a valuable educational program enrolling more than 20 undergraduate students directly involved in research or design projects and educational outreach to middle-school and high-school students.
In both of the last 2 years (2008 and 2009), students working on this project have presented a workshop on biodiesel production to middle school and high school students through the Center for Talented Youth program. The Drexel students prepared three experimental demonstrations and rotated teams of students through the experiments: (1) mixing reactants for biodiesel production, (2) washing biodiesel, and (3) testing biodiesel purity. Each year, approximately 60 participants have learned about biodiesel through these demonstrations.
The bubble column biodiesel reactor is robust for converting low-value and impure feedstocks into biodiesel. Oil feeds containing 0 percent and 100 percent FFA react with methanol and are converted completely to biodiesel within about 90 minutes at a reactor temperature of 120°C. Reaction with ethanol exhibits similar conversion kinetics to reaction with pure methanol. For both ethanol and methanol as the alcohol feed, high conversions are achieved when the alcohol contains up to 20 percent water. A reaction model has been developed that predicts the reaction kinetics, equilibrium limitations and mass transfer resistances within the bubble column reactor. Using model parameters based on results with pure methanol, the model accurately predicts changes in conversion when the feed contains up to 20 percent water. With baffles, similar conversion kinetics are observed in reactors ranging from 70 ml to over 1 liter; so the reactor is scalable to higher production capacity. This EPA P3 project has been a valuable educational program enrolling more than 20 undergraduate students directly involved in research or design projects and educational outreach to middle school and high school students.
Journal Articles:
No journal articles submitted with this report: View all 10 publications for this projectSupplemental Keywords:
Sustainability, biofuel, soybean oil, trap grease, FAME, FFA, biodiesel, RFA, Scientific Discipline, Sustainable Industry/Business, POLLUTION PREVENTION, Environmental Chemistry, Sustainable Environment, Energy, Technology for Sustainable Environment, Environmental Engineering, sustainable development, environmental sustainability, alternative materials, biomass, alternative fuel, biodiesel fuel, energy efficiency, energy technology, alternative energy sourceProgress and Final Reports:
Original AbstractP3 Phase I:
Novel Reactor Design for Biodiesel Production | Final ReportThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.