Degree Type

Dissertation

Date of Award

2013

Degree Name

Doctor of Philosophy

Department

Mechanical Engineering

First Advisor

Robert C. Brown

Abstract

The overall goal of this research is to develop methods for analyzing and recovering sugars and phenolic compounds from bio-oil. Specific objectives include (1) adapting analytical methods developed for sugar analysis in the food industry to measure total water-soluble sugars in the aqueous phase of bio-oil; (2) adapting analytical methods developed for total phenolic analysis of wine to measure total phenolic content of bio-oil; (3) separate the heavy fraction of bio-oil into a concentrated sugar solution and a phenolic oligomer-rich raffinate; and (4) determine the effect of pyrolysis temperature on the yield of sugars and phenolic compounds in bio-oil.

Recent research at Iowa State University suggests that bio-oil may be the most economical approach to advanced biofuels production. Produced from the fast pyrolysis of biomass, bio-oil contains hundreds of chemical compounds that complicate their accurate and cost-effective chemical analysis. Among the most commercially important components of bio-oil are sugars and phenolic compounds. Both are difficult to analyze because of the large number of variations that can occur and potential interferences with other bio-oil components.

The approach to this research was to adapt chemical analysis methods developed by food chemistry to quantify total sugars and phenolic compounds using standardized test methods. The Association of Analytical Communities, Inc. (AOAC) Official Method of Analysis 988.12 (44.1.30) Phenol-Sulfuric Acid Assay for Total Carbohydrate Determination was utilized to quantify water-soluble sugars and the Folin-Ciocalteu (FC) colorimetry method was used to quantify total phenolic compounds in bio-oil.

Bio-oil produced from fast pyrolysis of biomass contains sugars originating from cellulose. Traditional quantification of sugars in bio-oil is accomplished by gas chromatography/mass spectroscopy (GC/MS) via derivatization, high performance liquid chromatography (HPLC), ion chromatography (IC), or nuclear magnetic resonance (NMR) methodologies. These techniques are highly specific for each sugar, tedious to perform, expensive, and involve the use of hazardous solvents.

A standardized test method developed for food and agriculture applications, the Association of Analytical Communities, Inc. (AOAC) Method 988.12 (44.1.30) Phenol-Sulfuric Acid Assay for Total Carbohydrate Determination, was utilized to quantify total sugars in the water-soluble fraction of bio-oil. This study investigated accuracy relative to matrix effects caused by non-sugar compounds using positive and negative controls. Positive controls included levoglucosan, D-glucose, D-mannose, D-xylose, D-fructose, D-galactose, L-arabinose, L-fucose, and cellobiosan. Negative controls included phenol, acetic acid, formic acid, propionic acid, glycolic acid, acetol, furfural, 5-hydroxymethylfurfural (5HMF), furfuryl alcohol, 2-methylfuran and 2(5H)-furanone.

Potential interference with the quantification of total water-soluble sugars by the AOAC Method 988.12 (44.1.30) was calculated for all positive and negative controls by using data obtained when adding the contributor (positive controls) and the interferent (negative controls) into the water-soluble fraction of bio-oil with typical concentrations found in bio-oil. It was found that furfural, 2(5H)-furanone, 5HMF, and furfuryl alcohol influenced results with a range of potential errors of 9.56-29.7%, 9.52-29.8%, 2.91-24.8%, and 1.34-11.9%, respectively. A correction factor of 0.76 wt% was established to reduce or eliminate this influence. Total water-soluble sugars content in bio-oil detected by AOAC Method 988.12 (44.1.30) was comparable to the quantity of sugars detected using hydrolysis with quantification by HPLC. The uncertainty of measurement of water-soluble sugars in bio-oil at 95% confidence was ¡À2.0% using AOAC Method 988.12 (44.1.30) when the correction factor was employed.

Bio-oil from fast pyrolysis of biomass contains phenolic compounds derived from the lignin portion of the biomass. Traditional testing for total phenolic compounds in bio-oil is based on either a rough estimate of the weight percent water-insolubles in bio-oil or on tedious liquid-liquid extraction methods. The Folin-Ciocalteu (FC) colorimetry method used for quantifying total phenols in wine was used to determine total phenols in bio-oil. This method, based on the oxidation of phenolic compounds by the FC reagent, is fast and easy to perform. This study evaluated its accuracy relative to interferents by the use of positive and negative controls.

Positive controls included phenol, 4-methylphenol, 3-ethylphenol, guaiacol, 2,6-dimethoxyphenol and eugenol. The negative controls included sugars, furfural, and acids. Potential interferents with the quantification of total phenols by the FC method was calculated for all positive and negative controls by using data obtained when adding the contributor (positive controls) and the interferent (negative controls) into bio-oil using typical concentrations found in bio-oil. The positive and several of the negative controls produced strongly correlated linear relationships between the indicated phenolic content of the bio-oil and the amount of contributor or interferent added. However, the slopes of these relationships for the negative controls were much smaller than those for the positive controls, indicating that the error in the prediction of phenolic content was small even for large concentrations of interferent compounds.

For typical concentrations of non-phenolic compounds in bio-oil, the error in predicted phenolic content as a result of their presence was ¡Ü 5.8%. Total phenolic content in bio-oil detected by the FC method was comparable to the quantity of total phenolic compounds obtained by liquid-liquid extraction. All results fell within the margin of error and the uncertainty of the measurement by the FC method indicating there was no significant difference in the results between the two methods. The FC method uncertainty of measurement was ¡À1.1% at the 95% confidence level.

An investigation of sugar and phenolic oligomer recover from the heavy ends of fractionated bio-oil is performed. This study explores the separate recovery of sugars and phenolic oligomers produced during the fast pyrolysis of lignocellulosic biomass. The experiments were conducted in an 8 kg/h fluidized bed pyrolysis process development unit. Bio-oil fractionation was accomplished with a five-stage system that recovers bio-oil according to ¡°dew points¡± of the constituent compounds. The first two stages capture ¡°heavy ends¡± consisting mostly of water soluble sugars derived from polysaccharides and water insoluble phenolic oligomers derived from lignin. Exploiting differences in water solubility, a sugar-rich aqueous phase and a phenolic-rich raffinate were recovered. The soluble sugars were effectively washed from the phenolic oligomers allowing the production of ¡°pyrolytic sugars¡± and a carbohydrate-free raffinate comprised of phenolic oligomers that readily flowed at room temperature. Over 93 wt% sugars were removed with two wash stages for stage fractions (SF) 1 and 2.

The separated sugars from SF 1 and 2 are suitable for either fermentation or catalytic upgrading to biofuels. The phenolic oligomer-rich raffinate, which represents 44-47 wt% dry basis (db) of both SF1 and 2, is less sticky and viscous than the unwashed stage fractions. It has potential for production of fuels, aromatic chemicals, unique polymers, resins, binders, coatings, adhesives, asphalt, and preservatives.

Iowa State University¡¯s fluidized bed pyrolysis process development unit (PDU) with a condenser collection system is utilized to evaluate physicochemical properties of bio-oil produced at 350, 400, 450, 500, and 550 ¡ãC. A study of temperature effects on the production and distribution of specific chemical families and or chemicals is pertinent to quality bio-oil with specific end-use applications. The red oak biomass gave maximum bio-oil yield at 400¡ãC, highest non-condensable gases (NCG) yield at 550¡ãC, with the highest char yield at 350¡ãC. Carbon monoxide increased at the expense of carbon dioxide at 550¡ãC. There was a slight increase in methane as well. A higher conversion of cellulose and hemicellulose content to sugars resulted at 400¡ãC and was condensed in stage fractions (SF) 1-2. Total phenolic production was highest at 350¡ãC with the majority being larger lignin derived oligomers which condensed in SF1-2. The phenolic monomers were the most prevalent at 550¡ãC with the highest concentration condensed in SF3.The insolubles ranged from 40-45 wt% at 500-550¡ãC in SF1-2. Moisture content was highest at 550¡ã in SF5.

Copyright Owner

Marjorie Ruth Rover

Language

en

File Format

application/pdf

File Size

133 pages

Share

COinS